PACRIM 2004 Proceedings Cover 26/8/04 2:35 PM Page 1

5/2004

PA C R I M 2 0 0 4
Adelaide, South Australia

PACRIM 2004
Hi Tech and World
Competitive Mineral
Success Stories Around
the Pacific Rim

19-22 September 2004

Gold Sponsor Proceedings

Silver Sponsor
PACRIM 2004
Congress
19 - 22 September 2004
Adelaide, South Australia

The Australasian Institute of Mining and Metallurgy

Publication Series No 5/2004

Published by

THE AUSTRALASIAN INSTITUTE OF MINING AND METALLURGY

Level 3, 15-31 Pelham Street, Carlton Victoria Australia 3053
 The Australasian Institute of Mining and Metallurgy 2004

All papers published in this volume were reviewed prior to publication.

The Institute is not responsible as a body for the facts and opinions advanced in any of its publications.

ISBN 1 920806 19 9

Desktop published by:

Angie Spry and Kristy Burt for The Australasian Institute of Mining and Metallurgy

Compiled on CD ROM by:

Visual Image Processing Pty Ltd
113 King Street
Templestowe VIC 3106
Organising Committee
Trevor Ireland
Congress Chairman

Brian Levet

John Dow

Ian Gould

Carolyn Ireland

Dale Sims

Andrew Vigar

Kevin Wills

The AusIMM
Miriam Way (Events Manager)

Alison McKenzie (Senior Conference and Events Coordinator)

Angie Spry (Senior Conference Publications Coordinator)

Kristy Burt (Publications Assistant)

 Reviewers
We would like to thank the following people for their contribution towards
enhancing the quality of papers included in this volume.

Gerard Bosch Mark McGeough

Solomon Buckman Ian Pontifex

Serge Caplygin Nigel Radford

Lucy Chapman Andrew Robertson

Andrew Cuthbertson Andreas Schmidt Mumm

Brett Davies Michael Sexton

Greg English Dale Sims

Bob Gunthorpe David Turvey

John Hammond Andrew Vigar

Trevor Ireland Ron Wilmshurst

 Foreword
PACRIM congresses are held by The Australasian Institute of Mining and Metallurgy, once
every four years. Their objective is the presentation of a palette of current thinking across
many aspects of earth science and the mining industry of the plate margins of the Pacific
Basin. Despite the general commonality of geological processes and ore formations occurring
around and across our largest ocean, we apply our professional endeavours in an enthralling
variety of geological settings, technical environments and geopolitical situations. Such is the
range, it could be argued that the strongest thread of continuity is provided only by the
Pacific surf. Further, the rate of progress over each quadrennium ensures that every PACRIM
program glistens with excitement, revelation, and diversity. PACRIM 2004 is in this mould.
At once generalist and specialised, this congress brings together papers by keynote speakers
distinguished in fields as diverse as crustal geological process and ore genesis, resource
project finance and mining industry sustainability. Papers encompass the achievements of
the discoverers of new deposits, the developers of new geophysical and geochemical tools,
investigators from new frontiers whether geographic, technical or commercial; and
operators applying new techniques to old issues. The scope of this congress and the
applications of its content reach far beyond the coastal margins of the countries which
enjoy the metallic endowment associated with the Rim of Fire.
The assembly of so diverse a congress program would not have been possible without the
willing and dedicated work of our technical team, Adelaide based, but in total, resident
across the nation, and globally focussed. We are indebted to Gerard Bosch, Colin Brooks,
Chris Charles, John Dow, Ian Gould, Carolyn Ireland, Mike Lawrence, Brian Levet, Danielle
Martin, Dave Pollard, Dale Sims, Andrew Vigar, Kevin Wills, other occasional helpers, and
the reviewing team acknowledged elsewhere. The organisation of the congress and
coordination of this publication is the product of The AusIMM Events Department, and we
acknowledge the expertise, the commitment and the efficiency of the Melbourne-based
executive of Miriam Way, Alison McKenzie, Angie Spry, Kristy Burt and their office support
team.
On behalf of all delegates I thank the authors and presenters for their contribution to the
success of this event.
We are indebted to our many sponsors and exhibitors for their support.
Welcome to PACRIM 2004 and to Adelaide. On behalf of the organising team I wish you an
experience which is professionally and socially memorable.

Trevor J Ireland
Congress Chairman
 Sponsors
Gold Sponsor

Silver Sponsor Congress Dinner Welcome Cocktails

MINERALS
& ENERGY

Session Sponsor Official Publication Supporter

Exhibitors

MINERALS
& ENERGY
 Newmont Mining Corporation

Newmont Mining Corporation (NYSE & ASX: NEM; TSX: NMC) is the worlds largest gold
producer with significant assets or operations on five continents. Newmont is also engaged in the
exploration for and acquisition of gold properties in some of the worlds best gold districts. In
addition, Newmont is the worlds largest private sector precious metals royalty owner.
Newmont is well positioned as The Gold Company of choice for investors:
Worlds largest SEC-reportable gold reserves
No-hedging philosophy
Only S&P 500 gold stock
Newmont is dedicated to Creating Value With Every Ounce by:
Growing reserves
Strengthening our asset base
Increasing earnings per share
Improving our financial strength
Paying higher dividends
With the continued strength in the gold market, 2004 should be another exciting year for Newmont
with strong bottom-line earnings growth. The Company is optimistic about its ability to grow
reserves net of depletion at year-end 2004. For 2004, Newmont expects to sell between 7.0 million
and 7.2 million ounces of gold at total cash costs of between US$225 and US$235 per ounce.
Newmont offers attractive leverage to a rising gold price; for every US$10 change in the gold price,
Newmonts annual net income changes by approximately US$50 million and annual cash generated
by operating activities by approximately US$55 million, assuming all other factors remain constant.
Newmont operates core assets in the United States, Peru, Australia and Indonesia. The Company is
advancing two exploration projects in Ghana, which combined represented 11.9 million ounces of
reserves at year-end 2003, an increase of 143 per cent from year-end 2002, and are expected to become
the next core operating district. Newmont employs approximately 14,000 people worldwide
(approximately 3,500 in Australia) and is committed to the highest standards for environmental
management, health and safety for its employees and neighbouring communities.
Founded in 1921 in New York City, Newmont has been trading on the New York Stock Exchange
(NYSE) since 1925. In addition to the NYSE, Newmont trades on the Australian and Toronto stock
exchanges. Newmont is headquartered in Denver, Colorado.
 MINERALS
& ENERGY
Division of Minerals and Energy, PIRSA
The Minerals and Energy Division, Primary Industries and Resources South Australia facilitates mineral and
petroleum exploration and development. The division is focused on increasing the prosperity of South Australians
by ensuring responsible development of South Australias mineral and petroleum resources within a sustainable
framework. It provides geoscientific and specialist services as part of a process to acquire and update geological,
geophysical and engineering data. The group is also responsible for industry regulation, legislative development
and review, policy development and provision of advice.
In April 2004, the state government released A Plan for Accelerating Exploration PACE involving investment of
$15 million over five years.
Premier Mike Rann says South Australia stands on the brink of a boom in minerals exploration and this package
provides the incentive to accelerate the rate of exploration in this State.
The Minister for Mineral Resources Development Paul Holloway said Its the most comprehensive mining
exploration package ever developed in Australia bringing together exploration, education, research and
environmental considerations in a strategic and coordinated way.
PACE focuses on two areas critical for mineral and petroleum exploration in South Australia:
Facilitating access to land, under a multiple land use framework and in partnership with indigenous people
on Aboriginal lands
New pre-competitive geoscientific data to showcase the states mineral and petroleum prospectivity and
address critical uncertainties that impede exploration investment. This will demonstrate to the global
resources industry that South Australias mineral and energy endowment is competitive with other major
resource regions of the world and that there is a real chance of commercial success here (for further PACE
details visit www.minerals.pir.sa.gov.au/unlockingsa )
PIRSA has had an overwhelming response to the innovative Theme 2 Drilling Collaborative that matches
explorers dollar for dollar when they choose to drill mineral exploration holes in highly prospective regions in
South Australia. Applications were received from 47 mineral explorers hoping for a share in the $1.7 million made
available through the collaborative drilling program. Successful candidates will be announced in the second half of
September with drilling by some companies expected to commence soon after. Drilling and associated ancillary
costs as a result of this program are expected to total well in excess of $3.4 million (the total for drilling alone).
This program provides a major injection of funds for exploration and contributes to generating a wealth of new
geological understanding which increases the prospectivity potential of the region.
A world-class Centre of Excellence for research into successful exploration for minerals under covering rocks and
sediments has become a reality through PACE establishing the South Australian Chair in Mineral Exploration.
This will direct applied research into addressing opportunities to explore in areas under cover in South Australia
and bring discoveries forward. It will also result in the improved alignment between higher education and resource
education strategies and address a key recommendation of the Resources Task Force.
South Australia was recently ranked No.1 in the world (with a score of 100 per cent) in the Geological databases
category (Fraser Institute Annual Survey of Mining Companies 2003-04), with all explorers considering that our
data encourages exploration investment. Many of these databases including geophysical data can be accessed and
downloaded through the South Australian Resources Information Geoserver(SARIG) a world leader in online
geoscientific data availability. No third party software is required, simply access SARIG through the website
www.minerals.pir.sa.gov.au/sarig
The PACE program confirms the state governments commitment to developing its mineral and petroleum
resources and when coupled with Adelaides recent recognition the most competitive city for business in Australia
(KPMGs Competitive Alternatives 2004 Survey) provide compelling reasons to explore in South Australia.
We want and welcome mining in South Australia.

PowerPoint Presentation
 Contents
Keynote Addresses
Project Management for All Seasons Financial Risk Mitigation in a Q G Amos and O C Aitken 3
Challenging World

The Times They Are (Always) A-Changin S E Close 9

El Teniente and Rio Blanco Porphyry Cu-Mo Deposits, Chile D R Cooke, J Cannell, 13
Giant Ore Formation in an Active Continental Margin P H Frikken, P Hollings and
J L Walshe

Epithermal and Porphyry Gold Geological Models G Corbett 15

Digging In and Digging Out Where Weve Been, Where We Are and P Cristofani 25
Where Were Going

Meeting Quarter-Century Exploration Requirements in a Quarterly M S Enders and R A Leveille 27

Results Business World

Advances in Regolith Research A CRC LEME Perspective R D Gee and R R Anand 29

Overview of Mineral Project Risk Issues and Role of Mineral Industry M J Lawrence 45
Professionals (Especially Consultants) in Risk Reduction Strategies

The Many Dimensions of Sustainability I Lowe 53

New Mining Projects Expectations and Outcomes P L McCarthy 57

The International Mining Industry Linking the Upstream N Miskelly 63

Mineralisation With the Downstream Money

Modern Solutions to Old Day Mining Problems B Morrison 69

Issues Confronting Major Expansions and New Developments R Shaw 71

Recent Advances in Geophysical Tools R J Smith 77

Successful Management of Mining Companies What Have We A Trench 87

th st
Learned From the 16 Century to the 21 Century?

Deposit Models and Regional Reviews

Gold Distribution and Geochemistry at Batu Hijau, Indonesia J Arif, T Baker, C Clode, B Harlan 93
Geological and Economic Implications for Porphyry Copper-Gold and A Perdanakusumah
Deposits

Gold Exploration of the Phuoc Son Tam Ky Suture, Central Vietnam M J Banks, R H Murfitt, 95
A Case Study N N Quynh, L V Hai
Geology of the Gold Deposits of the Yanacocha District, Northern Peru P D Bell, J G Gmez, C E Loayza 105
and R M Pinto

Common Geological Characteristics of Prominent Hill and Olympic A Belperio and H Freeman 115
Dam Implications for Iron Oxide Copper-Gold Exploration Models

The Sams Creek Peralkaline Granite Hosted Gold Deposit, Northwest R L Brathwaite and K Faure 127
Nelson, New Zealand A New Variant on Alkaline Intrusion-Related
Gold Deposits

Disseminated Gold Mineralisation in a Schist-Hosted Mesothermal D Craw, D J MacKenzie and 135

Deposit, Macraes Mine, Otago, New Zealand B Petrie

The Jianchaling Mine, Shaanxi Province, Peoples Republic of China M Erceg, K Fan and 143
A Carlin-Type Gold Deposit? T Leach

Distribution of Alteration and Mineralisation in Northern Carlin Trend T M Leach 153

Gold Deposits, Nevada

Recent Investigations of Submarine Hydrothermal Activity in Indonesia T F McConachy, H Permana, 161

R A Binns, I Zulkarnain,
J M Parr, C J Yeats, N D Hananto,
B Priadi, S Burhanuddin and
E P Utomo

A Geological Review of the Tampakan Copper-Gold Deposit, Southern C Middleton, A Buenavista, 173
Mindanao, Philippines B Rohrlach, J Gonzalez, L Subang
and G Moreno

The Tintina Gold Province, Alaska and Yukon New World-Class S H Mueller, R J Goldfarb, 189
Gold Resources and Their Sustainable Development C J R Hart, J L Mair, E E Marsh
and C S Rombach

Data Metallogenica Around the Pacific Rim and Across the World K N OSullivan and A D T Goode 199

Discovery Techniques and Regolith

Seeing Through Alluvial Gold Fine Fraction Stream Sediment M C Arundell and S G Gatehouse 207
Sampling in the Sofala Area, Central New South Wales

HoistEM A New Airborne Electromagnetic System G W Boyd 211

Groundwater Geochemistry as a Tool for Mineral Exploration Under P de Caritat and D Kirste 219
Cover The Curnamona Province

221
TM
The Falcon Airborne Gravity System as an Exploration Driver in P Harman, D Isles and
Australia C Anderson

Automated Mineralogical Logging of Core From the Emmie Bluff, Iron J Huntington, A Mauger, 223
Oxide Copper-Gold Prospect, South Australia R Skirrow, E Bastrakov,
P Connor, P Mason, J Keeling,
D Coward, M Berman, R Phillips,
L Whitbourn and P Heithersay
New Ways to Explore Through the Regolith in Western New South K G McQueen and A McRae 231
Wales

Prospectivity Mapping Using GIS With Publicly Available Earth G A Partington and M J Sale 239
Science Data A New Targeting Tool Being Successfully Used for
Exploration in New Zealand

Recent Developments in Electrical Exploration T J Ritchie 251

Weathering of Base Metal Deposits and Implications for Exploration in K M Scott 257
the Lachlan Fold Belt

A Spatially Enabling Relational Geologic Database of Java Island, L D Setijadji and K Watanabe 263
Western Sunda Arc of Indonesia A Step to the Geoscience and
Mining Data Model

The Controls on Element Distribution in the Regolith at the Portia and K P Tan, P de Caritat, K M Scott, 273
North Portia Prospects, Curnamona, South Australia R A Eggleton and D Kirste

Compositional Variations of White Mica in the Footwall Hydrothermal K Yang, J F Huntington, 283
Alteration System of the Hellyer Zn-Pb Deposit, Tasmania K M Scott and P Mason

Exploration and Mining Geology Case Studies

Reducing Grade Uncertainty in High-Nugget Effect Gold Veins S C Dominy and G F Johansen 291
Application of Geological and Geochemical Proxies

Hydrothermal Alteration at the Bowdens Silver Deposit, NSW J Elliot, T Leach and I Pringle 303

Results From Induced Polarisation Surveys Over the Beruang P J Elliott 307
Copper-Gold Deposit in Central Kalimantan, Indonesia

A Solution to Grade Estimation in a High Nugget Environment G F Johansen 317

The Bendigo Experience

A Review of the Kapit Orebody, Lihir Island Group, Papua New Guinea R P Kidd and J R Robinson 323

The Tooloom Gold Project, NSW Forgotten But Not Gone! R M D Meares, G G Lowder, 333
Rediscovering a 147-Year-Old Gold Field M J White, B A Wake and
M D Vickers

The Perseverance Gold Deposit The Next Step at Peak S Munro and R Berthelsen 339

Geology, Mining and Grade Control at the Batu Hijau Porphyry A Prihananto 345
Copper-Gold Deposits, Sumbawa, Indonesia

The Bong Mieu Gold Project, Phuoc Son Tam Ky Suture, Central N N Quynh, R H Murfitt, 347
Vietnam T Sirinawin and W Shywolup

Discovery of the Toguraci Epithermal Au-Ag Deposits, Gosowong T H Richards and 359
Goldfield, Halmahera Island, East Indonesia M D Basuki Dwi Priyono

Discovery Performance of the Western World Gold Industry Over the R C Schodde 367
Period 1985 - 2003
The Hillgrove Gold-Antimony-Tungsten District, NSW, Australia C K Switzer, P M Ashley, 381
B Hooper and B Roach

The Discovery and Early History of the Mt Leyshon Gold Deposit, G S Teale and J E Lynch 385
North Queensland

A Search for Gold in Indonesian New Guinea M R Thirnbeck 391

Nagarze Gold Porphyry, Tibet PRC D S Tyrwhitt 401

Mining in Fiji B Wesson 403

A Three-Dimensional Structural Interpretation of the Olympic Dam H Widdup, T Fouet, J Hodgkison, 417
Deposit Implications for Mine Planning and Exploration T C McCuaig and J Miller

Industry Challenges Globalisation, Communities,

Developing Countries, Financing Exploration
Exploration and Project Development Risk Reduction Approaches in the A H Osman 429
PacRim The Indonesian Experience

Industrial Minerals Opportunities and Making Money in AusAsia D Turvey 435

Looking to the Future Sustainability, Education and Skills

An Environmental Collaborative Research Program Field Studies A Black, T Clemens and 443
Contributing to the Sustainability of New Zealands Mineral Industry D Trumm

Water Quality Trigger Values for Suspended Mud in the Jabiluka Mine K G Evans, D R Moliere, 447
Receiving Catchment, NT, Australia M J Saynor, W D Erskine and
M G Bellio

Pioneering Community Collaboration in Mineral Exploration C M Horn and I B Miller 453

Southern Flinders Ranges

Environmental Auditing as a Driver for Continuous Improvement in K W Perry 459

Environmental Performance

Transition Issues Discovery to Profitable Operations

Sampling and Analysis Without Fear or Favour C C Brooks 465

Assessing Ore Loss and Ore Dilution Quantifying the Effects of P C Dare-Bryan and P A Dowd 473
Geological and Technical Uncertainties

Authors Index 487

Keynote Addresses

Project Management for All Seasons Financial Risk Mitigation in a Q G Amos and O C Aitken 3
Challenging World

The Times They Are (Always) A-Changin S E Close 9

El Teniente and Rio Blanco Porphyry Cu-Mo Deposits, Chile D R Cooke, J Cannell, 13
Giant Ore Formation in an Active Continental Margin P H Frikken, P Hollings and
J L Walshe

Epithermal and Porphyry Gold Geological Models G Corbett 15

Digging In and Digging Out Where Weve Been, Where We Are and P Cristofani 25
Where Were Going

Meeting Quarter-Century Exploration Requirements in a Quarterly Results M S Enders and 27

Business World R A Leveille

Advances in Regolith Research A CRC LEME Perspective R D Gee and R R Anand 29

Overview of Mineral Project Risk Issues and Role of Mineral Industry M J Lawrence 45
Professionals (Especially Consultants) in Risk Reduction Strategies

The Many Dimensions of Sustainability I Lowe 53

New Mining Projects Expectations and Outcomes P L McCarthy 57

The International Mining Industry Linking the Upstream Mineralisation N Miskelly 63

With the Downstream Money

Modern Solutions to Old Day Mining Problems B Morrison 69

Issues Confronting Major Expansions and New Developments R Shaw 71

Recent Advances in Geophysical Tools R J Smith 77

Successful Management of Mining Companies What Have We Learned A Trench 87

From the 16th Century to the 21st Century?
Project Management for All Seasons Financial Risk Mitigation
in a Challenging World
Q G Amos1 and O C Aitken2
ABSTRACT
Mining project developers are committed to taking on all the challenges
and obligations of financial risk management. It is interesting that our
current models of project finance have not changed markedly in principle
for some 700 years, yet the rules which prepare the way for effective
project assessment and subsequent loan management still do not seem to
be well understood by project or corporate developers.
Probability and gravity fill the opposing trays of the golden balance of
financial risk. Understanding the relationship between these opposing, but
intimately related elements is critically important for successful risk
management. It is important to move beyond the qualitative approach and
use quantitative methodologies to measure the financial impact of the
occurrence of each major risk. We live in a world where the only constant
is change and thus it is not surprising that a bankers life is still governed
by assessing, measuring and pricing risk.
Banks are attracted to funding well managed companies with strong
cash flow projects. But we are more risk averse compared with the credit
appetites of late last century. This attitudinal shift has resulted in tougher
term sheets for the early 21st century, encompassing more conditions,
warranties, undertakings and stronger cash sweep provisions.
Banks require well balanced portfolios to achieve an acceptable spread
of risk, thus there is a more restricted market for small company lending
than that available to large multi-project corporate entities with less
concentrated risk profiles. Because of this focus, junior company debt
finance remains comparatively expensive and especially so for more
challenging geographical and political environments.
Understanding the most important components of financial risk for
each particular project development and establishing methods for the
probabilistic measure of adverse outcomes will allow a company the
opportunity to place a comparative value on each opportunity. Whilst
more commonly used in financial treasury analysis to address the
assessment of complex derivative products, this methodology should
assist in bringing together at least some of the many complex elements in
project development and allow project value comparisons.
Understanding and using the tools of risk management to ensure the
continued financial strength of the company will serve to best protect
shareholders from the vicissitudes of the changing political, corporate and
financial currents. To achieve excellence in their chosen operating
environments companies must focus on developing and maintaining
financial resilience through active and informed risk management.
It should be obvious that banks will seek to provide funds to those
companies that they perceive to be in charge of their own destinies and
are best prepared to withstand periods of adverse conditions on a number
of fronts. Banks are not interested in assisting those companies that
exhibit an archaic trust in the good auspices of the Fates because it is
more likely that the Sirens will lure them onto the sharp rocks of Scylla
or under the rock of Charybdis.
INTRODUCTION
Mining project developers are committed to taking on all the
challenges and obligations of financial risk management in many
different locations and social environments around the world.
This is of course in addition to managing all the technical,
environmental, legal and macro economic risks, which are
inherent in any project development. How can life be made less
risky for the intrepid explorer and project developer?
1. FAusIMM, HSBC Global Metals Group, Level 10, 580 George
Street, Sydney NSW 2000. E-mail: quentinamos@hsbc.com.au
2. HSBC Global Metals Group, Level 10, 580 George Street, Sydney
NSW 2000. E-mail: oliviaaitken@hsbc.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Understanding the relationship between the opposing risk
elements of probability and gravity is critically important for
successful risk management particularly when an explorer steps
away from a well known home base. There have been a great
many failures of enthusiastic would be project developers over
the last 20 years, we believe that the sad fact is that most of these
failures could have been foreseen and to a greater or lesser extent
mitigated if comprehensive databases had been available and
sound statistical analysis had been applied to that data.
Qualitative analyses such as that espoused in AS4360 have
proved very useful in focussing effort towards addressing risk
elements affecting mining projects. Our contention is that it is
imperative to move beyond the accepted qualitative approach to
risk analysis and use quantitative methodologies to more
accurately measure the likelihood and financial impact of the
occurrence of each major risk.
Every company needs to be well equipped technically to
ensure success in project development; this enterprise is
challenge enough without having to scramble for much needed
funds. Whilst many factors govern shareholder participation as
either speculator or investor it is still the company directors duty
as responsible stewards to minimise the risk profile and thus
provide the best chance for success. It should also be obvious
that banks are far more likely to provide funds to successful risk
managers, well managed companies with strong cash flow
projects. Therefore, it is in the best interests of any project
developer to be able to demonstrate a comprehensive
understanding and assessment of the risk elements attached to
each particular development to its shareholders and potential
debt financiers.
From a lenders viewpoint there is a lot of competition for
project funding and not simply from mining related opportunities
and all banks will be most interested in the companies that can
best demonstrate their capability to repay funds lent. A company
needs to demonstrate the impeccable nature of its technical and
commercial credentials and thus necessarily its comprehensive
risk assessment strategies, which will account for the many
internal and external elements, which may negatively impact
project development.
Mining project development is a difficult task in any
environment and the more so when the particular geo-political
environment is less well understood. The various opportunities
that may present themselves should be subject to rigorous and
methodical analysis and not simply contrasted based upon
subjective and often confined data from sources, which may have
conflicting experiences.
Understanding the most important components of financial
risk for each particular project development and establishing
methods for the probabilistic measure of adverse outcomes will
allow a company the opportunity to place a comparative value on
each opportunity. This is a goal that is well worth the time and
strenuous efforts of every participant in the mining project
development arena to work towards.
THE SEASONS OF RISK
Risk should be regarded as all pervasive, ever present and
multi-variant, very much akin to the weather on this fair planet
and just as meteorologists must painstakingly gather together
3
Q G AMOS and O C AITKEN
vast quantities of data to form a semi statistically reliable
database from which predictions can ever so tentatively be made,
so must miners collect their own risk data. There is nothing surer
than the observation that risk perception will increase with the
availability of data and that this data is often the data of
experience, the product of long years of hard work by many
parties in a multitude of environments.
Just as certain weather patterns are peculiar to specific
geographies or geographical regions there are particular patterns
and degrees of risk associated with mining technical,
environmental, social, political and legal environments. How do
we recognise the spring, summer, autumn and winter of risk? If it
looks like winter may come on fast then perhaps an ostensibly
attractive project should be avoided. It is very important to have
the best prediction tools available.
A very good database will assist a mining company in the
measurement of likely outcomes for specific project
developments and corporate directions. This then is the target
objective, to transfer the collective knowledge and experience of
many individuals to an effective database which can be
statistically analysed and shared by all users many of whom will
not have had 50 years of individual project development
experience in far flung locations. It is a database that must tap
into and assemble a vast array of knowledge and thence be able
to separate and somehow distil an objective record from
subjective observations and opinions.
Let us address the seasons of risk. There is no doubt that when
the final pre-development decision has been taken to move
forward into project construction and finance has been secured
there is an all pervasive air of enthusiasm, the very breath of
spring, with the board and senior management basking in the
warm sunlight of the promise of great things to come. Everyone,
except some embittered, old and battle hardened generals is filled
with enthusiasm and the conviction that all the big issues have
been addressed and that all that is really needed now is hard
work to get things up and running ahead of schedule and under
budget. The old generals have seen all this before; in fact they
were once participants in these raptures of spring before they
learnt to their detriment the reality of the so-called Murphys
Law.
The euphoria of spring soon passes and the project moves into
high summer, project managers and contractors, lots of
sub-contractors, all manner of preparation and building
development, consultants everywhere and generally the
schedules are being met except for a mill shell delivery problem
here, a difficult creek crossing there and an unfortunate lack of
planning in one of the contractors order books. The old generals
look around and start to audibly grumble, but nobody really
wants to hear a bunch of sour old whingers.
Things are falling behind, budgets are beginning to look too
lean, the owner and the contractors are falling out over slips in
the schedule and workforce is starting to agitate for improved
conditions because the timeframe for construction is extending.
The bankers are growing restless. Autumn is here. Any smart
squirrel knows that when autumn comes it is time to gather nuts
and prepare for winter, in other words take the time and effort to
get your house in order and deal with the risk issues even if that
is expensive so that you can weather the coming storm. By this
stage the board is beginning to listen to the old hands, but it is in
a tough position trying to fight a rearguard action to get back on
schedule before cash sweep provisions cut into loan availability.
Perhaps even more concerning, annual report time is looming on
the horizon.
Winter has come. The project is six months behind schedule,
the grinding mills are not performing to specification, recovery is
five per cent below budget and the rock is harder than originally
thought. The technical difficulties are coupled with adverse
public reaction to dust and noise and the bankers are cash
4 Adelaide, SA, 19 - 22 September 2004
sweeping and are likely to have to restructure repayment
schedules. The old generals are vindicated, but it is too late and
the project must now enter an expensive early recovery phase
just to get back on track, and all of this has been captured in the
annual report and presented to shareholders at the AGM in full
glowing colour. The share price has crashed and financial
flexibility is now severely limited.
Much of this trauma could have been avoided by the
availability of a comprehensive database coupled with sensible
statistical analysis of that data. This style of risk analysis may
have suggested that the project was doomed from the start or
should only have been considered when the company was very
financially resilient.
BEST PRACTICE IS THE BEST HABIT OF BEST
MANAGEMENT
When it comes to risk management what constitutes best
practice? In essence it is preparedness, having a deep
understanding of the importance of risk recognition and
measurement and being equipped with the appropriate tools to
collect data both historical and current from all of the associated
risk areas that are likely to impact project development. These
data must be categorised and thence statistically analysed and
weighted for both probability and gravity of occurrence. But
herein lies the most important element which is encapsulated by
Murphys Law what needs to be measured is not just the
potential result of a single event impact, but the interrelationship
between all of the risk elements and how one event may impact
concurrently upon the rest.
It can be imagined that a totally comprehensive mathematical
model which can encompass the mammoth task of comparing
and weighting all possible risk outcomes will take a great deal of
time and money to construct and still will not of itself provide
surety of a perfect result, that is, total preparedness. There will
always be some unknown factors, which may in rare cases cause
severe stress to project development, but it should be possible to
avoid or mitigate identifiable and likely impacts.
We believe that what is needed, as a useful management tool is
a comparatively small interactive mathematical model, which
encapsulates the medium to major risks, and can estimate their
likelihood of occurrence in order to give a reasonable basis for
making development decisions. This model should be able to be
interrogated to provide a workingmans guide as to the most
likely risk occurrences within each opportunity and which of
several investment and development opportunities presents the
comparative lowest general risk profile. In this way an
investment opportunity can be given a nominal risk weighting
which can be compared with an appropriate internal hurdle bar,
which will necessarily vary from company to company. The
questions, which must be answered, are, do we want to attempt
any of these projects, are they rewarding enough to compensate
for the perceived risks and which of these opportunities presents
the most favourable risk/reward profile?
Step 1 is to collect, collate and mathematically model the data
that pertain to each major risk category. This will involve getting
input from data providing organisations and experienced parties
with regard to as many previous projects as possible. Each
project will have the same general risk classes, but within these
classes specific data will reflect widely different outcomes. For
instance we can regard technical risk as a general class within
which there is geological, geotechnical, mining and metallurgical
risks as sub categories, and within the sub category of
metallurgical risk there are specific categories of crushing,
grinding, ore feed, material dissolution, recovery and concentrate
production. Particular styles of geological regimes and
mineralisation will be impacted by differing degrees of risk in
each of the major and sub categories discussed above.
PACRIM 2004
 PROJECT MANAGEMENT FOR ALL SEASONS FINANCIAL RISK MITIGATION IN A CHALLENGING WORLD

The company must have a reasonably reliable method to gauge A step-by-step statistically logical approach is required to add
which of these risk categories may be a showstopper or have a a quantitative dimension to current qualitative project assessment
seriously detrimental impact on cash flow. Swaying tropical models. A comprehensive database of worldwide mining projects
palms may be attractive, the people friendly, the investment with documented risk event information is essential for an
regime positive, labour rates low and the ore grade better than all-encompassing quantitative approach to add risk assessment
most, but if the mining conditions in similar orebodies are value to the total project evaluation process.
sometimes abysmal at depth, the tropical storms at times wash Information collected on all projects irrelevant of success or
out the roads for three months of each year and there may be a failure, size and profitability would be most beneficial for
related need to maintain a full storehouse then maybe this project
providing an unbiased database tool, a solid platform for
is not the gem everybody originally thought. The risks must be
comparisons that can be extrapolated across the commodity
measured and quantified as competently as possible because the
group and industry as a whole.
board will not want to pass up an excellent opportunity but
neither will it wish to see funds pour down a cash drain. This is a There is a clear need for all the involved parties, to endorse
critical cost benefit analysis for any organisation. and financially support a research organisation or similar body to
The traditional method of assessing opportunities has been to develop a comprehensive database encompassing the all the
construct a DCF style cash flow model and stress test the base important risk aspects of mining. The human safety, production
level assumptions by using a variety of sensitivity analyses such environmental, social and financial impacts of each specific risk
combining a five percent decrement in grade with a five percent event can then be classified and recorded. Some of the key risks
fall in recovery and a ten percent increase in operating costs. that need to be documented for each mining project are set out in
This methodology is at best an educated guess. It seldom Table 1.
captures sufficient relevant data to estimate statistically These specific risks fit within classes of events of varying
acceptable parameters for any rise or fall in base numbers and it impact levels (Behre Dolbear Australia Pty Ltd, 2003 adapted
certainly does not attempt to estimate the likely occurrence of from AS4360 Appendix E). The aim is to identify the gravity of
any variation. Such methodology is seen as better than nothing occurrence.
but that may not be so and indeed it may be severely misleading Catastrophic very large financial loss (>$5 M); death or
giving a false sense of security to all parties involved in the serious injury to multiple persons; major loss of plant resulting in
project and related financial transactions. There is nothing so >3 months loss of production capability; toxic environmental
dangerous as believing in your own ability to correctly prophesy. release off-site with detrimental effect.
Best practice requires responsible behaviour and it is not Major major financial loss ($1 M - $5 M); death or serious
responsible to take a gamblers approach to project development injury to multiple persons; extensive loss of plant resulting in one
using stakeholders money when tools are available to mitigate
to three months loss of production capability; offsite
risk.
environmental release without detrimental effect or on-site
release with detrimental effect.
GATHERING ACORNS FOR WINTER
If we accept that Step 1 is to collect, collate and mathematically
model the base data, which pertain to each major risk category TABLE 1
then how do we best achieve this? For the purposes of our Significant mine project risks.
demonstration example we have used the commodity type as
central to our decision process although the data sets could be 1. Country 6. Environmental
organised around other central pivots such as processing sovereign effluent issues
methodology for instance, if this was identified as a tactical legal tailings disposal
advantage for the company. We selected gold because this permitting issues conservation groups
commodity presents us with the largest database and thus the local sociological factors
conclusions drawn are better based because of the data climate
population size. 2. Reserves and resources 7. Costs
We have filtered the data sets based upon operating costs grade/tonnage capital
associated with producing projects. Our data set is not perfect, estimation procedure operating
for instance it does not account for unlisted companies and it is drilling and sampling
biased towards medium to larger projects. Certainly we regard geological and structural
this example as only a first and very small step towards achieving judgments
a useful and practical mathematical model for objective risk
3. Mining 8. Revenues
analysis and by no means even vaguely approaching the desired
design exchange rate
end point. In this sense we are involving the reader in a
planning metal price
developing process, but one, which we see as very valuable in dilution
understanding the risk parameters, associated with embarking labour
upon uncertain endeavours. technology
It is our attempt to generate a first pass guide and ready
reckoner which may serve to prevent over enthusiasm for one or 4. Processing 9. Management
two project aspects which in turn drive the company to make a plant performance communications
flawed or financially fatal decision. Undoubtedly the final design qualifications
models will need a lot of development, but eliminating the gross recovery structure
technology
and identifiable risks is an important step for sensible direction
and use of stakeholder funds and for gaining the support of 5. Infrastructure
shareholders and debt providers alike. Such a model should power
primarily provide a focus for understanding the risks that may water
seriously deteriorate project value and the timing and the scale of transport
that deterioration.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 5

Q G AMOS and O C AITKEN

Moderate high financial loss ($0.1 M - $1 M); serious injury

to multiple persons; moderate loss of plant resulting in one week TABLE 4
to one month loss of production capability; onsite environmental Classified impact level.
release contained with assistance without causing long-term Project A Samples Catastrophic Major Moderate
detrimental effect. Mine 1 Y
A very important issue in providing a comprehensive database Mine 2
will be the release of corporate data. How many current mining
companies are willing to divulge and classify their risk events Mine 3 Y
encountered that had a moderate or greater financial impact on Mine 4 Y
the company? We believe that it is vital for industry growth and Mine 5 Y
success that all but proprietary classified information is available
: : : :
to be shared.
: : : :
PROPOSED METHODOLOGY Mine 268

A selection tool filters the database resulting in sample

observations comparable to the project that is being evaluated. Total = 120 16 31 73
The major filters being commodity, country, mine type,
production and process route. However, if the database were Australian Mineral Economics (AME) kindly supported
large enough the filters could be more precise, for example HSBCs investigation by providing anonymous basic gold
including government development policy, social history, operating information and cash costs from their current database
regulatory intervention et cetera all the way down to grade, from 1998 to 2002 to give a real example of what comparisons
tonnage and specific metallurgy. From this point the user can can be made using the same logic, but obviously without a global
identify risk events that have damaged analogous projects (the risk database being available.
probability measure) and how financially detrimental they were
It is important to note this database is market focused towards
to that projects survival (the gravity measure).
major profitable resource projects and is therefore positively
To give a theoretical idea of how a company might benefit skewed to the upper end of the market. It is also important to
from historical comparisons consider the following scenario: note the number of observations or mines is limited.
The CEO of a gold company is interested in investing in one of Of the 124 mines with production and cash cost information
the two projects set out in Table 2. provided, there were only two countries with 25 or more
observations when filtered by commodity, country and mine
TABLE 2 status (Tables 5 and 6).
Project comparison.
Name Project A Project B TABLE 5
Mine selection.
Commodity Gold Gold
Country Australia South Africa Country Number of mines
Mine type UG/OP UG Australia 30
Production 100 000 oz 300 000 oz South Africa 25
Process route CIP/CIL Electrowinning CIP/CIL Electrowinning Total 55

To compare these two projects, the global risk database would TABLE 6
be consulted and when filtered could give a sample of mines with Summary statistics from 1998-2002.
similar characteristics as set out in Table 2. The sample would
then highlight the probabilistic measure of certain adverse Australia South Africa
outcomes as set out in Table 3. Number of mines 30 25
Number of underground 6 24
TABLE 3 Number of open pit or cast 8 0
Probabilistic measure of adverse outcomes. Number of OP/OC and UG 16 1
Type of risk Project A samples Project B samples Average operating cash costs $190.69 $229.86
Country 2% 4% 95% confidence interval t stat 1.7080 1.7110
Reserve 45 % 29 % Lower limit $176.32 $212.13
Mining 10 % 12 % Upper limit $205.06 $247.59
: : : Average ore treated (t) 2 354 927 2 419 299
Etc Lower limit 1 456 988 1 764 530
Upper limit 3 252 866 3 074 069
Then each risk percentage can be drilled down to determine Average gold grade (g/t) 5.19 7.22
the financial impact these events had on projects from the
Lower limit 3.98 6.16
sample. For example, 45 per cent risk of Reserve failure for a
project type A (Table 4). Upper limit 6.40 8.28
Therefore, from a sample size of 268 mines with similar Average gold production (oz) 212 496 463 452
characteristics to Project A, 45 per cent or 120 mines Lower limit 166 737 348 433
experienced a grade failure, 13 per cent of which, had
Upper limit 258 255 578 471
catastrophic consequences for the company.

6 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 PROJECT MANAGEMENT FOR ALL SEASONS FINANCIAL RISK MITIGATION IN A CHALLENGING WORLD

In this example the average operating cash costs per ounce for
a mine in Australia is $190.69, which is less than that of South
Africa at $229.86 per ounce. Of the mines in the South African
sample, 96 per cent are solely underground mines. Of the mines
in the Australian sample only 20 per cent are solely underground
mines.
Underground mining is generally technically more complex
than near surface work. It is more difficult to gain access to the
deposit and to control the behaviour of the rock mass
surrounding the mine openings. The dimensions of the mine
openings constrain equipment size. It is also necessary to provide
adequate ventilation and men cannot be withdrawn as rapidly as
from surface mines. Underground mining is generally more
hazardous and requires an even greater emphasis on safety and
training. For these reasons the unit costs of excavating and
removing material from an underground mine are normally
higher than for an equivalent surface operation.
Another issue having a small affect on the average operating
cash costs is the existence of copper or silver by-product credits.
In Australia, ten per cent of the samples cash costs are reduced
via by-product credits, however, in the South African sample
there are no by-product credits.
Therefore the combination of the type of mining and the
metallurgical differences could account for the higher operating
cash costs in South Africa. Although this may seem like an
obvious conclusion to draw, it does highlight how logical
comparisons can be made between two possible projects, using
historical information.
Other conclusions that can be made from the operational
information are highlighted as examples in Figure 1. The average
gold production per mine in South Africa (463 452 ounces per
year) is more than double that of Australia (212 496 ounces per
year). Fewer but larger mines in South Africa and the higher
average grade in South Africa (7.22 g/t) can explain this since
the average amount of ore treated is similar to that of Australia
(Figure 1).
CONCLUSION
Our observations derived from the real example cited here are
simple and a long way from a thorough risk conclusion based on
historical comparisons. However, they do demonstrate how
historical data can be used to draw relevant conclusions and with
a risk focused database probability of certain risk events and
their respective consequences can be highlighted. These initial
observations are the first tottering steps of a child. As in all
research we must learn to walk, then run, climb and leap.
The shortcomings of documented risk information are evident,
however, we believe that this type of quantitative approach is
necessary to move beyond inadequate solely qualitative
conclusions. With industry support and sponsorship, a risk based
unbiased global database is feasible and in turn the industry as a
whole would benefit from having a much needed tool to evaluate
and compare projects and reduce risk.
The business of risk minimisation is a serious and time
consuming endeavour, which must be well supported by
excellent databases, which are readily accessible.
Unquestionably a vast amount of data exists in research
companies and organisations and within the vaults of the various
departments of mines and corporations. The tendency is that the
more recent the data, the more likely it is to be found either via
net connection or on the particular organisations LAN. What is
clearly lacking is a coordinated approach to capturing all
available data within one database, which could be shared by all
mining companies.
Who would be responsible for such a large effort and just as
importantly how could it be paid for are open questions at this
stage, but the undoubted value of a successful outcome should be
obvious. The Australian mining industry is at the forefront of the
global business in most respects and risk minimisation for all
companies is in the direct best interest of all participants. At
present much data is shared between various companies within
particular regions and regimes, what is required is a
formalisation of this laudable practice.

Average Total Ore Treated Average Operating Cash Costs

4,000,000 $300.00
$250.00
3,000,000 $200.00
USD/oz

Australia Australia
2,000,000 $150.00
t

South Africa $100.00 South Africa

1,000,000 $50.00
0 $0.00
1998
1999
2000
2001
2002

98

00

02
19

20

20

Average Mine Production Average Total Head Grade

600,000 8.00
500,000 6.00
Oz of Au

400,000 Australia Australia

4.00
g/t

300,000
200,000 South Africa South Africa
2.00
100,000
0 0.00
98

00

02

98

99
00
01

02
19

20

20

19

19
20
20

20

FIG 1 - Australia/South Africa graphic comparisons.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 7

Q G AMOS and O C AITKEN
As an example there is a far greater chance of exerting
significant pressure for change and investment climate
improvements in developing nations if there is a well established
and large expatriate mining community contributing significantly
to a countrys GDP. To support greater participation within
offshore locations there has to be a lot more encouragement than
just a few exciting assays. Understanding the risks associated
with stepping into someone elses backyard and being able to see
and most importantly, use the successes and failures of other
participants (albeit anonymously) vastly increases the chance of
building better investment infrastructure at all levels through
greater market participation.
Our fundamental tenet espoused in this paper is that risk
measurement can and should be a quantitative exercise not
simply a qualitative one. Hard numbers speak louder and more
clearly than simple anecdotal evidence, particularly to potential
debt providers. We contend that all project development
stakeholders are standing on the bottom step of a tall ladder, but
that does not mean that we should be afraid to climb or that the
rungs are too slippery, it just means that the climb will be
strenuous. It is time to better use the tools available to us with
our current technology to sharpen the play in the exploration and
mine development game and move it onto a safer playing field
thus increasing the chance of scoring vital goals.
8 Adelaide, SA, 19 - 22 September 2004
We believe that risk can be far more effectively measured and
quantified than in the past and that this in turn will lead to greater
chances of financial support from shareholders and potential debt
providers for project development. Ultimately such a positive
outcome must result in greater corporate growth and a more
successful mining industry.
REFERENCES
Amos, Q G, 2001. Resources and risk A lenders view, in Mineral
Resource and Ore Reserve Estimation The AusIMM Guide to Good
Practice (Ed: A C Edwards), pp 511-518 (The Australasian Institute
of Mining and Metallurgy: Melbourne).
Amos, Q, 2003. Project development, financial risk and corporate
obligations, in Proceedings Mining Risk Management Conference
2003, p 3 (The Australasian Institute of Mining and Metallurgy:
Melbourne).
Amos, Q and Breaden, P, 2001. The JORC code A bankers view, in
The AusIMM Bulletin, 6:43-47.
JORC, 1999. Australasian Code for Reporting of Mineral Resources and
Ore Reserves (The JORC Code) The Joint Ore Reserves Committee
of The Australasian Institute of Mining and Metallurgy, Australian
Institute of Geoscientists and Minerals Council of Australia, 16 pp.
Joint Standards Australia/Standards New Zealand, 1999. Risk
Management.
Mase Westpac Limited, 1990. The Winning of Gold.
PACRIM 2004
The Times They Are (Always) A-Changin
S E Close1
The papers in this section of the conference and indeed the
conference as a whole will rightly focus on the recent the state
of play around the Pacific Rim.
However, to set the scene I propose to take a longer-term
perspective. In so doing I wish to illustrate and emphasise that
change is normal and natural to paraphrase Bob Dylans 1963
classic, The times they are (always) a-changin .
The challenge is how to cope with such change, in an industry,
a corporate and a personal sense.
A further challenge for geologists, both exploration and
mine-based, and indeed for anyone, is to understand where they
fit in the broader picture.
For several reasons it is appropriate first to look briefly at the
minerals industry in the Pacific Rim around a century ago.
It was here in Adelaide in April 1893 that the first general
meeting of the Australasian Institute of Mining Engineers (or
The Australasian Institute of Mining and Metallurgy as it later
became) took place and where the Centenary Meeting was held
in 1993.
The late 1800s and early 1900s was a period of extraordinary
change in the minerals industry regionally and worldwide. The
industry benefited from the economies of scale provided by
improved mechanisation. It further benefited from the
introduction of technically qualified professionals, such as
chemists, metallurgists and mining engineers, although the
usefulness of geologists had yet to be recognised commercially.
With the rise of such professionals, technical societies formed
and flourished.
At the same time international financial markets and trade
were expanding, a gold standard was widely supported and the
financing of mining ventures was well established. Mining was
an international business and its people travelled widely.
As a consequence of the better-educated mining professional,
the ready exchange of knowledge and the availability of finance,
new processing techniques such as pyritic smelting, the cyanide
process and flotation were developed. These, together with much
improved materials handling in the mining and processing
phases, dramatically changed the economics of the minerals
industry.
In Australia, for example, in 1903 the Second Gold Boom
reached its peak. Gold accounted for around two thirds of the
value of minerals produced in that year. Coal contributed some
ten per cent of the total value and silver-lead-zinc and copper
each contributed seven to eight per cent, while tin accounted for
just a few per cent. At that time gold and silver were also
significant in several other areas of the Pacific Rim including
New Zealand, South America and parts of South East Asia
covering modern-day Malaysia and Indonesia.
There is insufficient time to trace the various changes wrought
by World War I, the Depression and then World War II.
Therefore, I shall focus on the changes that are within living
memory, say from the 1950s and 1960s onwards. This post-war
period saw a phase of renewed growth and development, with an
influx of newly trained people and the increased engagement of
geologists in mineral exploration. As the region recovered from
the effects of World War II, new discoveries began to expand the
minerals industry significantly, especially in Australia.
1. FAusIMM, Director, Surbiton Associates Pty Ltd, 71 Stevenson
Street, Kew Vic 3101. E-mail: info@surbiton.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
By this time, on the production side, silver-lead-zinc and
copper were the backbone of the Australian minerals industry,
with the majority of output coming from the stalwarts of Broken
Hill, Mount Isa and the West Coast of Tasmania. Coal was mined
for domestic use and some was also exported. Mineral sands
were a growing contributor to exports but gold was fading in
significance.
The 1950s saw the discovery and development of several small
uranium deposits, while Harry Evans recognition of bauxite at
Weipa in 1955, when he was carrying out exploration for oil and
gas, was the fore-runner for the development a significant
bauxite-alumina-aluminium industry.
Following the lifting of the iron ore export embargo in
December 1959, discoveries of large, high-grade haematite
deposits led to the development of one of the worlds great iron
ore provinces in the Pilbara region of Western Australia.
Manganese was discovered at Groote Eylandt in the early
1960s and subsequently developed by The Broken Hill
Proprietary Company Limited (BHP).
In 1965, Conzinc Riotinto of Australia Limited (CRA)
announced the discovery of Bougainville which sparked a search
for porphyry copper type deposits over much of the prospective
ground around the Pacific Rim. This subsequently led to the
discovery and development of a number of significant copper
and gold deposits.
The domestic coal industry was well established to provide
coal for power raising, town gas manufacture and for BHPs iron
and steel works. Although small volumes had long been
exported, the 1960s saw a rapid rise in exports. By the mid
1960s, this amounted to some eight million tonnes annually,
mostly from New South Wales, of which 95 per cent was
purchased by Japan. Regional exploration in Queensland also
began in earnest during the 1960s.
While the focus is on minerals, it would be remiss not to
mention the oil and gas industry a noteworthy event was the
discovery of oil at Rough Range in 1953. There followed a series
of oil and gas discoveries in the 1960s including Moonie in 1961,
Barrow in 1964, Moomba in 1966, and the Bass Strait field in
1965/1966.
In 1966, Western Mining Corporation Limited (WMC)
announced that several drillholes had intersected high-grade
nickel sulfide in the Kambalda area of Western Australia. The
subsequent nickel boom created much excitement, on the ground,
on the stock exchanges and even in the movies! Exploration
continued apace, while WMC developed mines around the
Kambalda dome and a few other small operations also
commenced.
Then came the First Oil Shock in 1973, when OPECs
response to the Arab Israeli war disrupted oil supplies and the oil
price rose rapidly. A period of economic recession and high
inflation followed.
The 1970s became the energy decade.
A major thermal and coking coal export industry was
developed in Australia and coal mines were also opened up in
Indonesia. Uranium was once again sought with major deposits
discovered in the Northern Territory, although development
proved somewhat controversial. The discovery of the North-West
Shelf oil and gas fields and the Phase I domestic gas
development dates from this period.
9
S E CLOSE
The First Oil Shock was reinforced by the Second Oil Shock at
the end of the 1970s, due to revolution in Iran, the subsequent
Iran/Iraq war and the Soviet invasion of Afghanistan. Between
1970 and 1980, the price of crude oil rose from around $US2 per
barrel to $US38 per barrel at its peak.
During the 1970s mineral development continued as well. In
Australia, further nickel operations commenced and there were
several base metal discoveries of which Roxby Downs was
outstanding. The 1970s also saw the discovery and development
of the diamond industry in the Kimberley region of Western
Australia and the continued expansion of the iron ore and the
bauxite-alumina-aluminium industries.
In summary, the 1960s and 1970s saw a swing from
conventional underground base metals mining to bulk
commodity, large-scale, predominantly open pit operations, both
for coal and a range of metallics and non-metallics. The
Australian industry had changed dramatically.
Then, as the 1980s approached there was yet another major
change as might be expected its impact was first felt on the
exploration front, which was in the doldrums. By this time gold
had almost faded from the scene but it was about to stage a
comeback, in Australia and also in several other countries around
the Pacific Rim.
I shall now focus more closely on the renaissance of the gold
sector and its impact on the industry as a whole, using Australia
as an example. Why single out gold? There are several good
reasons.
In Australia, gold production rose some 20-fold as a result of
the modern gold boom and gold has become one of the largest
single exports.
Gold as proportion of total mineral exploration expenditure in
Australia, rose significantly from around ten per cent in 1980 to
70 per cent in 1987 and thereafter has stood at 50 - 60 per cent.
Such a revival of gold exploration and development also
occurred elsewhere in the region.
By its nature the gold sector in Australia is mainly
characterised by a large number of small to medium sized
deposits. Thus, it provided expanded employment opportunities
for professionals such as geologists, mining engineers and
metallurgists, as well as support staff in many other professions
and service sectors.
The gold developments were different from the predominantly
large, capital intensive, bulk commodity developments that
characterised much of the 1960s and 1970s. Therefore, the
overall nature of the industry changed considerably yet again.
There is also another reason for examining the gold industry
more closely, although this may seem a long way removed from
the detailed geological case studies that will comprise this
session of the conference. You might recall that earlier I pointed
out that geologists, and indeed everyone, need to be aware of the
broader picture. Certainly, the expertise of the exploration
geologist in locating a viable economic deposit an orebody
and the expertise of the mine geologist in delineating the
orebody to enable it to be mined effectively and efficiently to its
full potential, are both essential in todays mineral industry. But
the simple bush boy or girl should never forget that the minerals
industry is comprised of many elements and requires a huge
range of expertise. This is well illustrated by the renaissance of
gold.
A complex set of circumstances led to the modern gold boom
in Australia. Two of the key factors were the fundamental change
in the gold price regime and the development of new processing
technology.
The price of gold was fixed until August 1971 at $US35 per
ounce ($A31.25) but thereafter it was subject to market forces.
Following the Second Oil Shock, gold captured considerable
attention when the price rose to a peak of $US850 per ounce
(~$A770) in January 1980.
10 Adelaide, SA, 19 - 22 September 2004
On the processing side, by the early 1980s carbon-in-pulp
technology was just beginning to emerge as a viable new
technology, although its origins can actually be traced back
around a century.
The investment climate and the availability of equity were also
important factors. Locally, the 1980s saw a new interest in gold
stocks, in a market which had experienced a dramatic nickel
boom and bust just ten to 15 years earlier and had then ridden the
energy roller-coaster. While the large swings in the gold price
caused some heartache in the early 1980s, by the mid 1980s the
market was hot for gold stocks and there were a host of small
producers as well as a swag of hopeful explorers. Several
hundred gold companies were listed on the Australian Stock
Exchange by the time of the stockmarket crash in October 1987.
The modern gold boom in Australia and the Pacific Rim both
benefited from and was responsible for a range of exploration
developments. Some of these were applicable in the wider
exploration field while others were specific to gold.
For example, the high cost of analysis for gold was one of the
principal problems in gold exploration. Although atomic
absorption spectrometry had been introduced in the 1960s it was
not sensitive enough for gold. It took until the mid 1980s to
develop analytical techniques for gold which were sufficiently
cheap, sensitive and reliable.
The gold boom saw the development of rapid, cost-effective
drilling with rotary air blast then reverse circulation drilling
utilised for exploration purposes. While this had begun towards
the end of the nickel boom, it really became cost effective for
gold.
The shift to large-scale open pits in the 1960s and 1970s has
already been mentioned. Many of the gold developments were
not necessarily large, rather they were characterised by the
development of relatively short life, steep wall pits. Of course, as
we have seen subsequently, there is now a swing back to
underground mining in the gold industry as ore is exploited
below the original open pits.
The gold boom saw a major change in the approach to plant
design and construction. Typically treatment plants were designed
and built individually to suit a specific deposit. However, as the
gold boom in Australia developed during the 1980s, a new
approach was pioneered by the local manufacturers. A generic
gold treatment plant could be built in just over six months, often
on a modular principle. The plants were not over-engineered and
they were designed so that they could be dismantled and
reassembled at a new location when the relatively short-life gold
deposits had been exhausted. Not only was the lead time short but
such plants were also relatively cheap compared with using
conventional engineering practices. The expertise developed was
readily transferable throughout the region.
It would be remiss not to mention the extraordinary impact of
computers have had on all aspects of the gold and the entire
minerals industry. Indeed, much of the minerals industry- related
software that is now recognised and exported worldwide,
originated from the local industry.
We have already briefly looked at equity aspects of financing.
On the debt side, it is hard to realise today how simple the local
financial markets once were. For example, trading banks in
Australia were only allowed to lend funds for a maximum period
of four years and exchange controls meant all foreign dealings
had to be approved by the Reserve Bank. It was not until the mid
1970s that a specific bank was set up to borrow long and lend
long to the rapidly expanding local minerals industry.
Furthermore, it was not until 1983 that exchange controls in
Australia were lifted, at the same time as the Australian dollar
was allowed to float freely.
Although the established overseas and local mining houses
were able to finance their gold developments by equity and/or
debt, many of the small, newly formed companies of little
PACRIM 2004

financial standing had difficulty procuring funding. Little wonder
then that Australia led the way in creating new forms of
financing to enable the newly emerging gold producers to
develop their deposits rapidly. Such financing was often tied in
with new forms of gold sales. Thus developed forward selling of
gold and subsequently a whole range of forward sales
mechanisms and option arrangements remember this occurred
initially at a time when inflation was high for example, interest
rates were 15 per cent and higher. Gold loans, whereby the loan
was repaid in physical gold, were developed and structured to
suit the needs of borrowers. There was however, little product
marketing as opposed to selling.
Tax incentives also played role in promoting the modern gold
boom. Companies which principally mined gold were exempt
from income tax but since this had been the case since 1924, it
alone was not a sufficient incentive, although it was far more
attractive once the gold price rose appreciably. In the 1980s, an
investment allowance of up to 40 per cent plus accelerated
depreciation allowances were very attractive. While technically
these were not available to a non-income tax paying gold
company, structuring made it possible for some companies to
take advantage of such favourable tax deductions.
Overall, the modern gold boom had some wide ranging effects
and industrial impacts that have spread throughout the rest of the
minerals industry. One of the major elements was the huge rise in
contracting, in mining, as well as the more traditional service
industries. The ready acceptance of contract mining is usually
attributed to operational aspects such as flexibility of plant and
equipment supply, broader experience and industrial relations
issues, as well as the substantially reduced capital requirement
for the mine owner. Also, although rarely acknowledged today,
tax incentives played a significant role, as the contractor could
utilise significant tax deductions and pass on some of the benefit
to mine owners.
Fly-in/fly-out operations were another outcome of the gold
boom. As it was uneconomic to develop fully serviced towns for
the relatively short life of the small to medium size deposits
which characterised the industry, the new system was developed
that has now spread to other sectors.
There were also huge changes at the corporate level and some
20 years after it commenced most of the early players in the
Australian gold boom have ceased to be. It is interesting to note
that the initial size of the company was of little significance in
determining its success as a gold miner.
In essence, the renaissance of the gold industry considerably
changed the overall profile of Australian minerals industry.
Now, to summarise.
Over the last century there has been extraordinary change in
the minerals industry in Australia and around the Pacific Rim
and there is no reason to doubt that it will continue to change.
The re-emergence of the gold sector alone has had a marked
impact on the industry overall and its people.
The challenge for the future is how to cope with such change. I
suggest that it is essential to accept that change is normal, to
embrace it and to plan accordingly. That is, take a strategic
approach and be proactive rather than reactive. This applies in
the industry, the corporate and the personal sense.
Conventionally, one of the early steps in a strategic appraisal is
a review of the relevant strengths, weaknesses, opportunities and
threats.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
THE TIMES THEY ARE (ALWAYS) A-CHANGIN
I shall leave the personal and corporate aspects for each
individual to ponder and turn to the minerals industry in the
region. How has it performed? What has been done well? What
needs improvement?
Certainly, the industry has shown considerable technical
expertise and innovation in a number of fields, be it exploration,
development, engineering, finance or environmental aspects, for
example. I have outlined some of these and there will be many
more recent, specific case studies and success stories discussed
in the next few days, I am sure. Equally, I hope some failures are
also examined these are rarely documented but often provide
very valuable lessons.
However, I would suggest that the industry has been less
successful in the management aspects, in the broadest sense.
Consider the following. The minerals industry is certainly an
international one but in Australia, given the countrys substantial
mineral endowment and relatively large minerals sector, does it
enjoy a commensurate position locally, say in terms of
investment and control? Why, despite the economic importance
of the industry in export terms alone, is it now so poorly
supported by the local financial markets, the government and the
community as a whole?
Those inside the industry cannot expect others to understand
and support it if they do not lead the way. They themselves need
to understand their industry better overall not just their
particular field. They need to appreciate the many factors that are
fundamental to the success of their industry, their organisation
and their own success, in the context of a changing industry and
a changing world.
In formulating your strategies and planning for change, I
suggest you also need an understanding of the past. In other
words, it is hard to know where you are going if you do not know
where you have come from and what has gone before.
I began this address with reference to Bob Dylans The Times
They Are A-Changin .
Those words are as relevant today as they were 40 years ago.
In fact, you might do well to track down a copy of song, read it
carefully and consider it in the context of this industry and your
career. For example:
For the politicians:
Come senators, congressmen
Please heed the call
Dont stand in the doorway
Dont block up the hall...
For those hesitant to embrace change:
Your old road is
Rapidly agin.
Please get out of the new one
If you cant lend your hand
For the times they are a-changin .
REFERENCES
Close, S E, 2002. The Great Gold Renaissance The Untold Story of the
Modern Australian Gold Boom 1982-2002 (Surbiton Associates Pty
Ltd: Melbourne).
Dylan, B, 1963. http://www.bobdylan.com
Raggatt, H G, 1968. Mountains of Ore (Lansdowne Press: Melbourne).
11
El Teniente and Rio Blanco Porphyry Cu-Mo Deposits, Chile
Giant Ore Formation in an Active Continental Margin
D R Cooke1, J Cannell, P H Frikken, P Hollings and J L Walshe

ABSTRACT

The central Chile Cu-Mo porphyry province contains several of the worlds largest mineralised
porphyry systems in terms of contained copper metal. The Andean continental margin of Central
Chile contains a 550 km long flat slab segment, around which a wealth of giant porphyry Cu-Mo,
Cu-Au and high sulfidation gold deposits have formed. A transient peak compressive event began in
the Late Miocene and continued into the Pliocene due to the subduction of the Juan Fernandez ridge.
This small collision has caused slab shallowing, crustal thickening, rapid uplift and exhumation,
producing the highest mountains outside of the Himalayas. Ridge subduction mostly shut down
volcanism during the late Miocene-Pliocene, producing the Chilean volcanic gap above the flat
subduction zone. At the same time, the giant porphyry Cu-Mo deposits of Central Chile formed on the
southern flexure from flat to normal subduction.
El Teniente is the largest porphyry deposit in Central Chile (11 845 Mt @ 0.63 per cent Cu,
0.02 per cent Mo and 0.035 g/t Au). It is a classic example of stockwork vein mineralisation
associated with a dacite porphyry complex, despite recent claims by Skewes et al (2002) that the
deposit is primarily a breccia-hosted system. El Teniente has been cut by a large, spectacular
late-stage diatreme (the Braden Pipe) which hosts Cu-Mo mineralisation in a marginal tourmaline
breccia facies. The deposit formed between 6 and 4.5 Ma, based on Re-Os and U-Pb geochronology.
By integrating paragenetic information with structural analysis of the stockwork veins, it is apparent
that the centre of hydrothermal activity at El Teniente was focused on the location now occupied by
the Braden Pipe. Episodes of intrusion, magma withdrawal and resurgence best explain the radial and
concentric vein arrays developed within the deposit.
The Rio Blanco-Los Bronces deposit is the second-largest of the Central Chile porphyry-related
hydrothermal systems (6900 Mt @ 0.75 per cent Cu, 0.018 per cent Mo and 0.035 g/t Au). While it
formed synchronous with El Teniente (6 - 4.5 Ma), it is an extreme contrast to the El Teniente system,
because mineralisation at Rio Blanco occurs primarily as cement within biotite and tourmaline
breccias. Late stage porphyritic intrusions cut the breccias, but are only weakly mineralised to barren.
The tourmaline breccia in the Sur-Sur sector of the deposit is zoned mineralogically from deep level
biotite to high level tourmaline breccia. Similarly, oxide minerals are zoned from magnetite at depth
to specularite at high levels. Sulfur isotope zonation progresses from near-zero values deep within the
breccia body to values around -5 per mil at the highest elevations, coincident with the zone of highest
grade mineralisation. These features, together with the radiogenic isotope systematics of gangue
minerals, imply that ore formation at Sur-Sur was driven by fluid mixing within the breccia column.

1. Centre for Ore Deposit Research, University of Tasmania, Private Bag 79, Hobart Tas 7001.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 13

Epithermal and Porphyry Gold Geological Models
G Corbett1
ABSTRACT
Geological models for the exploration and evaluation of epithermal Au
and porphyry Cu-Au deposits rely upon a classification of different
deposit styles and an understanding of the evolutionary processes of
deposit formation. These models guide the interpretation of subsurface
deposit anatomy in order to target drill testing and assist the
explorationist to focus on the more prospective projects. It is important
for the explorationist to distinguish between different styles of magmatic
arc Au deposits as each style displays distinctive characteristics with
exploration implications.
Magmatic arc Au deposits are distinguished as porphyry Cu-Au
formed at depths of 1 - 2 km at the apophysis to larger, buried, magmatic
sources which are also the ultimate metal sources for the varying styles of
shallower level epithermal Au deposits. Low sulfidation epithermal Au
deposits characteristically develop from dilute circulating meteoric waters
and are distinguished as the group with a closer relationship to intrusive
source rocks, and slightly higher sulfide contents (still generally <10 per
cent), which are classed with decreasing crustal level as: quartz-sulfide
Au + Cu, carbonate-base metal Au (including polymetallic Au-Ag), and
epithermal quartz Au-Ag, and contrast with the low sulfidation
adularia-sericite banded epithermal Au-Ag quartz veins. Sediment hosted
replacement Au deposits result from the reaction of a quartz-sulfide style
fluid with reactive carbonate host rocks. High sulfidation Au deposits
develop from strongly acidic evolved magmatically-derived fluids with
characteristic zoned alteration and pyrite-enargite mineralisation.
INTRODUCTION
Our understanding of epithermal Au deposits has advanced
considerably since the first Pacrim conference in 1987, and many
important contributions to our understanding have been
presented at subsequent Pacrim meetings. While we still draw on
the outstanding work on porphyry Cu-Au deposits prior to the
1980s new data enables the refinement of porphyry Cu-Au
mineralisation models. This study outlines some important
features of these deposits discernible to the field geologist and
demonstrates the benefit to explorationists of an understanding of
the styles of magmatic arc Cu-Au-Ag mineralisation, and the
processes of ore formation. It provides an update of the
Australian Institute of Geoscientists Presidents Lecture
Epithermal Gold for Explorationists (Corbett, 2002a), and
draws on earlier studies (Corbett and Leach, 1998; Corbett,
2002b), which are therefore not extensively quoted in this text.
Geological models characterise different styles of magmatic
arc Au mineralisation in the exploration environment, and so aid
the explorationist in either the more rapid identification of
economic mineralisation, or rejection of unsuitable projects. It is
important for explorationsits to be able to envisage the
subsurface anatomy of an exploration project, and comprehend
the exploration implications of different deposit styles. However,
the use of geological models in mineral exploration is a mixed
blessing. While models provide a framework for the expeditious
use of precious exploration resources, it is imperative to maintain
flexibility in the use of any model. As the database emerges,
explorationists should be ready to modify any existing
exploration model, or change to another which is more
appropriate. The science we use is constantly developing, and so
we must always be ready to accommodate geological scenarios
which do not fit our existing framework. Careful preservation of
the original factual data is essential in order to later apply
interpretive geological models in mineral exploration.
1. Consultant, 29 Carr Street, North Sydney NSW 1544.
E-mail: office@corbettgeology.com
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
STYLES OF MINERALISATION
Magmatic Arc Cu-Au-Ag deposits are distinguished mainly on
the basis of ore and gangue mineralogy in order to characterise
varying deposit styles which are derived from dramatically
different fluids, at varying crustal levels. Deposits are initially
divided between porphyry Cu-Au and skarns, developed at
crustal depths of 1 - 2 km, and epithermal Au-Ag + Cu which
form at shallower crustal levels, and are categorised as high and
low sulfidation styles (Figure 1). In the latter class it is possible
to further distinguish between the adularia-sericite style banded
epithermal Au-Ag quartz veins, and a group of more sulfide-rich
deposits with common associations with intrusive source rocks,
which vary with often continuous progressions from deeper to
shallower crustal levels as: quartz-sulfide Au + Cu,
carbonate-base metal Au, (including the Andean polymetallic
Au-Ag veins), and epithermal quartz Au-Ag deposits. Sediment
hosted Au (Carlin-style) deposits develop from the interaction of
quartz-sulfide style fluids with reactive calcareous rocks. The
class of intermediate sulfidation deposits defined recently
(Sillitoe and Hedenquist, 2004) in part corresponds to the earlier
documented carbonate-base metal Au association (Leach and
Corbett, 1993, 1994, 1995; Corbett and Leach, 1998; Corbett,
2002a). High sulfidation Au deposits develop from a
characteristic evolved strongly acidic magmatically-derived fluid
described below in detail.
LOW SULFIDATION EPITHERMAL GOLD
DEPOSITS
Low sulfidation epithermal Au-Ag + Cu deposits develop from
near neutral dilute fluids, which are dominated by meteoric
waters within cells of circulating hydrothermal fluids, commonly
driven by the intrusive source rocks for metals, at considerable
depth. Low sulfidation deposits therefore tend to dominate in
reactivated dilational structural settings, and so are commonly
characterised by banded veins comprising many individual
events of hydrothermal mineral deposition. Some events of
mineral deposition will be dominated by Au-bearing fluids
derived from the magmatic source, deep circulating meteoric
waters will entrain a magmatic component and so may exhibit
lower grade Au mineralisation, while shallow circulating
meteoric waters are sometimes barren. Groundwaters may
collapse into the hydrothermal system or otherwise interact with
the hydrothermal cells as an important feature of the ore
deposition process (Figure 2).
Varying mechanisms of mineral deposition are apparent within
multi-generational veins. While boiling or phase separation by
rapid pressure drop has long been proposed as a possible
mechanism of mineral deposition, detailed character sampling
has often failed to identify the bulk of Au-Ag mineralisation in
the minerals deposited at this stage adularia, bladed calcite,
quartz pseudomorphing calcite and to a certain extent
chalcedony. Rather, these minerals constitute much of the gangue
mineralogy. Some workers (Corbett and Leach, 1998) have
proposed that Au deposition may be promoted by rapid cooling
of the ore fluid, enhanced by wall rock reaction, or mixing with
varying ground waters. Rapid cooling of an ore fluid, which
promotes high-grade Au deposition, is often evidenced by the
presence of Au within chalcedony, while fluid mixing is apparent
from the presence of kaolin for low pH acid sulfate waters,
manganese oxide for bicarbonate waters, and hypogene
haematite and jarosite for oxygenated ground waters.
15
G CORBETT

FIG 1 - Conceptual model illustrating the styles of Pacific rim porphyry to epithermal Cu-Au mineralisation (modified from Corbett, 2002a).

FIG 2 - Conceptual model for the generation of fluids in high and low sulfidation epithermal Au deposits (modified from Corbett 2002a).

16 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


Varying styles of low sulfidation epithermal Au deposits,
which commonly form in different geological environments, are
distinguished on the basis of vein mineralogy, and display
sufficiently different characteristics to warrant distinction by
explorationists. The group of low sulfidation Au-Ag deposits
with higher sulfide contents, although in many instances only in
the order of one to two per cent, display a closer association with
intrusive source rocks. These display transitional relationships
and vary spatially and temporally from early to later in a vein
paragenetic sequence, and generally from deeper to shallower
levels from: quartz-sulfide Au + Cu, to carbonate-base metal Au,
and epithermal quartz Au-Ag deposits. Mechanism of mineral
deposition provides one of the important distinguishing
characteristics, and is of interest to the explorationist as a
significant influence upon precious metal grades.
Quartz-sulfide Au + Cu deposits
Quartz-sulfide Au + Cu deposits characteristically contain iron
sulfides with a quartz-rich gangue, varying to Kfeldspar-rich in
the silica-poor alkaline intrusion-related deposits, and
demonstrate pronounced mineral zonation with varying crustal
levels of formation. Barite is present in many Andean examples.
At deepest crustal levels these deposits may contain pyrite,
pyrrhotite and chalcopyrite, with lesser specular haematite and
magnetite, in a comb or druzy quartz gangue. Gold is deposited
with sulfides, typically by fluid cooling, and so slow cooled
coarse-grained veins often display lower Au grades, but good
metallurgy, especially where oxidised. While most quartz-sulfide
Au deposits contain pyrite as the main iron sulfide, at elevated
crustal settings this may pass to marcasite and arsenean pyrite, in
combination with opal to chalcedony as the silica component.
Many high level deposits are therefore characterised by an
arsenean pyrite bearing Au-As-Ag anomalous grey silica
(silica gris in Latin America), which commonly displays poor
metallurgy as Au is encapsulated in the sulfide lattice. The
refractory ores at the Ladolam deposit Lihir Island, Papua New
Guinea, are of this style. Here, Kfeldspar alteration dominates in
the alkaline host rocks.
As the quartz-sulfide Au + Cu deposits form at deeper crustal
levels they tend to exploit pre-existing structures and commonly
display a close relationship to porphyry Cu-Au intrusions, locally
representing the porphyry-epithermal transition. Many
quartz-sulfide Au + Cu vein systems fit into the D vein
classification of the early porphyry Cu literature (Gustafson and
Hunt, 1975). In dilational structural settings often evidenced by
the presence of sheeted veins, ore fluids may migrate
considerable distances from source porphyry Cu-Au intrusions to
form wall rock porphyry Au deposits. Consequently, sheeted
veins in some Maricunga Belt (Chile) porphyry Au deposits are
typical of quartz-sulfide Au + Cu mineralisation. Similarly, the
giant Cadia Hill wallrock porphyry, Australia, displays
significantly higher Au than Cu contents, than is typical of
porphyry Cu-Au deposits, and so represents a transition between
quartz-sulfide and porphyry mineralisation. At La Arena in Peru,
quartz-sulfide Au mineralisation occurs as a fracture coating
pyrite in a quartzite immediately overlying a porphyry intrusion.
Sector collapse at Ladolam initiated the change from porphyry
Au to epithermal Au mineral deposition (Corbett et al, 2001).
Many quartz-sulfide Au + Cu deposits occur as steeply dipping
lodes (Mineral Hill and Adelong, Australia; Bilimoia
[Irumafimpa], Papua New Guinea; Jaing Cha Ling, China;
Rawas, Indonesia) while fracture vein networks (Nolans,
Australia; deeper parts of Porgera, Papua New Guinea) may vary
to more dilational sheeted veins (Kelian, Indonesia). Flat
structures may host ore where collapse of volcanic edifices is
interpreted as a trigger for ore formation, such as the flatmakes at
Emperor Gold Mine, Fiji, initiated as a reactivation of bedding
planes during adjacent caldera collapse, or the listric faults at
Ladolam formed by Mt St Helens style sector collapse, and
where ores vary from a deeper structural, to higher level
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
EPITHERMAN AND PORPHYRY GOLD GEOLOGICAL MODELS
lithological control. At both these deposits most ore is hosted
within fine variably arsenean pyrite deposited as the ore fluid has
rapidly cooled by contact with the wall rocks. Quartz-sulfide Au
+ Cu deposits are not restricted to magmatic arcs and provide a
link to other intrusion-related mineralisation styles, such as the
Paulsens Precambrian quartz-pyrite lodes formed adjacent to a
granite batholith in the Ashburton district of Western Australia.
Exploration implications
Gold is readily liberated from oxidised coarse-grained
quartz-sulfide Au + Cu ores and so very low metal grades may be
worked as bulk low-grade heap leach operations (San Cristobal,
Chile). However, explorationists should be aware that
quartz-sulfide Au deposits are notorious for surficial supergene
enrichment, particularly in steeply dipping structures as sites of
chemical and mechanical concentration. Elevated assay results of
surficial samples should be treated with caution. Supergene
settings are evidenced by boxworks after pyrite, or Au anomalous
jarosite at the surface, base of oxidation and deep within faults.
Many cases where soil geochemical anomalies cannot be
substantiated during drill testing, are accounted for by supergene
enriched quartz-sulfide Au + Cu mineralisation, even in
association with other deposit styles such as adularia-sericite
banded epithermal Au-Ag quartz vein deposits. Fine grained
As-rich ores (Lihir and Kerimenge in Papua New Guinea), formed
by fluid quenching display poor metallurgy, and high As contents
in these ores may prove to be an environmental liability. However,
the silica gris may be used as a vector to buried ores in Andean
settings where extreme topographic variations allow access to
much deeper level, and possibly more prospective, vein portions
where higher precious metal grade polymetallic ores might occur.
Higher hypogene grade ores are recognised in settings of fluid
quenching, either by wall rock reaction, or by mixing with
varying ground waters as evidenced by kaolin (low pH waters),
or more commonly manganese oxide (bicarbonate waters), as a
reflection of the transition to higher crustal level carbonate-base
metal Au mineralisation. Overprinting epithermal quartz Au-Ag
mineralisation also provides higher Au-Ag grades.
Carbonate-base metal Au
As the group of sulfide-bearing low sulfidation intrusion-related
Au deposits display a fluid evolution from quartz-sulfide Au +
Cu, to carbonate-base metal Au, and later epithermal Au-Ag,
many deposits are characterised by all three deposit styles either
grading temporally in overprinting events (Porgera), laterally
(Kelian, Indonesia), or vertically (Kerimenge), while others
display telescoping into single lodes (Tuvatu, Fiji), or fracture
networks (Mt Kare, Papua New Guinea). Similarly, district scale
vertical zonation may be evident, such as in the Morobe
Goldfield in Papua New Guinea, where the Hamata
quartz-sulfide Au deposit occurs at deepest level, Hidden Valley,
Kerimenge and Upper Ridges carbonate base metal Au deposits
lie at intermediate levels, and the Edie Creek bonanza grade
epithermal Au deposit is mined at an elevated crustal setting.
Furthermore, Kerimenge displays a vertical zonation over several
hundred metres from quartz-sulfide Au + Cu, to carbonate-base
metal Au and highest and lateral epithermal Au-Ag
mineralisation, all localised at the contact of a fault with the
margin of a diatreme breccia pipe. Although most deposits
display some transitional relationships, a significant portion of
the Au in many southwest Pacific rim Au deposits is clearly of
the carbonate-base metal Au affiliation, as these represent the
most prolific producers in the region. In older more deeply
eroded terranes such as the Ordovician Lachlan Fold Belt of
eastern Australia, quartz-sulfide Au systems attain economic
status due to the overprinting carbonate-base metal Au
mineralisation (Kidston, Lake Cowal and London-Victoria in
Australia). Carbonate-base metal deposits display more efficient
17
G CORBETT
mechanisms of Au deposition than the quartz sulfide and so
commonly display higher precious metal grades.
Carbonate-base metal Au deposits are characterised by one to
ten per cent sulfides, commonly as pyrite > sphalerite > galena
with a gangue of carbonate and variable quartz and display
pronounced zonation (Corbett and Leach, 1998). At deeper levels
the transition to quartz-sulfide style may reflect minor pyrrhotite
(Porgera, Kelian). Vertical zonation in sphalerite type is evident
as a composition-controlled colour change related to temperature
(depth), varying from black, Fe>Zn, high temperature at depth,
through brown, red, yellow and locally clear, Zn>Fe, low
temperature sphalerite, at highest crustal levels. Carbonates are
zoned as the collapsing weakly acidic bicarbonate fluids undergo
a progressive rise in pH with depth by wall rock reaction, and so
vary with increasing depth from carbonates dominated by Fe
(siderite) at higher crustal levels, to Mn (rhodochrosite), Mg
(ankerite, dolomite) at intermediate levels, and Ca (calcite) at
deepest crustal levels. Much of the mineral deposition results
from the mixing of rising ore fluids with bicarbonate waters,
often derived from high level felsic intrusions.
The form of carbonate-base metal Au deposits varies
considerably from banded veins (Antamok and Acupan,
Philippines; Cikotok and Pongkor, Indonesia), or composite
lodes (Edie Creek and Woodlark Island, in Papua New Guinea;
Tavatu), fracture vein networks (Porgera, Mt Kare, Lake Cowal),
sheeted vein systems (Kelian, Kidston), and matrix to diatreme
breccias (Mt Leyshon, Australia; Montana Tunnels, USA, Rosa
Montana, Romania), or brecciated intrusion margins (Bulawan,
Philippines). This association with phreatomagmatic (diatreme)
breccias commonly provides a link to magmatic source rocks
(frequently dacite and rhyodacite). At higher crustal levels the
clay altered diatreme breccias are incompetent and so fracture
mineralisation occurs in the adjacent host rocks (Kelian,
Kerimenge), or at the diatreme-host rock contact (Acupan), and
only at deeper crustal levels does mineralisation reside in the
diatreme matrix (Mt Leyshon, Montana Tunnels), including
cross-cutting fracture systems (Cripple Creek, USA). Elsewhere,
pre-mineral phreatomagmatic breccia dykes exploit structures
and provide ground preparation (Woodlark Island).
While in many instances carbonate-base metal Au deposits
occur in the same terranes as high sulfidation deposits (Rio de
Medio at El Indio in Chile, Victoria at Lepanto in the
Philippines), it is also possible, but only rarely noted, for the
fluids responsible for the formation of high sulfidation Au-Ag
deposits to undergo progressive cooling and neutralisation by
rock reaction to evolve to low sulfidation mineralisation.
Consequently, the Link Zone Prospect provides a higher grade
and better metallurgy Au mineralisation as quartz-sulfide veins
with a carbonate overprint, recognised on the margin of the Wafi
high sulfidation deposit, Papua New Guinea.
Exploration implications
Carbonate-base metal Au deposits are important Au producers
but vary considerably, with resultant metallurgical implications,
especially where overprinting relationships are recognised with
other styles of low sulfidation Au mineralisation (Kelian,
Porgera). Explorationists should be aware that carbonate-base
metal Au deposits often display very irregular Au distribution
which must be taken into account in ore reserve estimation,
particularly within oxide zones, or where overprinted by
telescoped epithermal quartz Au-Ag mineralisation.
Carbonate-base metal Au deposits display highly variable forms
requiring an estimation of the sub surface three dimensions and
ore controls prior to drill testing. It is imperative that drilling
transect sheeted veins or lodes at the best possible angles, and
that kinematic analyses attempt to determine the setting of
structurally controlled ore shoots within throughgoing veins.
While high level diatreme breccia systems may be barren at
18 Adelaide, SA, 19 - 22 September 2004
surface, Au mineralisation may occur at breccia pipe margins at
this level, or within the breccia matrix at depth. The
characteristic Mn wad formed by the weathering of common Mn
carbonates provides a ready tool for the recognition of these
deposits, but may scavenge Au.
Epithermal quartz Au-Ag
Epithermal quartz Au-Ag deposits form at the highest crustal
levels and late stage in the paragenetic sequence of
intrusion-related low sulfidation Au deposits. They consequently
overprint both quartz-sulfide Au (Ladolam, Emperor), and
carbonate-base metal Au deposits (Porgera Zone VII, Mt Kare,
parts of Edie Creek). Some occur marginal to porphyry Cu-Au
deposits (Thames, New Zealand), or at the margins of
carbonate-base metal deposits (Kelian and Kerimenge). Most
display a strong structural control as they form at great distances
from the magma source as fluids deposit to cooler epithermal
crustal settings. Consequently, much of the higher grade ore
commonly occurs in ore shoots formed at preferential sites of
fluid flow or Au deposition (Porgera). One of the most notable
features of these deposits is that the characteristic bonanza Au
grades often overprint existing veins with little new gangue
mineral deposition, and so are difficult to identify. Many deposits
contain anomalous exotic metals such as tellurium (Emperor) or
selenium (Selene, Peru) and minerals such as tellurobismuthinite
are a common associate with Au (Bilimoia). Mineral deposition
is considered to be promoted by rapid fluid cooling, which may
be enhanced by the mixing of ore fluids with groundwaters
entering the deposits at elevated crustal settings.
Recent age data for the Porgera Au deposit (Ronacher et al,
1999) provides a 5.9 m.y. age for two contrasting mineralisation
events, with a very small age separation (0.26 m.y. within the
error estimation). The initial (Stage I) bulk low-grade
quartz-sulfide grading to a later carbonate-base metal Au event
currently being mined at the Waruwari open pit probably formed
at a deep crustal level as evidenced by the presence of high
temperature pyrrhotite and dark sphalerite. The later (Stage II)
low temperature epithermal quartz Au-Ag mineralisation well
developed in the Zone VII underground is associated with
renewed more felsic magmatism (Corbett et al, 1995), and
probably formed with at least 600 m less overburden. Recent
exposures from the deepest portion of the mine demonstrate that
the (Stage II) epithermal mineralisation grades rapidly through
quartz-sulfide and low temperature carbonate base metal
alteration, as a renewed mineralisation associated with the later
felsic magmatism. Thrusting exposed within the Porgera open pit
and common throughout the region, is suggested to account for
the rapid unroofing of the Porgera Intrusion Complex and
provision of a trigger for renewed epithermal mineralisation.
Exploration implications
The free milling bonanza Au has led to the ready identification
of many epithermal Au-Ag deposits by panning (Porgera, Edie
Creek), but may also promote the presence of artisan miners (Mt
Kare, Kelian). The improved metallurgy and higher metal grades
within epithermal quartz Au-Ag mineralisation have enhanced
the economics of otherwise lower Au grade or metallurgically
difficult quartz-sulfide Au and carbonate-base metal Au deposits
(Porgera, Emperor). Great care must be exercised in the drill
testing of these deposits to ensure that vein ores are intersected at
the best possible angles, and maintenance of good core
recoveries is essential within mineralised fault zones. Similarly,
in careful drill core sampling, a geologist should mark where
core should be sawn and sampled to ensure best possible assay
returns. The bonanza Au grades, commonly with only minor
gangue minerals, are difficult to recognise and so may provide
challenges in ore reserve determinations. Explorationists must
PACRIM 2004

also be aware that most ore and higher precious metal grades
commonly occurs in ore shoots formed as dilatant structural sites
(flexures or fault jogs) of enhanced fluid flow (Porgera Zone
VII), or enhanced metal deposition at cross structures (Thames),
or hanging wall splits (Porgera Zone VII).
Sediment hosted replacement Au
Sediment hosted replacement Au (Carlin style) deposits are well
documented as major Au producers in the western US (Carlin,
Goldstrike, Cortez), and although less important Au producers in
the southwest Pacific (Mesel, Indonesia; Bau, Malaysia; and
Sepon, Laos), remain as important exploration targets
worldwide, and are well represented in emerging provinces such
as China. They are also recognised in rocks of varying ages
including the Precambrian of the Asburton District of Western
Australia.
These deposits develop from the interaction of an ore fluid
typical of quartz-sulfide Au style deposits, with reactive host
rocks, typically impure calcareous sediments (marls of the
Popovich Formation, Nevada). Regional scale extensional
structures facilitate the transport of ore fluids from magmatic
source rocks at depth to elevated crustal settings, where mineral
deposition occurs. Deposits of the Carlin trend are localised
along the Post Fault system, and Mesel occurs within a smaller
scale fault jog, where dilatant fractures have focused ore fluid
flow. Sediment hosted replacement Au deposits display
important internal variations from structurally controlled feeder
structures at deeper levels, commonly with higher Au grades, to
lithologically controlled lower grade ores at higher crustal levels.
At Mesel Au contents decline rapidly moving away from dilatant
feeder structures, and in the Carlin trend structurally controlled
higher grade deposits such as Meikle are mined underground,
while the lithologically controlled ores (Carlin, Goldstrike)
represent large open pit mines.
Sediment hosted replacement Au deposits are characterised by
the dolomite-silica-kaolin alteration associated with the
introduction of auriferous arsenean pyrite, with anomalous Hg
and Sb. The early dolomitisation of calcite creates open space
and so provides secondary permeability for ore fluid flow
associated with local silicification. This dissolution is commonly
evidenced as collapse breccias. Silicification is also apparent as
barren jasperoid alteration, common in the upper portions of
these deposits.
Exploration implications
At the reconnaissance exploration stage, the resistant jasperoid
rocks, which are commonly preserved in the float train, are an
indication of this style of mineralisation within a region. While
barren in outcrop these rocks may provide indicators of
mineralisation at depth (Mesel). Because sediment hosted Au
deposits are commonly termed classic no see em gold deposits,
gold panning may not be reliable, and so advocated geochemical
tools include BLEG stream sediment sampling, followed by
analyses of soil samples for elements such as As, Sb, W, and Hg.
During evaluation, analysis of structural controls may allow
explorationists to target higher grade ores within feeder
structures at depth, which will compensate for the additional
costs of dealing with these metallurgically difficult fine As-rich
pyritic ores. Consequently, oxide ores are favoured for mining
operations. The environmental aspects of the As, Sb and Hg
bearing ores should be taken into account.
Adularia-sericite banded epithermal Au-Ag quartz
vein deposits
The adularia-sericite banded epithermal Au-Ag quartz vein
deposits are the most extensively documented low sulfidation
Au-Ag deposits, particularly using the parallels with the New
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
EPITHERMAN AND PORPHYRY GOLD GEOLOGICAL MODELS
Zealand geothermal systems. While many of these deposits are
well developed in back arc environments (Drummond Basin,
Australia; Taupo Volcanic Zone, New Zealand; Argentine
Patagonia; Japan; western US), or some are noted within
intra-arc rifts (Tolukuma, Papua New Guinea), other individual
deposits occur within magmatic arcs (El Peon, Chile; Ares,
Peru), or other linear magmatic arcs are dominated by these
deposits (Coromandel Peninsula, New Zealand; Kamchatka
Peninsula, Eastern Russia). This style of mineralisation is also
recognised in the mid oceanic ridge hotspot environment at
Iceland. All these environments display characteristic of bimodal
volcanism, commonly as an interpreted association of Au-Ag
mineralisation with felsic magmatism, commonly hosted with
andesitic or basaltic rock sequences.
The low sulfidation adularia-sericite banded epithermal Au-Ag
quartz veins typically comprise fine interlayers of chalcedony
varying to opal, with lesser adularia, quartz pseudomorphing
platy calcite, and black sulfidic ginguro bands (named by the
nineteenth century Japanese miners). Most genetic
interpretations suggest that meteoric dominant waters rise rapidly
up dilatant fracture systems hosted within competent rock
packages and boil to deposit much of the vein mineralogy.
Comparisons with geothermal systems link adularia and quartz
pseudomorphing platy calcite to boiling fluids. However, these
mineral assemblages tend not to contain Au-Ag mineralisation,
which dominates in the ginguro bands and to a lesser extent
chalcedony and so, as discussed above, some workers therefore
invoke rapid cooling, locally aided by mixing of the ore fluid
with varying groundwaters, as a mechanism for deposition of
bonanza Au-Ag mineralisation.
Adularia-sericite banded epithermal Au-Ag quartz vein
deposits display pronounced vertical zonation. At surficial levels
eruption breccias (Champagne Pool and Puhipuhi in New
Zealand; Toka Tindung, Indonesia; Twin Hills, Australia),
represent sites of venting hydrothermal fluids which form
laminated sinter deposits. Although anomalous in toxic elements
(Hg, As, Sb, W) many silica sinter deposits are barren with
respect to Au, unless proximal to fluid up flows (Champagne
Pool). Sheeted veins often extend into the deeper portions of
eruption breccias (Twin Hills; McLaughlin, USA) and also cap
the upper portions of some fissure vein systems (Golden Cross
and Karangahake in New Zealand). Most ore systems occur
within deeper fissure veins, commonly localised by dilatant
fractures within competent host rocks (Pajingo and Vera Nancy,
Australia; Tolukuma; Hishikari and Sado in Japan; Waihi, New
Zealand; Asacha; Eastern Russia). Wall rock clay alteration
varies from illite at deeper levels marginal to fissure veins and
grades vertically and laterally to assemblages dominated by
illite-smectite and thence smectite. Acid sulfate alteration
(alunite, cristobalite, kaolin) forms at near surficial settings by
the reaction with wall rocks of low pH condensate waters, which
may also collapse into the ore system to promote mineral
deposition.
Factors which influence the localisation of higher grade ore
systems include, structure, hot rock competency, and mechanism
of Au-Ag deposition. In many vein systems most ore, including
of higher grades, occurs in ore shoots developed as preferential
sites of mineralised fluid flow within flexures in otherwise
poorly mineralised throughgoing structures (Vera Nancy; Sims,
2000), fault jogs between structures, or splays. Structures
dominated by strike-slip fault movement host steeply plunging
ore shoots (Vera Nancy), while those within listric faults will
display flat plunges (Sierra Madre, Mexico; Arcata, Peru). Rock
competency influences the manner in which host rocks fracture
during vein formation and less competent sequences may cap
ore. At Hishikari, the Shimanto Group shale, which hosts vertical
quartz veins, is overlain by volcanic breccias, and preferred
mineral deposition of bonanza ores has occurred at the
intersection of these veins with the lithological contact, resulting
19
G CORBETT
in the development of flat plunging higher grade ore zones. At
Karangahake, the andesite-hosted mineralised fissure veins
become a less well mineralised stockwork in overlying
incompetent rhyolite. While the Chon Aike ignimbrite hosts
veins in the Cerro Vanguardia region, Argentine Patagonia,
elsewhere veins are limited to the competent margins of felsic
domes at Ares and Asacha (Kamchatka). Enhanced Au-Ag
deposition often occurs at sites of fluid mixing where ground
waters come in contact with ore fluids such as at splay faults
(Tolukuma), hanging wall splits (Asacha), or changes in host
rock competency (Hishikari). These may be recognised by the
presence of kaolin (Ares) or manganese oxide (Karangahake) in
the upper and better mineralised portions of the vein systems.
Exploration implications
Adularia-sericite banded epithermal Au-Ag quartz vein deposits
provide attractive exploration targets which commonly contain
bonanza Au-Ag grades, and are amenable to mining in semi
urban areas (Hishikari, Waihi) or in difficult terranes (Tolukuma
lacks a road link and is wholly supplied by helicopter).
Gold panning (Tolukuma; Chatree, Thailand), and BLEG
stream sampling, continue as useful first pass prospecting tools,
aided by the recognition of characteristic banded quartz as
stream float downstream (Tolukuma), or at prospect scale
(Chatree, Cerro Vanguadia). Resistive geophysical techniques
(CSAMT) have targeted quartz veins for drill testing (Hishikari),
locally within throughgoing structures (Vera Nancy). PIMA now
replaces slower XRD to readily provide studies of clay alteration
zonation to target higher temperature illite alteration close to
veins (see Golden Cross in Corbett and Leach, 1998).
Explorationists should be conscious of the importance of ore
shoots as sites of bonanza Au grades within vein systems. While
many ore shoots display steep plunges (Vera Nancy), flat
plunging ore shoots occur in settings of listric fault control, or
changes in host rock (Hishikari), and the intersection of hanging
wall splits with main the fault (Asacha). Often unmineralised
eruption breccias and acid sulfate alteration may cap mineralised
vein systems and so vector towards mineralisation.
Careful geological mapping is essential in order to plan drill
testing at the best possible angle, and the angle of veins to the
core axis should be monitored to ensure optimum drill direction
is maintained. As bonanza ores comprising ginguro sulfidic vein
portions locally occur with clays, and are commonly
fault-controlled, good drill core recoveries are essential for
accurate ore reserve determinations. Poor ore recoveries
sometimes downgrade Au contents.
BARREN HIGH ADVANCED ARGILLIC
ALTERATION
Many varying styles of alteration result from the interaction of
acid waters with host rocks, and while some alteration systems
clearly vector towards Au mineralisation, it is imperative for the
explorationist to distinguish prospective from barren
hydrothermal alteration. There is clearly great benefit for
explorationists to adequately understand varying styles of barren
advanced argillic alteration.
Magmatic-derived advanced argillic alteration
Magmatic arcs contain common bodies of barren advanced
argillic alteration which typically outcrop as resistive ledges
developed by alteration within structures or permeable
lithologies, and extending from porphyry to epithermal crustal
levels. While these alteration systems display characteristic
zonation outwards from silica cores, mineral assemblages vary
with depth such that at deeper levels mineral assemblages are
dominated by alunite, pyrophyllite and diaspore grading laterally
20 Adelaide, SA, 19 - 22 September 2004
to dickite and kaolin clays, and may contain minerals such as
corundum and andalusite, formed in very high temperature
settings. By contrast, at epithermal crustal levels, silica cores
grade to mineral assemblages dominated by alunite and more
marginal dickite and kaolin clays. Erosion of the softer marginal
clays commonly facilitates formation of prominent topographic
highs by preservation of the relict silica cores which are resistant
to erosion.
These alteration zones are interpreted to have developed from
the reaction with host rocks of magmatic volatiles venting
directly from intrusive source rocks early in their evolution, and
are compared to magma plumes in geothermal systems (Reyes et
al, 1993). Alteration zones are recognised outcropping on the
margins of porphyry Cu-Au systems where they have been
termed barren shoulders (Corbett and Leach, 1998). Importantly,
the fluids responsible for these alteration zones were rising at the
time of alteration, but have not evolved as the volatile-rich high
sulfidation fluids, and so are not mineralised.
Explorationist should note that these alteration zones are
common in the vicinity of porphyry Cu-Au deposits and other
intrusion-related manifestations such as low sulfidation
quartz-sulfide veins. Many exploration case histories include
instances where Au anomalism derived from nearby low
sulfidation epithermal veins has been attributed to these
topographically obvious alteration zones which are themselves
barren.
Exploration implications
The early recognition of these barren alteration zones may aid in
prioritisation of exploration programs to focus upon more fertile
alteration systems. These bodies of zoned magmatically derived
advanced argillic alteration might commonly be recognised by
the explorationist by the presence of massive rather than vughy
silica, typical of high sulfidation epithermal Au deposits. At
deeper crustal levels high temperature minerals such as
corundum and andalusite are characteristic, and minerals such as
alunite and diaspore may display very coarse-grained shapes
indicative of slow formation at near porphyry levels. The
distinction of this non-fertile alteration at epithermal crustal
levels is less definitive. Higher temperature dickite clays
dominate over kaolin. They commonly form in settings such as
within eroded volcanic edifices, and contain ledges of
structurally or lithologically controlled silica in intense
clay-pyrite alteration, which weathers to provide distinctive
liesegang rings and ferricrete deposits. Any Au mineralisation
may be in later cross-cutting veins or breccias and importantly
generally occurs outside to core of actual alteration zone.
HIGH SULFIDATION Au - Ag - Cu DEPOSITS
High sulfidation epithermal Au deposits result from the
interaction with host rocks of magmatically-derived ore fluids,
which have evolved to attain a characteristic very acidic
character. In simple terms, a volatile-rich fluid (dominantly SO2,
but also containing CO2, H2S, HCl) leaves the magmatic source
and becomes depressurised during the rapid migration to
epithermal crustal levels, causing the volatiles to come out of
solution and oxidise (as O2 and H2O also evolve from the same
depressurising fluid) to form a hot acidic fluid. The fluid has not
interacted with host rocks or groundwaters during rapid upward
migration. Consequently the fluid has evolved during rapid
ascent, from a near neutral fluid in the porphyry environment, to
strongly acidic at epithermal levels, where host rock reaction
results in the development of the characteristic high sulfidation
alteration zonation and mineralisation.
High sulfidation deposits commonly develop without the
repeated activation of dilational structures, which in low
sulfidation systems drive hydrothermal cells of
PACRIM 2004
 EPITHERMAN AND PORPHYRY GOLD GEOLOGICAL MODELS

meteoric-dominated waters to facilitate banded quartz vein in remote sensing imagery as colour anomalies, and more
formation. Rather, the development of high sulfidation deposits recently using multispectural clay analyses.
might be promoted as single magmatically dominated The enargite ores often display difficult metallurgy and so
hydrothermal events during transient relaxation in compressional many deposits are preferentially worked in the oxide zone
magmatic arcs. The kinematics of individual deposits therefore (Sipan, El Guanaco, La Coipa) and the bulk low-grade ores are
commonly contrast with observed regional tectonics. conducive to extraction as heap leach operations (Sipan,
The volatile portion of the high sulfidation fluid travels more Yanancocha). Acid mine waters produced by weathering of the
rapidly than the fluid-rich portion, and reacts with the host rocks pyritic clay alteration, and the high As content of sulfide ores,
to produce the characteristic alteration zonation. At the core of and local Hg by products, may all provide environmental
the alteration zone the host rocks undergo intense leaching by the concerns.
extremely acidic fluid to produce a rock composed almost One of the most important exploration procedures is the use of
entirely of silica, termed residual or vughy silica, indicative of careful PIMA studies to supplement field identification of the
the characteristic open space texture. As the hot acidic fluid is alteration zonation in order to vector to the central vughy silica
cooled and neutralised by rock reaction the zoned alteration which hosts most mineralisation. It is possible for silica-alunite
grades outwards through alteration assemblages characterised by to form resistant barren caps and so mask mineralised vughy
alunite, pyrophyllite, diaspore, and dickite/kaolin, to neutral silica. In some cases the barren alteration has previously been
clays such as illite/smectite, and eventual marginal porphyritic mined for industrial minerals (pyrophyllite at Peak Hill,
alteration (chlorite-carbonate). Alteration zonation also varies Australia). The Nansatsu high sulfidation deposits in Japan have
according to, proximity to the fluid up flow, host rock been mined for silica flux and the Gidginbung silica was initially
permeability, and crustal level (eg dickite at deeper levels passes used for road construction.
to kaolin in higher level cooler settings).
High sulfidation fluid flow is influenced by permeability PORPHYRY COPPER-GOLD
controls classed as structural, lithological and breccia. While
many deposits are localised either on major throughgoing Porphyry Cu-Au deposits develop as a result of focusing of the
structures (Gidginbung, Australia, on the Gilmore Suture; Wafi, mineralising fluids at depths of 1 - 2 km in the cooler apophyses
Papua New Guinea on a transfer structure; Mt Kasi, Fiji), or on to magmatic sources at greater depths, and so extend from
dilational fractures between throughgoing faults (Lepanto and intrusion host rocks into the wall rocks. Some of the better ore
Nena each lie on splays; El Indio occurs within a sigmoidal systems (Grasberg, Indonesia; Oyu Tolgoi, Mongolia; Ridgeway
loop), the contacts of these structures with permeable lithologies and Goonumbla in Australia) are characterised by multi phase
(Sipan, Peru; El Guanaco, Chile; Nena and Gidginbung), or intrusion emplacement into spine-like vertically attenuated
diatreme breccias (Lepanto), provide suitable settings for intrusion complexes. Overprinting intrusions provide multiple
development of alteration zones. Other deposits are wholly events of mineralisation and locally recycle ore minerals into
controlled within permeable lithologies in volcanic sequences settings with higher metal grades, but may also overprint and
(Pierina, Peru; La Coipa Chile). Many high sulfidation deposits obliterate mineralisation related to earlier porphyry Cu-Au
are associated with phreatomagmatic breccias (Wafi; Pascua, intrusions, therefore downgrading the total ore system (Figure 3).
Chile; Veladero, Argentina; Miwah, Indonesia), which no doubt
facilitate the rapid rise of ore fluids from porphyry to epithermal
levels, and so some high sulfidation deposits also occur within
felsic domes (Mt Kasi), or dome/breccia complexes (Yanacocha,
Peru). Felsic domes are an important link to magmatic source
rocks at depth, and therefore commonly associated with high
sulfidation Au deposits. Some deposits occur as veins (El Indio)
while other deeper level systems collapse upon porphyry Cu-Au
deposits (Monywa, Myanmar; Tampakan, Philippines).
A liquid-dominated fluid component enters the zoned
alteration via the same plumbing system as supplied the volatile
portion of the ore fluid, and so deposits sulfides comprising
pyrite, enargite (including the low temperature polymorph
luzonite) and additional alunite, along with barite and late stage
sulfur, mainly within the vughy silica, locally extending into the
adjacent silica-alunite portion of the zoned alteration. High
sulfidation deposits display a zonation from Cu-rich at deeper
levels, grading to Au-rich at higher crustal levels. Ores occur as
veins, breccia matrix, or filling vughy silica. While most
southwest Pacific deposits contain very little Ag, many Andean
deposits are Ag-rich and higher level deposits may contain Hg
and Te.

Exploration implications
The characteristic siliceous float provides a ready prospecting
tool, which may be recognised in the float train well downstream
and traced to source (Miwah). Covered silicification is
discernible on resistivity geophysical studies such as CSAMT
(Mt Kasi; Maragorik, Papua New Guinea), most apparent in
higher level systems where there is a strong contrast with the
adjacent conductive clay alteration. The high sulfidation clay
alteration may be identified on geophysical studies as regions of
magnetite destruction and coincident electrical conductivity, or FIG 3 - Some important features of porphyry Cu-Au anatomy.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 21

G CORBETT
In generally compressional subduction-related magmatic arcs,
changes in the nature of convergence may act triggers to
facilitate the emplacement of vertically attenuated intrusions
from magmatic source rocks at depth, into higher crustal level
dilational structural settings (Corbett and Leach, 1998; Corbett,
2002b). Ore fluids then continue to evolve from the magmatic
source into higher level ore settings utilising dilational fracture
systems such as sheeted quartz veins. Many quality porphyry
Cu-Au deposits are not linked to associated extrusive volcanic
rocks (Grasberg; Oyu Tolgoi; Bingham Canyon, USA)
suggesting volatiles and mineralisation may have been retained
within the cupola rather than vented.
Initial intrusion emplacement is characterised by zoned
prograde alteration grading outwards as potassic (magnetite,
secondary biotite and Kfeldspar), to inner propylitic (actinolite,
epidote), and outer propylitic (chlorite, carbonate) alteration
(Figure 3). In sodic rocks albite alteration may dominate over
potassic-propylitic alteration assemblages. The early disjointed,
high temperature ptygmatic A-style quartz veins (in the
terminology of Gustafson and Hunt, 1975) develop within the
cooling intrusion, while quartz-magnetite (M-style) quartz veins
dominate within the prograde magnetite-bearing alteration, and
locally contain chalcopyrite-bornite-pyrite mineralisation.
As also recognised in active geothermal systems (Reyes et al,
1993), volatiles venting from intrusions early in the cooling
history react with host rocks to produce barren advanced argillic
alteration, often migrating laterally as altered ledges, described
above.
Hydrothermal fluids accumulate at the apophyses of cooling
intrusions which become overpressured and eventually fracture.
The resulting pronounced pressure drop promotes the
development of B-style quartz veins (in the terminology of
Gustafson and Hunt, 1975). Whereas traditional geological
models utilise an excess of fluid pressure over rock tensile
strength to promote fracture development, structural systems
which localise intrusions may also crack an overpressured
carapace. Consequently, sheeted (parallel) quartz veins display
dilational tensile fracture/vein relationships, and so may
transport ore fluids from magmatic sources at depth, to the cooler
carapace, and extending into the wall rocks where mineral
deposition occurs. At Cadia Hill, Australia, wall rock hosted
sheeted quartz veins vary little through 700 m vertical extent.
Low pH condensate waters develop in the upper portion of the
hydrothermal system and react with the host rocks to promote
retrograde alteration of the prograde assemblages, including
demagnetisation, typically as phyllic (sericite-silica-pyrite)
grading to argillic (chlorite-kaolin-carbonate) alteration, as the
collapsing fluids are progressively cooled and neutralised by
rock reaction (Figure 3). Strongly acid fluids will also promote
the development of advanced argillic alteration, which contains
additional alunite-pyrophyllite in addition to the phyllic
alteration assemblage. This, and early venting magmatic
volatiles, contribute towards the development of lithocaps
recognised in the upper levels of porphyry systems. Cooling of
the intrusion apophysis promotes collapse of the hydrothermal
system and re-entry of ground waters into the porphyry
environment and development of extensive pyrite-bearing acid
alteration.
While sulfides may be exsolved from the major intrusion source
at depth throughout the life of the cooling porphyry, mineral
deposition is promoted as the apophysis cools in the at later stages
of the porphyry evolution, especially if ore fluids come in contact
with low pH condensate waters collapsing into the hydrothermal
system. Consequently, sulfides (chalcopyrite-bornite-pyrite)
commonly cross-cut or occur in the central portions of quartz
veins, and are focused within breccias. Dilational structural
settings promote the evolution of ore fluids from the intrusion
source at depth to form intrusion-related epithermal deposits
(including D veins) at higher crustal levels.
22 Adelaide, SA, 19 - 22 September 2004
Exploration implications
Explorationsits need to gain an understanding of the subsurface
anatomy of the overprinting intrusion-hydrothermal system to
target ore, generally best developed in stockwork and sheeted
quartz veins about the intrusion carapace and wall rocks. The
magnetic and zoned prograde alteration provides vectors towards
the intrusion carapace. However, magnetite destruction and
chargeability anomalies recognised in association with the
strongly pyritic retrograde phyllic-argillic and advanced argillic
alteration styles need not display a direct relationship with
mineralisation, and require an understanding of the overprinting
events. Regional structural analysis may be projected into the ore
environment combine with careful geological mapping in order
to plan the best direction for drill testing of preferentially
oriented sheeted vein systems, which both host and transport ore
as tension veins. While multiple intrusion events may upgrade
some systems, post-mineral intrusions can also stope out ore and
so explorationists should carefully consider the implications of
barren drill holes within mineralised zones.
Explorationists should not neglect the importance of skarns in
porphyry exploration. Many important porphyry systems (Frieda
River and Ok Tedi in Papua New Guinea) were identified by
follow up of skarn float well downstream, while the presence of
skarns helped maintain interest in the Cadia alteration system
during exploration, and eventual identification of the Ridgeway
deposit, and the Grasberg discovery in West Papua arose during
the mining of the Ertsberg skarn.
CONCLUSION
Geological models assist the in the prioritisation of exploration
projects and play an important role in the estimation of the
subsurface anatomy of hydrothermal systems in order plan
effective drill testing.
Porphyry Cu-Au mineralisation is best developed at intrusion
apophyses and high sulfidation Au mineralisation is focused in
the central portions of zoned alteration. Once categorised, barren
high zones of advanced argillic alteration can be avoided. An
understanding of the different styles of porphyry and epithermal
Au mineralisation allows the exploration implications of each
style of mineralisation to be applied to individual projects, and so
provide a framework for mineral exploration in magmatic arc
settings. For instance, quartz-sulfide Au deposits often undergo
surficial supergene enrichment and so surface assay data must be
treated with caution. The alteration zonation in high sulfidation
Au deposits varies according to crustal level, relationship to fluid
up flow and style of host rocks. Similarly, the application of the
understanding of the processes of ore formation assist in
accessing whether an alteration system warrants continued
evaluation, and locally aid in provision of vectors to guide
further exploration. In low sulfidation gold deposits shallow
circulating meteoric waters may deposit banded chalcedony
veins which are essentially unmineralised.
While many exploration successes formerly relied upon the
recognition of mineralised float during first pass prospecting
(Tolukuma, Chatree, Miwah), where ore systems have been
exposed by erosion, it is often now necessary to focus upon
techniques which aid in the identification of buried
mineralisation, either not exposed by erosion (Nena, Ridgeway),
or obscured by post-mineral cover (Raffertys Porphyry, Wafi;
Vera Nancy). It is hoped a better understanding of the nature of
porphyry and epithermal Au mineralisation will assist in the
evaluation of subsurface geology during exploration.
ACKNOWLEDGEMENTS
Too many colleagues and clients to mention over 25 years of
Pacific rim travel are thanked for their participations in the
PACRIM 2004

development of the concepts presented herein. Mike Smith and
Denese Oates kindly commented on the manuscript and Denese
also drafted the figures. I thank Dale Sims of The AusIMM for
asking me to present this paper, and the Australian Institute of
Geoscientists for permission to provide it as an update of the
AIG Presidents Lecture (Corbett, 2002a).
REFERENCES
Corbett, G J, 2002a. Epithermal gold for explorationists: AIG Presidents
Lecture, AIG On Line Journal April 2002, AIG website
www.aig.asn.au.
Corbett, G J, 2002b. Structural controls to porphyry Cu-Au and
epithermal Au-Ag deposits, Applied Structural Geology for Mineral
Exploration, Australian Institute of Geoscientists Bulletin, 36:32-35.
Corbett, G, Hunt, S, Cook, A, Tamaduk, P and Leach, T, 2001. Geology
of the Ladolam gold deposit, Lihir Island, from exposures in the
Minifie open pit in Proceedings Geology, Exploration and Mining
Conference, (Ed: G Hancock) pp 69-78 (The Australasian Institute of
Mining and Metallurgy: PNG Branch).
Corbett, G J and Leach, T M, 1998. Southwest Pacific rim gold-copper
systems: structure, alteration, and mineralisation, Society of
Economic Geologists Special Publication 6, 234 p.
Corbett, G J, Leach, T M, Stewart, R and Fulton, B, 1995. The Porgera
gold deposit: structure, alteration and mineralisation, in Proceedings
PACRIM 95, pp 151-156 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Gustafson, L B and Hunt, J P, 1975. The porphyry copper deposit at El
Salvador, Chile, Economic Geology, 70:857-912.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
EPITHERMAN AND PORPHYRY GOLD GEOLOGICAL MODELS
Leach, T M and Corbett, G J, 1993. Porphyry-related carbonate base
metal gold systems: The transition between the epithermal and
porphyry environments, in Second National Meeting, Specialist
Group in Economic Geology, Geological Society of Australia
Abstracts, Vol 34, pp 39-40.
Leach, T M and Corbett, G J, 1994. Porphyry-related carbonate-base
metal gold systems in the southwest Pacific: characteristics, in
Proceedings PNG Geology, Exploration and Mining Conference,
(Ed: R Rogerson) pp 84-91 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Leach, T M and Corbett, G J, 1995. Characteristics of low sulfidation
gold-copper systems in the southwest Pacific, in Proceedings
PACRIM 95, pp 327-332 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Reyes, A G, Giggenbach, W F, Saleros, J D M, Salonga, N D and
Vergara, M C, 1993. Petrology and geochemistry of Alto Peak, a
vapour-cored hydrothermal system, Lyete Province, Philippines, in
Geothermal Systems of the Philippines: Geothermics, (Eds:
D Sussman, J R Ruaya, A G Reyes and J W Hedenquist) 22:479-519.
Ronacher, E, Richards, J P, Villeneuve, M E and Johnston, M D, 2002.
Short life span of the ore forming system at the Porgera Gold Deposit
PNG: Laser 40Ar/39Ar dated from roscoelite, biotite and hornblende,
Mineralium Deposita, 37:75-86.
Sillitoe, R H and Hedenquist, J W, 2003. Linkages between
volcanotectonic settings, ore-fluid compositions, and epithermal
precious metal deposits in volcanic, geothermal, and ore-forming
fluids: rulers and witnesses of processes within the earth, Special
Publication No 10, Society of Economic Geologists, pp 315-345.
Sims, D, 2000. Controls on high-grade gold distribution at Vera Nancy
Mine: Northern Queensland Exploration and Mining 2000,
Australian Institute of Geoscientists Bulletin 31, pp 51-60.
23
Digging In and Digging Out Where Weve Been, Where We Are
and Where Were Going
P Cristofani1

ABSTRACT

The Australian minerals industry is carving out a leadership role on the sustainable development
imperative, not only in the global minerals industry but also in the Australian public and private
sectors. The industry has come a long way in recent years in grappling with its understanding of its
broader responsibilities and how it can address them.
Beyond providing the materials and wealth that have underpinned the evolution of modern
society, from the mills of England and Europe in the early days of the industrial revolution, through
the modern era and to the emerging post-industrial world of today, the minerals industry has been a
major contributor toward regional and national economic development for more than a century.
From the late 1960s societal standards and community expectations of the resource industries
have been rapidly changing. The evolution from environmental awareness to increasing expectations
of social performance under the Sustainable Development imperative has signalled a major turning
point in the way the minerals industry sees itself, its responsibilities to society and future generations,
as well as its shareholders.
The Australian minerals industry was one of the early movers in accepting and embracing
Sustainable Development during the 1990s. However, it has not been easy moving from acceptance
to a full, industry-wide understanding of how we can minimise our impacts and add maximum value.
Yet that is exactly what we are doing. As an industry we are always improving the management of our
operations in complex ecosystems, to not only minimise impact but to maximise environmental
benefits. And we are adding the most value directly to the community, particularly our neighbours
and those directly affected, through our work while being a price-taker in a highly competitive
global market.
It may be that the industry will always have its challenges cut-out for it in continuously improving
its performance in the key areas of environmental and social sustainability. By their very nature
continuous improvement and true sustainability have no end points. Nor are they destinations. They
are a perpetual journey and we must always be mindful of that. What the Minerals Council of
Australia is doing has profound implications. We are establishing a robust national platform for this
industry to operate in a manner that is attuned to community expectations and which seeks to
maximise the industrys contributions across all aspects of sustainable development, from
pre-feasibility to post closure stages, for as long as we all live. In so doing we are, indeed, digging. We
are digging out and sharing value for all people, now and those following.

1. Director Environmental and Social Policy, Minerals Council of Australia, PO Box 4497, Kingston ACT 2604. E-mail: p.cristofani@minerals.org.au

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 25

Meeting Quarter-Century Exploration Requirements in a
Quarterly Results Business World
M S Enders1 and R A Leveille2

ABSTRACT

At first glance it would seem ludicrous to propose a 25-year planning framework for exploration
goals in a business climate that is obsessed with quarterly results for the copper industry there is
little choice. The mean time from initial discovery to first production of copper for projects put into
production in the 1990s to early 2000s was 18 years with a standard deviation of 11 years.
Approximately 2000 recon exams, 350 first-pass drill tests, and five scoping studies were required to
produce one feasibility study at a typical cost of $134 million or $270 million per mine if only 50 per
cent go into production. Approximately 80 per cent of new reserve additions, however, have been
from near-mine discoveries within the same period. This all has occurred in an environment of
significant technological changes that have occurred at a pace of one every ten to 20 years over the
last 100 years. Greenfield copper exploration must be a long-range, consistently funded effort with
the expectation that production from new deposits is not likely to occur much sooner than seven years
and possibly as late as 29 years from discovery under different economic and technological
conditions than the current quarter. In the mean time, the burden falls on near mine exploration to
deliver the new reserves in the short-term that sustain and/or grow production between major
stand-alone discoveries.

1. Vice President, Worldwide Exploration, Newmont Mining Corporation, 1700 Lincoln Street, Denver CO 80203, USA.
E-mail: steve.enders@newmont.com
2. President, Phelps Dodge Exploration Corporation, One N Central Avenue, Phoenix AZ 85004, USA. E-mail: rleveille@phelpsdodge.com

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 27

Advances in Regolith Research A CRC LEME Perspective
R D Gee1 and R R Anand2
ABSTRACT
Recent research by CRC LEME from an Australian mineral exploration
perspective, has focussed on regolith architecture and mechanisms of
geochemical anomaly formation, within transported regolith. Methods
have been developed for the rapid production of regolith-landform maps
which is the starting point for understanding regolith architecture in the
third dimension. Methods for development of 3D regolith models are less
well advanced, and require integration of all available geological and
geophysical datasets, but still with a heavy reliance on drill-hole
information. The base of the transported regolith the unconformity
remains the most important regolith boundary for geochemical sampling
in transported regolith terrains. A new generation portable spectral logger,
employing a broader spectrum can interpret important features from drill
chips within the regolith. This allows internal architecture, redox fronts
and the all-important interface to be objectively identified. Transported
lateritic components can provide vectors to mineralisation in depositional
regimes. Palaeomagnetic and new radiometric methods which date
individual phases of the regolith provide evidence of different episodes,
each characterised by different geochemical dispersion patterns in
transported regolith. Along with lateritic residuum, calcrete is prime
sampling medium. Calcrete is probably enriched in gold through physical
and biological processes, rather than straight chemical processes. In
South Australia, the lowermost calcrete layer buried within the
depositional regime, especially where it lies directly over the silcrete
layer at the buried interface, is the desired sampling medium. Certain
gum trees, with deep roots in transported regolith have been shown to
record anomalous values of gold and basemetals in leaves, twigs and
bark. Anomalous metals in acid-sulfate soil seeps can also provide a
window through sedimentary cover in respect of mineralisation in the
basement. A number of Yilgarn occurrences of gold anomalies in
transported cover, strongly infer hydromorphic and biological
mobilisation of gold. This inference is supported by micro-analyses of
regolith mineral hosts in hardpans and mottled sediments, which show
gold and pathfinder elements prefer to reside in goethite and specific
types of clay minerals. This creates anomalies in specific mineral phases.
However the inference that that these are sites of mobile or labile metal
ions remains enigmatic. Similarly there is inconclusive evidence that
selective extractions will provide an anomaly not otherwise detected by
conventional total dissolution methods. Hydrogeochemistry is a
promising exploration tool for basemetal exploration. The importance of
microbes in both the dissolution and precipitation of gold in the
Australian regolith is demonstrated. We are on the verge of identifying
the functional gene sequence of individual active microbes. A
consideration of possible mechanisms for creating anomalies in soils
above transported regolith, presents difficulties in facilitating upward
movement of ions from the watertable to the surface. Perhaps the best
avenues of research lies with surrogate soil gases, and biotic transfers.
THE IMPORTANCE OF REGOLITH
Regolith is the surficial blanket of material including weathered
rock, sediments, soils and biota that forms by the natural
processes of weathering, erosion, transport and deposition. It has
complex architecture and may vary in thickness from a few
centimetres to hundreds of metres. It hosts or hides valuable
mineral deposits, we live on it, we grow our food in it, it is the
foundation of many major engineering works, and much of our
water supplies are stored in it. It underpins our economic, social
and infrastructure systems.
1. CEO of CRC for Landscape Environments and Mineral Exploration,
C/O- CSIRO Exploration and Mining, PO Box 1130, Bentley WA
6102.
2. FAusIMM, Leader of Mineral Exploration Program CRC LEME,
C/O- CSIRO Exploration and Mining, 26 Dick Perry Avenue,
Kensington WA 6152.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Prolonged deep weathering over the last ten to 250 million
years, on a predominantly stable continent of antiquity, has
created a unique Australian regolith. An understanding of
regolith architecture and the processes that act within it, are
essential to address the challenges of sustainable economic
development. Regolith science has important applications in the
fields of mineral exploration and natural resource management.
However the uniqueness of the Australian regolith means
research has to be done here, and cannot be borrowed from
anywhere else in the world.
EXPLORATION IN REGOLITH COVERED
TERRAINS
Despite some impressive advances in exploration technology in
the last two decades, geochemistry remains the prime direct
sensor in the armoury of explorers. This is because many
mineralisation styles have polar symmetry that is pipe or
sheet-like bodies of large vertical dimension. Inevitably they cut
the Earths surface and become disturbed in the regolith.
However most surface expressions of mineralisation in residual
regolith terrains have been identified and tested by explorers,
many of them with considerable success. There remains now the
subtle expression of near surface orebodies, and most
importantly, the undiscovered mineralisation under sedimentary
cover which presents a special challenge. Of even greater
challenge, from the geochemical viewpoint, is the non-polar
blind orebodies, that are encased in lithified rock.
Some of the exploration success of the last two decades, at least
in Australia, relates to the work of LEME1. The Cooperative
Research Centres (CRC) program is an Australian government
initiative which aims to bring together separate research groups
and industry to work on projects of national interest. The CRC for
Landscape Evolution and Mineral Exploration (commonly known
as LEME1) was set up in 1995 for a seven-year term. It produced
outcomes that are universally regarded as truly outstanding, and
their work is highly regarded in the exploration industry. Some of
the achievements of LEME1 are:
development of the new discipline of regolith geoscience;
development of practical techniques for the identification and
discrimination of geochemical anomalies in residual regolith
material;
development of regolith maps as an important dataset for
mineral exploration; and
an understanding of 3D geometry of metal distribution in the
residual environment particularly in regard to supergene
gold.
LEME1 research concentrated on regolith descriptions,
empirical geochemistry, and genetic models, in various
geological regions. Perhaps the lasting legacy of LEME1, which
incorporates earlier CSIRO work, is the lateritic-
depletion-supergene model as shown in Figure 1. These
achievements resulted in a paradigm shift in the use of regolith
knowledge in mineral exploration. The second stage renewal is
called Landscapes Environments and Mineral Exploration
(LEME2). In the field of mineral exploration, LEME2 research
has two priorities:
understanding the dynamics of processes that determine the
3D architecture of transported regolith; and
29
R D GEE and R R ANAND
FIG 1 - Idealised profile of the lateritic-depletion-supergene model
of gold distribution in typical Yilgarn regolith.
understanding the processes of metal mobility so as to make
geochemistry work through transported regolith.
This paper reviews some of the advances made by LEME in
these two strategic areas.
REGOLITH-LANDFORM DATASETS
Regolith mapping is now widely practiced in Australia, and is
increasingly being used as an aid to geochemical exploration.
The mapping concepts have been developed by Colin Pain et al
(Pain et al, 2001; Craig, 2001) of Geoscience Australia, and the
methodology developed by Steve Hill of Adelaide University and
Peter Buckley of New South Wales of Department of Mineral
Resources. Regolith features are an integrated expression of
geology, climate, landforms, geomorphic processes and
landscape evolution, and generally have close empirical
relationships with landforms. Thus, landforms become a
surrogate for regolith and are the basis of mapping procedures.
Regolith-landform mapping units are shown on regolith maps
as symbols depicted as alpha-numeric codes in the RTMAP
system. The first two letters are capitalised and indicate the code
for the regolith material type. The second two letters are lower
case and indicate the landform associated with the regolith.
Variation on the map symbol coding can be applied to thematic
maps, such as the Kurnalpi-Kalgoorlie Special Regolith
Landforms Map (Craig and Anand, 1993). Here the legend was
customised according to interpreted regolith models (relict,
erosional and depositional the RED scheme). Despite
conceptual limitations discussed by Anand and Paine (2002), the
RED scheme provides a practical guide for geochemical
sampling, and seems to work well for simple weathering cycles.
Guidelines for regolith-landform mapping are documented by
Pain et al (2001), Craig (2001) and Anand and Paine (2002).
Regolith-landform maps can be constructed by on-screen
digitising using a number of datasets such as air photographs,
digital terrain data, Landsat imagery, multi-channel radiometrics,
hyperspectra, radar and aeromagnetics. These products are
considered to be a proof-of-concept development, and it is now
up to other mapping agencies and users to generate these
products. In this respect it is important that they are produced in
a rapid and systematic manner, so they can be treated as one of a
number of spatial datasets within a GIS framework, of use to the
mineral explorer.
Understanding the regolith setting is essential in planning
geochemical surveys, and most importantly, in evaluating
previous geochemical exploration surveys an increasing
amount of which is being released into pre-competitive
30 Adelaide, SA, 19 - 22 September 2004
information systems in digital format compatible with other GIS
datasets. Regolith-landform maps are not intended to be end
products in themselves, but they go some way to providing the
two basic questions required by mineral explorers is the
near-surface material residual or transported, and how thick is
the transported material? Regolith-landform maps are the starting
point for understanding the third dimension.
THE THIRD DIMENSION
Existing models to date have been built around large drill-hole
datasets. The Yandal project, one of the flagships of LEME1,
included a 3D representation of residual and transported regolith
over a major gold system in an Archaean greenstone belt (Anand,
2003). The model, shown in Figure 2, was based on a drill-hole
database of over 70 000 holes, and high-resolution
aeromagnetics which spectacularly identified the maghemite-
defined palaeodrainages (Figure 3). These models which define
the base of the transported regolith, weathering front, and the
internal architecture, have genuine predictive value. Every
important greenstone belt should have one. But they are difficult
to construct even from good datasets for two reasons,
inconsistencies of logging, and the difficulty in actually defining
the interface.
A 3D model of the Harris Greenstone Belt has been
constructed from a limited number of drill-holes and the
magnetically modelled interface between mafic rocks and
overlying lake sediments (Sheard and Robertson, 2004).
The low-frequency, broad-bandwidth, time-domain TEMPEST
airborne EM system developed by Cooperative Research Centre
for Australian Mineral Exploration Technologies (CRC AMET)
has the promise to map regolith thickness and architecture, with
a minimal amount of drill-hole calibration. Integration and
modelling of semi-regional spatial datasets, built around
interpretations of the AEM TEMPEST system, has important
applications equally in mineral exploration and land
management, as shown by the systems approach of the
GILMORE project (Lawrie et al, 2000). In this project,
subsurface regolith framework was mapped using a range of
geophysical techniques hyperspectral, magnetic, gradiometery,
electromagnetic, ground penetrating radar, DEM laser scanning
and electrical soundings.
In 2000, trial Tempest surveys at 150 m line spacing were
conducted around Challenger and Tunkillia gold deposits in the
Central Gawler Gold Province. These non-outcropping areas are
characterised by sand dunes, calcrete, and complex patterns of
transported and in situ regolith. Lane and Worrall (2002)
successfully identified the regional long-wavelength conductors
as palaeochannels, which were as much as 180 m deep, with
basal sands saturated in saline water. However the CDI
conductive units in the shallower regolith areas related to the
saprolite-saprock interval, and therefore did not distinguish
between transported and in situ regolith.
Anand (2003) developed maps showing thickness of
transported cover for the Yandal belt, 30 per cent of which is
covered by more than 5 m of sediments. This map provided a
basis for the interpretation of regolith geochemistry. McQueen
(2004) is developing prototype derivative maps in the
Cobar-Girilambone area (Figure 4), showing residual areas,
transported material less that 5 m thick, and areas deeper than
5 m. These go-maps are created by fitting a form surface based
on the extrapolation of the residual areas under major areas of
transported regolith, guided by digital terrain and landform
models, together with high-resolution aeromagnetics and
drill-holes. Trials using nano-TEM (Davey et al, 2004) show
significant contrasts between low resistive palaeochannel
sediments and high resistive saprolite basement. This enables the
production of depth-resistivity profiles that relate closely to the
palaeochannel reconstructions from drill-hole sections. These
PACRIM 2004
 ADVANCES IN REGOLITH RESEARCH A CRC LEME PERSPECTIVE

Mt Joel

Transported overburden
Oxide zone
Transition zone
Fresh rock

RRAf025-04

FIG 2 - 3D model of regolith over a 100 km strike length of the Yandal greenstone belt showing transported and oxide zones, based on a
database of 70 000 drill-holes.

Sundowner
NN

Bronzewing

Mt McClure Palaeodrainage outlines

Yandal Belt outline
0 5 km

RRRAR
f0A2f70-2074-04

FIG 3 - Palaeodrainage lines interpreted from buried maghemite trails visible on first vertical derivative aeromagnetic imagery of the
Mt McClure-Bronzwing-Sundower area of the Yandal greenstone belt.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 31

R D GEE and R R ANAND

FIG 4 - Derivative depth-of-overburden map of the Cobar-Hermidale-Girilambone area showing areas of less than five and more than 5 m
thickness of transported overburden, derived from regolith-landform maps, geophysical interpretation, and inverse-distance-weighting of
drill-hole information; courtesy of Ken McQueen.

techniques require considerable more development before they
can be considered accurate and rapidly produced tools for the
mineral explorer.
A significant, but perhaps not surprising conclusion from these
excursions into transported regolith models, is that despite the
subdued landscapes across much of Australia, the
palaeo-topography below the interface is quite rugged, and
characterised by vigorous streams. It is this elusive form surface
that will continue to host the best direct geochemical indicators
of buried mineral systems.
OBJECTIVE AND AUTOMATIC REGOLITH
LOGGING
During exploratory drilling programs, especially for gold, the
challenge in regolith-covered terrains is to identify:
the interface between transported and in situ regolith,
the nature of the buried in situ regolith,
the parent rock from which regolith has formed,
potential depletion zones,
the presence of ferruginous (redox) zones,
alteration zones, and
calcareous layers in the transported material.
The Yilgarn regolith is complex, with some indications of
three-layered strato-regolith architecture (Anand and Paine,
2002), as shown in the generalised profile in Figure 5. But even
when the general regional model has been elucidated, manual
logging is slow, subjective, inconsistent, expensive, and fallible.
In consequence, only a small proportion of potentially available
information is obtained from many hundreds of kilometres of
drilling undertaken annually by mineral explorers.
Portable spectroscopy is promising to be a reliable and rapid
technique for identification of regolith and primary minerals.
LEME is participating in the development of a new technology
to make drill-chip logging more objective, efficient, and even
automatic. It builds on the prototype spectral core logger
(Hy-Logger), developed by CSIRO Mineral Mapping
Technologies Group. Hy-Logger can scan as much as 1000 m of
core per day, and the output is a mineralogical log coupled to a
high-resolution digital image of the core. Spectra can be
recorded at 1 cm intervals with mineralogy determined from
spectral absorption features in the visible to shortwave infrared
range (400 - 2300 nm).
In regard to regolith, it has long been known that the
ubiquitous clay mineral kaolinite, has an inferior level of
crystallinity in transported regolith than in the in situ regolith
(Pontual and Merry, 1996). However, the transported in situ
interface may be less well defined where transported regolith
directly overlies saprock or lower saprolite, because of low
kaolinite crystallinity in saprock, and possible interference by
smectite and muscovite/illite (Anand and Phang, unpublished
data). Figure 6 shows a cross-section (A) based on lithological
logging of a drill-hole section, compared to a grided image (B)
of kaolinite crystallinity index, portrayed as a simulated
cross-section. Lower crystallinity can be closely correlated with
the unconformity at the base of the transported regolith.
Similarly the water absorption peak of clays can also identify the
transported in situ interface. Thus if Rotary Air Blast (RAB)
chips are zapped with a PIMA instrument, the important
interface can be identified. PIMA records spectra in the limited
range of 1300 - 2500 nm, and cannot always detect important
mineral phases needed for understanding regolith framework.

32 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 ADVANCES IN REGOLITH RESEARCH A CRC LEME PERSPECTIVE

LEME has acquired a new spectral analytical device called the

Red-brown ASD FieldSpecPro Spectro-radiometer (Figure 7). This has a
Quaternary
sandy clay much wider spectral window covering the 350 - 2500 nm range,
and is sensitive to clays, hydrous silicates, Fe oxides (eg
Hardpanised goethite) and carbonates. Under certain environments these
sandy silty mineral phases are the preferred hosts for low-level metal
and gravelly signatures that disperse through the regolith from primary buried
colluvium and
gold deposits. Goethite also defines redox fronts, which are
alluvium
prime sampling intervals. The technique does not recognise
Goethitisation feldspars, quartz, sulfides and other non-hydrous minerals
Pisolitic (hematite present in many rocks.
ferrricrete goethite) The ASD device is portable, and can be taken to the drill site,
or the chip store. Depending on the feeding mechanism, it can
read 600 - 700 samples per day. The current task is to write
Transported Overburden

Hematitisation
special software to convert spectral absorption data into
(dissolution of
semi-quantitative mineralogical interpretations. This information
Tertiary

Fe around tree
Mega mottled roots and can then be imported into standard exploration data bases, along
palaeochannel precipitation of with other geochemical and drill-hole data where it can be
clays hematite) plotted directly onto cross-sections. This is another theme in the
Transported rapid and automatic generation of 3D models.
gravel
LATERITIC RESIDUUM AND LAG SAMPLING
The ferruginous upper part of a weathering profile may be
residuum, ferruginised sediment (ferricrete), ferruginised
saprolite, or a mixture of these, to name some of the most
common situations (Figure 5). Focus has been where the
ferruginous upper part is a residuum. Lateritic residuum
Clays commonly evolves by partial collapse, involving some lateral
movement, over say 10 - 50 m, following chemical wasting, as
well as introduction and mixing of exotic material. It may be
exposed or buried at depth.
The use of lateritic residuum geochemistry for mineral
exploration in Western Australia was first recognised by
Base of sediments
Goethitisation Mazzuchelli and James (1966), and perfected in an operational
Lateritic sense by Smith and Perdrix (1983). The outstanding success of
residuum this method as applied to lateritic residuum is documented by
(some mixing Anand (2001) and Cornelius and other (2001). Residual
with overlying accumulation, combined with hydromorphic and mechanical
material) dispersion, results in widespread dispersion haloes related to
Regolith on Archaean rock

concealed ore deposits. It provides an easily-collected sampling

Mottled Hematitisation medium with the ability to scavenge and concentrate gold and
saprolite pathfinder elements even where adjacent parts of the profile are
depleted or transported. The haloes tend to be large, typically
100 to 400 times larger in area than the deposits from which they
have been derived (Smith et al, 2000). Sampling of lateritic
residuum can, thus, be useful for reconnaissance surveys through
to early stages of prospect evaluation.
Saprolite The recognition of buried lateritic residuum provides a useful
sampling medium during exploration and was important in the
discovery of Bronzewing (Wright et al, 1999). Studies around Mt
McClure in the early 1990s (Anand et al, 1993) recognised
buried lateritic residuum at some depth, and extrapolated the
model further east in the depositional regime that is now
Bronzewing (Anand, 2003). Buried residuum is now known from
many other parts of the Yilgarn Craton, and criteria have been
established to distinguish this from transported gravel and
ferricrete (Anand, 2001). Stratigraphic drilling is necessary to
establish whether the overburden overlies a buried lateritic
Saprock residuum or an erosion surface cut into saprolite.
Cornelius (2004) is now looking at extending the application of
Bedrock ferruginous materials into transported and depositional terrains.
RRAf015-04
Several case studies in the Yilgarn Craton have shown the
continuity of multi-element mineralisation signatures of
transported ferruginous gravels derived by degradation and
FIG 5 - Generalised regolith profile for the Yilgarn, showing a dispersion from residual duricrusts. For example, at Jaguar Cu-Zn
three-layered strato-regolith architecture, each with different VMS deposit, lateritic gravels have signatures characteristic of
degrees of ferruginisation, probably relating to episodic both the Teutonic Cu-Zn deposit four kilometres up-drainage, and
weathering events. the underlying Jaguar deposit (Cornelius, 2004).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 33

R D GEE and R R ANAND

FIG 6 - Cross-section (A) based on lithological logging of drill-holes, compared to a grided image, (B) of kaolinite crystallinity index, portrayed
as a simulated cross-section. Lower crystallinity can be closely correlated with the unconformity at the base of the transported regolith.

The interpretation of lateritic geochemical datasets using
multi-variate discriminate analysis, in the context of current
topography, palaeo-topography and dispersion directions can
provide vectors to mineralisation under transported terrains.
CSIRO/LEME and GSWA have commenced a program of
regional lateritic residuum/lag sampling over the entire Yilgarn
Craton, aiming for a 9 km triangular sampling pattern. The
considerable number of samples already in the residual databases
will be included in the new database. With the benefit of digital
terrain and regolith models, the successful techniques of lateritic
residuum/lag sampling can be of significant benefit for surface
exploration. The chemical changes that residual ferruginous
nodules undergo during redistribution into lags is not fully
understood, but in the Yilgarn the original signatures seem to still
be recognised using multivariate analysis, over distances of tens
of kilometres.
McQueen (2004) has examined the partitioning of ore elements
and pathfinder elements in residual and transported lags related to
Cu-Au systems in the Cobar Mineral Field, In the in situ regolith,
goethite is an important host for Zn, Cu, As and to a lesser extent
Pb, Bi and Sb. Hematite is the predominant host for Cu. In
goethite-dominant near-surface caps there is a strong correlation of
As and Zn with Fe, whereas in hematite-dominant phases there is
a strong correlation of Cu with Fe. However in the redistributed
lags there is a progressive conversion of goethite-hematite phases
to hematite-maghemite phases. At the same time there is an
increasing correlation of Pb, As, Sb, Bi, Ba, Cr and Th with Fe, as
these pathfinder elements increase in the mature lags. Significantly
Zn, Cu and Au do not increase in the mature lags. Thus the actual
ore elements fail to report, and the relatively immobile Pb once
fixed into hematite can become widely dispersed by mechanical
methods. These results indicate that pathfinder elements should be
normalised against Fe for their correct interpretation.

34 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


RRAf020-04
ASD Field Spec Pro Spectroradiometer
FIG 7 - Portable ASD spectro-radiometer producing automatic
mineralogical analysis from rapidly read VSWIR spectra.
INTERFACE SAMPLING
A feature of all reconstructions of the palaeo-surfaces below
transported regolith is its considerable relief and ruggedness. It is
therefore not surprising that the interface between the basement
collectively including saprock, saprolite and lateritic residuum
and the first overlying transported sediments, remains the
prime sampling surface, of equal ranking with lateritic residuum
and calcrete on the surface. The problem of course is that it may
be buried below tens of metres of cover. The saprolite-
transported interface may incorporate residual or partly
transported palaeosols, with greater scope for mechanical and
hydromorphic dispersion, during and after sedimentation. The
material on the interface can include diverse materials from a
variety of sources, depending on the degree of
palaeo-exhumation of the palaeo-weathering profile. As well as
containing metals in gossan particles, free particles (placer gold),
and anomalous fossil lags, the ferruginised material may act as
sorption traps for hydromorphic gold and other metals.
Experience has shown that, although the interface anomalies are
of low contrast, their sizes are significantly greater than those in
the top of the basement. Butt et al (2000), Gray et al, 2001,
Anand (2001), Robertson (2001), Robertson et al (2001) and
Anand (2003) summarises numerous examples of striking gold
anomalies in the immediate basal sediment in many areas (eg
Calista, Golden Delicious, Mt Joel, Bronzewing, Lawlers, Stellar,
Quasar ( Mt Magnet ) and Harmony (Peak Hill) gold deposits).
This is illustrated on a 3D image of the 50 ppb gold enveloping
surface for the Mt Joel gold deposit (Figure 8). This shows a
spread of elevated gold values on the interface, at the base of the
first transported layer. In this case, the gold anomalies are at the
base of the transported cover (itself composed of lateritic waste
material) at the unconformity with the truncated saprolith, at
depths of 40 - 60 m.
Modern micro-analytical techniques now allow the diagnosis
of carefully selected individual grains from the interface. This
can reveal not only the proximity but the style of mineralisation.
For example Smith (2004) can recognise the textural and
compositional characteristics of the Golden Grove Gossan Hill
Cu-Zn-Au VMS deposit in clasts, lateritic nodules and other
detrital grains in the basal conglomerate.
This reinforces the importance of precisely defining the
palaeo-surface, and more importantly the loci of the buried
thalwegs of the earliest palaeodrainages.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
ADVANCES IN REGOLITH RESEARCH A CRC LEME PERSPECTIVE
Concentration
Barren and dispersion
transported cover at interface
RL
(m) 470
450
430
410
Alluvium
390 Oxide
200 Transition
100 370 Rock
0
Northing -100
(m) -200
9200 9300
9000 9100
RRAf026-04 Easting (m)
FIG 8 - 3D image of gold distribution above the 50 ppb level for the
Mt Joel gold deposit, showing a spread of elevated gold values at
the interface immediately below the transported regolith; courtesy
of David Gray, CSIRO Exploration and Mining.
DATING EVENTS
Geological evidence indicated that most of the Australian
continent has experienced extensive subaerial weathering over
long periods with evidence of superimposition of effects. From a
mineral exploration viewpoint, it is important to know if the
weathering processes, and hence anomaly formation, is an
expression of continuous or episodic weathering. The LEME
geochronology project involves both technique development and
dating of events.
The basic geochronological framework of the Australian
regolith is provided by palaeomagnetic dating (Pillans, 2004).
This is done (Figure 9) by plotting the pole of the chemical
remnant magnetism (CRM) of secondary mineral phases on the
trace of palaeomagnetic poles of known age commonly known
as the Australian apparent polar wander path (Schmidt and Clark,
2000). Most useful is the phase change from goethite which
has no CRM to hematite which does. Pillans (2004) records
palaeomagnetic determinations from some 30 sites (mostly mine
sites) throughout Australia, with a focus on the Yilgarn Craton.
There is a distinct bi-modality appearing in this data, pointing to
events around 10 Ma and 50 - 60 Ma, with some evidence of
Mesozoic or even earlier weathering, as shown in Figure 8. At
around 60 Ma, the climate was wet, cool temperate and the
vegetation was dominated by conifer forests and woodlands. The
second cluster at late Tertiary (10 Ma) represents climates that
were seasonally drier (though rainfall was still higher than
present) and warmer, with consequent flora changes, as southern
Australia drifted to lower latitudes. Thus, the two major episodes
of hematite formation in the Tertiary occurred under differing
bioclimatic regimes.
This spread of palaeomagnetic ages is revealed at Lancefield
gold pit (WA), where saprolite (Archaean basement) is overlain
by Permian fluvio-glacials, Tertiary palaeochannel clays, and
Quaternary alluvium (Anand et al, 2004). The Permian and
Tertiary sediments each show different suites of hydromorphic
metal dispersion, particularly Au, As, Cu and Zn. It is likely that
each suite represents a different weathering and hydromorphic
event. If this is so, it emphasises the importance of dating the
regolithic events so that the more favourable sampling media
either bulk sample or mineral phase can be identified.
More precise isotopic techniques are being developed in
LEME, notwithstanding the problems of finding suitable
minerals, and the uncertainties regarding closed systems.
35
R D GEE and R R ANAND

150 E
120 E
180
E
90 E
150 Ma

250 Ma

LAN
S
30
DBS1
60
E
60 Ma DBS2 E
BRZ 180 Ma
M 0
NPK MP 21
MTP NCB

MCK
K

360 Ma
PB
6 0 S

LAN
AN

RRAf023-04

FIG 9 - Australian Apparent Polar Wander Curve from 360 Ma to present (Schmidt and Clark, 2000) with some representative
palaeomagnetic poles (95 per cent confidence circles) from weathered regoliths from some Australian mineral deposits (Pillans, 2004).
PB = Perth Basin (Schmidt and Embleton, 1976), MCK = MacKinnons (McQueen et al, 2002), MP = Morney Profile (Idnurm and Senior,
1978), BRZ = Bronzwing (Anand and Paine, 2002), DBS = Bead bullock soak (Pillans and Idnurm, unpublished), MTP = Mt Percy
(Pillans and Bateman, 2000), NCB = New Cobar (McQueen et al, 2002), LAN = Lancefield (Pillans, unpublished) NPK = North Parkes
(Pillans et al, 1999).

Pidgeon et al (in press) are developing the (U+Th)/4He method
on lateritic duricrust from the Darling Range, and have produced
a date around 10 Ma, that is compatible with the youngest age of
lateritic duricrust in the Yilgarn, recorded by palaeomagnetism.
Other developments are U-Pb and U-series of goethite by Alex
Nemchin (Curtin University), SHRIMP dating of opal by Susan
Symons (PhD student, Curtin University), and U-Pb dating of
anatase in silcrete by Martin Smith (PhD student, ANU). The
challenge with these novel applications is the low levels of
uranium in residual material.
Documenting the timing of regolith-forming events goes
further that good-to-know science. In Australia there is debate
whether the lateritisation process is episodic or continuous
(Taylor and Shirtliff, 2003) on the continental scale. The answer
to this has a bearing on modelling the distribution of gold in
deeply weathered terrains. The work in the Yilgarn Craton
(Anand and Paine, 2002) indicates the presence of possibly three
extensive strato-regolithic units (Figure 5) the lateritic duricrust
(possibly 60 Ma), an intermediate ferruginous transported unit of
red clay and grits, and the Wiluna Hardpan (possibly 0.76 Ma).
Each of these strato-regolith units has their own regolith gold
signatures, which variously can record proximal, distal or
spurious geochemical signatures.
CALCRETE (PEDOGENIC CARBONATES)
Calcrete geochemistry, along with lateritic residuum sampling, is
one of the most effective techniques for gold exploration in
deeply weathered terrains. A close correlation between calcium
and gold was first demonstrated in 1989 from calcareous soils in
a residual environment, over the Bounty gold deposit in the
Yilgarn Craton by Lintern (1999). No other ore-related elements
show this association; indeed most are diluted by the carbonates.
The presence of calcrete in both residual and transported
regolith is a common feature in many semi-arid regions, whether
these are juvenile terrains having soils developed from fresh
rocks, or ancient landscapes with deep regolith. Calcrete forms
by the cementation, displacive or replacive introduction of
predominantly calcium carbonate, into the vadose zone (above
the watertable), where it may occur as friable powders, nodules
or massive layer-like accumulations. It forms by precipitation
from near-saturated calcium carbonate solutions by processes
that may include evaporation, evapotranspiration, degassing of
CO2, or bacterial consumption of CO2. Gold content may be due
to trapped detrital particles inherited from the host material, or it
may be due to the stabilisation of hydromorphic gold, during or
after the formation of calcrete. The mechanism is still uncertain.
Calcrete is the favoured medium for gold in South Australia, to
the same extent as lateritic residuum in Western Australia.
Whereas in the Yilgarn craton, calcrete sampling is generally
used for in early prospect identification, in the Gawler Craton it
is employed at all stages of exploration, including target
definition. Calcrete geochemistry is credited with the discovery
of Challenger and Barnes deposits in the Gawler Craton.
However it is a fallible technique. There are examples of gold not
reporting in calcrete above known mineralisation, and examples
of large dispersed low-level anomalies that do not obviously pin
point gold in the basement. To better understand the process of
anomaly formation in calcrete, LEME is researching regolith
relationships, micro-morphology of gold, and role of biota.
Lintern (2004) studied the regolith profiles and 3D
distributions of gold at 19 mineralised sites in South Australia.
These sites had different degrees of preservation of the residual
profile, and different thicknesses of transported regolith. The
model profile consists of saprolite capped by residual silcrete,
commonly overlain by transported alluvial, colluvial and aeolian
material. Calcrete has intergrown within these upper materials,
generally within the top 5 m. But the most important is that layer
developed just above the interface silcrete, which forms a
permeability barrier. Major conclusions are:

36 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 ADVANCES IN REGOLITH RESEARCH A CRC LEME PERSPECTIVE

anomalous gold in calcrete registers well with underlying
gold mineralisation where calcrete is formed immediately
above residual silcrete, irrespective of the extent of current
exhumation or burial;
where thickness of transported regolith is greater than 5 m,
basement gold signatures are totally suppressed in
near-surface calcrete;
highly anomalous concentrations of gold (hundreds of ppb)
may occur in calcrete at the interface; and
in erosional and depositional regimes, gold may be dispersed
laterally in both degraded silcrete and younger calcrete, thus
providing false anomalies.
Current studies are underway to investigate the
micro-morphology of non-detrital gold particles using x-ray
photon microscopy, secondary ion mass spectrometery, and
electron beam technologies. This should help to provide evidence
on precipitation mechanisms. Experience with partial leach
suggests that calcrete has a high proportion of weakly bound,
easily soluble gold, indicative of a hydromorphic origin. For
example, Gray et al (1990) note significant dissolution in iodide
and deionised water, indicating a chemical origin. Chemical gold
may accumulate in calcrete because like Ca, it behaves like an
evaporite. Microbes may play dual roles, firstly by providing
ligands to dissolve and mobilise gold, and secondly to provide
nucleation sites on their surfaces for gold precipitation.
BIOGEOCHEMISTRY
Research under the direction of Steve Hill (Adelaide University)
and Ravi Anand is revisiting the vexed question of
biogeochemical prospecting. Hulme (2003) at Adelaide
University is focussing on river red gums (E camaldulensis)
which present an ideal sampling medium because of their
widespread occurrence in transported regolith in arid
environments, their confinement to watercourses, and their
extensive tap roots. Orientation sites have been set up in the
Curnamona Craton, on the basis of proximity of various styles of
mineralisation. Multi-element analyses have been done on
leaves, twigs and bark, with repeat sampling to test for seasonal
variation (Figure 10). Significantly, two sites over gold
mineralisation had detectable gold with 0.6 - 1.4 ppb Au in
twigs, and 0.2 - 0.4 ppb Au in leaves. Significant levels of As, Cu
and Zn report in leaves and twigs. All other elements in the suite
of 24 were below detection. These preliminary results offer
promise of a rapid sampling medium in areas of transported arid
regolith.

FIG 10 - Element uptake through transported regolith via the roots of E Camamdulensis and reporting as anomalous values of Au, As, Pb,
Zn and Cu in various parts of the tree, above some mineral prospects; courtesy of Karen Hulme, Adelaide University.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 37

R D GEE and R R ANAND
Similarly, Ravi Anand with assistance of Professor Pauline
Grierson (University of WA) is conducting biogeochemical
surveys over basemetal and gold mineralisation in the Yilgarn
Craton. Phyllodes, branch wood, bark, litter and roots of several
plant species are being analysed. Many pits were excavated to
collect soil samples and plant roots for microbiological studies.
ANOMALIES FROM DEEP ACID-SULFATE SOIL
SEEPS
Research on the environmental geochemistry of acid-sulfate soils
(ASS) has spin-offs for mineral exploration. ASS seeps develop
in advance of a rising watertable as a result of land clearing,
which brings salt and other solutes to the surface. These seeps
present a potential window on blind mineral deposits.
In the Mt Lofty area Skwarnecki et al (2002) developed a
model to account for high metal discharges. In areas with
sulfide-rich basement rocks, rising ground waters can be rich in
sulfate, and have elevated As, Pb and Zn. These become further
concentrated by evaporative transpiration. In soils of high
organic carbon in waterlogged conditions, cyanobacteria reduce
these sulfates, forming secondary framboidal pyrite and
micro-filamentous authigenic sphalerite in soils near the surface
(Figure 11). With further rise of the watertable these re-oxidise
and produce scums and gels of Al and Fe hydroxy minerals (eg
ferrihydrite) in discharge areas, with high element levels. This
pilot study identified a multitude of anomalies, many of which
correlate with known mineralisation, and some of which are new
unexplained anomalies. This new sampling medium has potential
to produce enhanced anomalies of large footprint, drawn from a
wider basement substrate.
RRAf021-04
FIG 11 - Secondary framboidal pyrite and micro-filamentous
authigenic sphalerite in soils near the surface above acid-sulfate
seeps in the Lofty Ranges. Such occurrences can provide leads to
buried mineralisation at depth; courtesy of Rob Fitzpatrick CSIRO
Land and Water.
Baker and Fitzpatrick (2003) have used lead isotopes to model
the interaction of groundwater with bedrock mineralisation, in
the vicinity of ASS seeps. Lead isotope signatures from
mineralised basement are shown to extend laterally for about one
kilometre in groundwaters. The lessons for exploration are that
soils in ASS seeps are good sampling points to detect blind
mineralisation, but the bulls-eye anomalies are not necessarily
point-source with respect to mineralisation.
GEOCHEMICAL ANOMALIES IN TRANSPORTED
MEDIA
Experience in the Yilgarn, North Queensland and Gawler
goldfields, suggests that sampling in transported overburden of
more than five metres has little predictive value in gold
38 Adelaide, SA, 19 - 22 September 2004
exploration (Butt et al, 2000; Anand et al, 2002; Anand, 2003).
Thus sampling cannot distinguish the negative result (nothing
there) from the null case (not adequately tested). Yet there are
many examples of strong near-surface gold anomalies in 5 - 25 m
of transported material above known mineralisation for
example Matt Dam, Callista, Bronzwing, Deep South, Lawlers
(Madden et al, 1998; Anand and Williamson, 2000; Anand,
2001), Quasar, Stellar (Robertson et al, 2001), Steinway, Safari
Bore, Argo, Apollo, Higginsville (Lintern, 2001).
Some of these cases are best explained by mechanical
redistribution of gold from residual environments. However there
are a few instances (for example Deep South at Mt Gibson,
Safari Bore near Lake Raeside, and Matt Dam prospect near Ora
Banda) where physical mixing is not obvious, and hydromorphic
or biogenic dispersion is suspected, with possible sorption of
gold and pathfinder elements onto clays, goethite, manganese
oxides and carbonates.
Tonui et al (2003) note elevated levels of As, Ni, Cu and Co in
thick transported clays above the Tunderdome deposit in the
Broken Hill region. Keeling et al (2003) describe evidence of
upward capillary movement of copper-bearing solutions into
transported clays 5 - 15 m thick, above the Poona (Moonta SA)
CuAu deposit. Attacamite nodules developed in
alunite-halloysite clays are interpreted to form by recent
acid-sulfate weathering of depositional kaolinite/illite/smectite
clays. This only happens where there is direct contact between
transported clay and underlying weathered porphyry.
Many of these case histories remain inconclusive in terms of
process, and require further detailed research. To evaluate the
hypotheses of metal mobility in transported regolith we need to
understand the nature of the sites of metals in regolith mineral
phases.
MINERAL HOSTS FOR GOLD AND TRACE
ELEMENTS
During weathering of mineral deposits, trace metals including
gold, are displaced from their original host minerals, and after
dispersion become fixed in newly formed secondary mineral
hosts in the regolith. Of most interest are clay minerals, and iron
and manganese hydroxides. Such mineral phases interact
geochemically with weathering fluids, and so have the potential
to trap trace elements, which creates an anomaly in a specific
mineral phase. The new range of modern analytical instruments
enables us to better understand the formation of geochemical
anomalies in the regolith. Micro-mineral mapping combined
with in situ geochemical analyses at the micron scale, points to
new techniques of anomaly detection. The Mineral Hosts
program at LEME employs electron microprobe, micro-
RAMAN, electron back-scattered diffraction and now recently
the laser ablation ICP-MS. The latter instrument uses a focused
laser beam, allowing a full quantitative chemical analysis of
virtually all elements, in an area of 20 - 30 microns diameter.
The advent of LA-ICPMS and a new transmission electron
microscope with full analytical facility is enabling high quality
trace element analysis on much smaller solid samples than has
hitherto been possible (Figure 12). Quantitative analysis is
achieved either by normalising intensities of the observed peaks
to the weight of the sample, or to an internal standard. Le
Gleuher (2003) and references therein, provide a full description
of the LA-ICPMS technique and the development of its
application to regolith samples.
We will also be using the new Australian synchrotron for
characterisation of the adsorption of elements on specific
minerals. These will be the first in situ experiments to
characterise the true situation of metals in weathering products
above a mineral deposit, and crucially, identify whether they
were incorporated as the weathering phases grew, or were
PACRIM 2004
 ADVANCES IN REGOLITH RESEARCH A CRC LEME PERSPECTIVE

A
(Mt Gibson gold project), occurs within a variety of hosts,
including kaolinite spherules, hematite overprints of hematite
clasts, hematite cutans, and calcite matrix. All of this is
consistent with late stage hydromorphic dispersion of gold in
Hematite/Kaolinite-rich core
Au: 107 ppb, transported regolith.
As: 94 ppm Evidence of remobilised gold and pathfinder elements is most
instructive at the Lancefield goldmine, where the oxidised
orebody is overlain by sediments comprising 10 - 20 m of
mottled Permian fluvio-glacial sediments, 3 - 8 m of mottled
Tertiary palaeochannel clays, and 2 m of hardpanised colluvium
750 m (Anand et al, 2004). There is evidence (Pillans, 2004) from
Hematite-rich rim
palaeomagnetic dating of multiple oxidising (weathering) events;
Au: 384 ppb, for example, pre-Permian, post-Permian, post-palaeochannels
As: 179 ppm and post-hardpan events. An early As-Cu-Au hydromorphic
Goethite nodules
system seems to have been adsorbed on the goethite/hematite
Au: 26 ppb, As: 6 ppm mottles in the Permian, and a later Cu-Zn(-Au) system been
2.5 mm
RRAf018-04
adsorbed onto goethite in mottles in the palaeochannels clays.
B
MOBILITY AND MICROBES
2 There is much evidence to show that microbial processes
influence weathering processes in contemporary regolith
environments. Welch and McPhail (2003) are looking at rates
and phases of mineral dissolution reactions using soil,
groundwater micro-organisms and microbial ligands, and
comparing these with inorganic abiotic reactions. Laboratory
5 1 experiments show that in granite samples, there is nearly a
4 ten-fold increase in release of major ions like Fe and Al in
6 organic solutions, compared to inorganic controls. Trace
elements Ga, Ti, Li and REE show preferential mobilisation by
3 organic acids. Presumably this is due to the formation of
metal-organic complexes.
500 m Frank Reith (Reith, 2003a, 2003b) at Australian National
University (ANU) is looking at the role of heterotrophic bacteria
1 Hematite-rich 2 Hematite-rich 3 Goethite-rich in the dissolution, transport and stabilisation of gold in three sites
Au:30 ppb Au: 22 ppb Au: 36 ppb
As: 11 ppm As: 13 ppm As: 71 ppm around Australia Tomakin in temperate southeast NSW, Peak
Cu: 77 ppm Cu: 39 ppm Cu: 267 ppm Hill in semi-arid central NSW, and Palmer River in tropical north
Zn: 16 ppm Zn: 15 ppm Zn: 10 ppm
Queensland. On the dissolution side, selective sequential
4 Goethite-rich 5 Hematite-rich 6 Goethite-rich
Au: 35 ppb Au: 28 ppb Au: 37 ppb leaching shows that most gold in soil is associated with
As: 61 ppm As: 67 ppm As:68 ppm exchangeable clay-bound and carbonate-bound fractions, as well
Cu: 48 ppm Cu: 68 ppm Cu: 59 ppm
Zn: 7 ppm Zn:4 ppm Zn: 18 ppm as organic fractions. Most of this gold can be extracted in the
laboratory with mild organic leaches in the presence of living
FIG 12 - Example of micro mineral mapping by LA-ICPMS. This microflora, over a period of ten to 20 days. In contrast, sterilised
technique is a powerful in situ method of identifying the sites of Au samples released little or no gold to solution. The dissolution
and other element distribution in regolith minerals; courtesy of Rob agent may be amino acids, organic acids or cyanide secreted
Hough CSIRO Exploration and Mining. from soil heterotrophic bacteria like Chromobacterium
violaceum.
hydromorphically dispersed later. It will also indicate in what On the precipitation side, Reith notes that micronuggets
form the metals were transported as ions, complexes or (0.1 - 1.0 mm) have the form of budding cells (Figure 13),
colloids. These fundamental issues need to be addressed to strikingly like Pedomicrobium australiensis. These are presumed
enable a bottom-up approach in elucidating how metal anomalies to be bacterial colonies now fossilised by native gold. Reith
form in the regolith. demonstrates in the laboratory that microbes in the soil are
One important research project addresses the trace element active, and capable of precipitating amorphous ferrihydrite on
content of the certain diagnostic clay minerals (kaolinite, planted gold flakes in a form identical to the micron-scale fossil
interstratified clay minerals), iron oxides, phosphates, as well as buds. Some species of bacteria (and fungi) are able to
amorphous materials. These investigations will establish the accumulate gold in cell walls, replacing ferrihydrite (Figure 13).
nature of geochemical dispersion patterns in various regolith DNA staining on gold flakes shows the presence of biofilms on
units and the probable mechanisms involved. It will enable us to natural gold flakes. In conjunction with Rogers (2004), the
determine which selective leaches will, or will not, work. One functional 16s gene sequence of a bio-film from a single gold
significant finding to date is that trace metals of Cu, Zn and Au flake from Palmer River, has been amplified and identified by gel
are preferentially absorbed and trapped in some specific types of electrophoresis. This characterises the particular bacterium
clay minerals. If we could develop a cost effective way to species.
beneficiate or selectively analyses these specific host phases, it Reith also noted an anomalously large count of B cereus
would provide a new technique to regolith sampling, to enhance spores in soils over the Tomakin deposit in effect a biological
the otherwise swamped anomaly. anomaly. The role of bacteria in the precipitation of gold has
Recently Hough et al (2003) have shown that significant been recognised for several years. However this new research has
anomalous gold in hardpans and clastic ferricrete some 20 m for the first time isolated the molecular gene structure of the
above the primary/supergene gold deposit in the Enterprise Pit mobilising agent.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 39

R D GEE and R R ANAND
A B
20m 2m
C D
10m 10m
RRAf017-04
FIG 13 - (A) Fossilised colonies of Pedomicrobium in depressions
of natural gold flake; (B) close-up of budding cells Tomakin
NSW; (C) Baciliius cerus in ferrihydrite, with striking analogy; (D) to
form of gold globules precipitated on gold flakes incubated in
Tomakin soils; courtesy of Frank Reith ANU.
SELECTIVE/PARTIAL EXTRACTIONS
Experimentation with partial extraction techniques started in the
1980s, in the pursuit of subtle less-obvious geochemical
signatures that were swamped in non-diagnostic elements of the
background noise. This follows the advent of ultra-fine
detection limits. The basis of this technique is the selective use of
one or more leaches within a range of weak, moderate or strong
dissolution agents. The leachate can include a spectrum from
de-ionised water, enzyme leach, hydroxylamine hydrochloride,
acetic acid, dilute HCl and complete dissolution with aqua regia.
The weakest leaches dissolve mainly the most recently
introduced metals that remain in water-soluble form. Other
leaches dissolve specific secondary minerals. The basic
presumption behind these techniques is that orebodies
continuously but slowly release metal ions during weathering.
These ions are presumed to rise more-or-less vertically to the
surface where they are variously adsorbed by weathering phases
such as clays, carbonates, amorphous or crystalline Fe and Mn
oxides.
One example is the Mobile Metal Ion (MMI-proprietary leach)
technique developed by the Geochemistry Research Centre in
WA. This focuses on the more loosely bound metal ions in the
soil, and employs as many as eight weak leaches to extract a
range of elements diagnostic of specific ore deposit associations.
The signals are a series of stacked response ratios (the ratio of
element response to background). The values of this technique
are that it:
gives a vertical response,
produces sharper anomalies,
assists in reducing drilling costs,
discriminates mechanical and detrital anomalies, and
applies to a range of climatic and regolithic conditions.
Another proprietary line Regoleach utilises a moderate leach,
and samples the lower soil horizons. It focuses on amorphous
iron and manganese oxides, as well as organic materials and
carbonates.
These techniques clearly give positive responses in shallow
regolith, and there are claims of responses from orebodies as
deep as 700 m. Comparative studies (Gray et al, 1999) in
40 Adelaide, SA, 19 - 22 September 2004
different regolith environments, including transported regolith,
show that weak, moderate and total dissolution methods give
similar responses in areas where the transported cover is less
than 10 m. Where the overburden was thicker, total analysis and
both proprietary and conventional partial leaches all yielded
some elevated concentrations of one or more elements unrelated
to the ore composition. However, Carey and Dusci (1999)
attribute anomalies in light rare earth elements detected by
enzyme leach analysis to the presence of gold mineralisation
buried beneath 10 - 15 m of saturated elements in saline playas.
With gold, much depends on whether it is loosely bound or hard
bound. Gold adsorbed onto clays is presumably hydromorphic in
origin. But it also seems that hydromorphic gold can with time
become hard bound into ferruginous nodules.
What is not universally agreed in the exploration industry is
whether anomalies produced by labile metals can be shown to
work where conventional total dissolution methods do not. This
is one of the unresolved areas of gold exploration in Australia,
which is not without it emotional protagonists and sceptics on
either side.
Perhaps the greatest uncertainty with the partial leach
techniques is that ion mobility is presumed to happen in nature,
but we do not know by what mechanism this may happen.
PARTIAL LEACH ISOTOPE GEOCHEMISTRY
If labile elements really do exist in the regolith, one means of
testing the connectivity between covered orebodies and surface
geochemical expression, is to use the fingerprinting and tracing
capability of Pb isotopes. This is being done under the
industry-sponsored AMIRA Project 618 on six sites where
partial leach techniques are reputed to have registered. This
should throw light on whether such mechanisms as gas phase
diffusion, electrochemical transport, biogenic factors or
groundwater movement are responsible, and indeed if the partial
leach techniques do in fact register mineralisation under cover. In
theory partial extraction techniques will preferentially dissolve
transported metal ion anomalies, which in turn should have
isotopic signatures different to background rock lead. Results to
date are encouraging in that the analytical technique for such
ultra-low levels of lead is successfully developed, and
connectivity is confirmed in some sites. However at these
ultra-low levels of metal content, the anthropogenic effects from
previous exploration, mining and processing operations may
submerge the signature of genuine labile lead. This technique
remains confidential to the industry sponsors.
SOIL GAS
Dating back to the mid 1950s in the Soviet Union, explorers have
considered the gaseous products of sulfide oxidation as a suitable
sample medium for targeting hidden deposits. Western
geochemists and geologists made serious forays into soil gas
geochemistry in the early 1980s. Klusman (1993) authored the
definitive text on the subject, incorporating summaries of past
work and numerous case studies from different sample media in
many applications.
A new gas technique (www.sdpsoilgas.com.au), again largely
an empirical method, seeks to identify the diagnostic volatile
compounds associated with different types of metalliferous
deposits, rather than attempting to directly sense ore and
pathfinder elements. Soil desorption pyrolysis (SDP) measures
traces of a range of volatiles that presumably have moved into
the soils as micro bubbles, and become adsorbed onto clays and
organics in the ultra-fine 0.2 - 2.0 m size range. As has been
demonstrated above oil and gas deposits, hydrocarbons up to C7
chain migrate upward as a flux through rock, regolith and water.
However gases from ore deposits are different to those from
PACRIM 2004
 ADVANCES IN REGOLITH RESEARCH A CRC LEME PERSPECTIVE
enveloping rock sequences. Gases desorbed from their temporary
hosts clays and soil organics are detached by pyrolysis and
analysed. These include:
aliphatic and aromatic hydrocarbons,
halogenated hydrocarbons,
organic sulfur gases,
carbonyl sulfides,
He and Ar, and
SO2, H2S, CO2.
The proof of concept for this technique will require building a
large library of SDP analyses from about 7000 test cases to
produce templates for various styles of mineralisation, for
example, MVT, BHT, VMS, Carlin, IOCG, lode gold, kimberlite.
Processing and interpretation will involve use of ratios of
various sulfur and hydrocarbon gases. The sum of all anomalies
in a mineral-system template would define the existence of an
anomaly, and the proportion of the total number of ratios that are
anomalous would define the strength of the anomaly. LEME is a
participant in this soil gas consortium.
HYDROGEOCHEMISTRY
Ground water should be an ideal sampling medium because it of
its interaction with the materials through which it moves, and the
apparent convenience and ease of sampling. However, it has not
yet proven reliable for gold search in the Yilgarn Craton. Gray
(2001) believes this is because gold in solution is liberally and
extensively smeared in the saline groundwaters. Sampling to date
shows no obvious local patterns that could pin-point gold
deposits. He further argues that gold in solution relates more to
regional patterns of salinity, pH and Eh, rather than local effects.
However indications to date are that hydrogeochemistry may
well be applicable to basemetals. For example de Caritat and
Kirste (2004) have collected 300 bore water samples from
around Broken Hill, some in transported sediments 100 m thick.
These have been analysed for major and trace elements, sulfur
strontium and lead isotopes, and then subject to geochemical
modelling. Some samples show an excess of sulfur (as sulfate)
over what can be ascribed to evaporation or mixing with connate
waters. All these have low 34S values indicative of derivation
from primary sulfides. Moreover they also record 207Pb/204Pb and
206
Pb/204/Pb ratios comparable to known mineral deposits in the
Broken Hill region. This technique is sensing nearby sulfide
accumulations under as much as 100 m of sediments in the
basins around Broken Hill. In terms of practical application, the
diagnostic key to this technique is the ease and cost of analyses
for sulfate and sulfur isotopes the latter can be of the order of
$25 per sample. This gives encouragement that a method for
first-pass screening in greenfield areas can be developed.
There is also some evidence of striking and local anomalies
related to nickel sulfide deposits (Figure 14). This is a fertile
field for future research which LEME will pick address.
DISPERSION MECHANISM THROUGH
TRANSPORTED REGOLITH
The selective, partial and mobile phase techniques have indeed
passed the proof-of-concept stage, but at least in the Australian
regolith, the relative advantages over total analyses is yet to be
demonstrated. In general, partial extraction analyses may yield
sharper anomalies, with higher contrast. Perhaps the most
convincing examples of success of selective methods over total
bulk approaches lie in the glacial tills of Canada (Cameron et al,
2004; Hamilton, 1998). Examples of partial-leach anomalies
over base metal deposits, buried by 10 m-thick glacial tills about
10 000 years old, provide parameters and constraints on the
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
FIG 14 - Highly anomalous dissolved Ni content in groundwaters
around Harmony nickel deposit, in comparison with various Yilgarn
groundwaters; courtesy of David Gray CSIRO Exploration and
Mining.
processes that may be operating. For example, the most striking
example of advective transport occurs at Spence Porphyry
deposit in Chile where the earthquake-induced (seismic)
pumping of mineralised groundwater to the surface through
fracture zones is recognised. The assemblage of elements found
in the soils above fracture zones is similar to that found in
groundwaters at 60 m depth. Thus, it is necessary to understand
the processes involved, so that the evaluation of the techniques
for Australian conditions can be made.
A number of mechanisms have been proposed as summarised
below.
Groundwater advection is the lateral movement of water
and its contained solutes as a result of piezometric potential.
Variants of this, known as convection, can be due to thermal
plumes related to exothermal oxidation, or density
adjustment of hypersaline layers.
Ionic diffusion can only take place in the water saturated
zone. According to Cameron et al (2004) the diffusion
coefficient in tills is too low to account for the generation of
surface geochemical anomalies in glacial tills, in the available
time.
Capillary is the vertical migration of pore water due to
surface tension of water films on grains. In porous sands it is
likely to be totally counteracted by influx from rainfall.
However capillary rise should be greater, albeit slower in
clays where the grain size is smaller and the infiltration is
less. John Keeling (pers comm) attributes the generation of
Cu in the Moonta clays to this process.
Barometric pumping refers to the process where cycles of
high and low barometric pressure first force air into the earth
and then withdraw a mixture of the air plus gases that were in
the rock (Cameron et al, 2004). For all practical purposes,
barometric pumping applies only to fractured rock. At the
Mike gold deposit, Nevada, tests of metal content of soil air
using collectors containing activated carbon failed to detect
greater amounts of metals than blank collectors sealed in
plastic bags (Cameron et al, 2004).
Electrochemical transport involves the formation of
electrochemical cells that are generated around sulfide bodies
as a result of their oxidation. Such cells are often detectable by
negative self-potential (SP) anomalies above the bodies.
Cameron et al (2004) invoked transport along redox gradients
rather than advection or diffusion process which are
ineffective in moving waters over the last 10 000 years.
41
R D GEE and R R ANAND
Gas transport of metals can take place by attachment of
metal ions or organometallic complexes to gas bubbles as
they rise from bedrock through regolith.
Biochemical transformation by vegetation is a potential
mechanism of rapid metal transfer. Vegetation is known to
take up Au, Ni, Cu, Pb, Zn and As with specific flora being
adapted to substrates with high concentration of these
elements. In drier climates, many pheratophyte plants have
dimorphic root systems with laterals and sinker-tap roots.
The latter roots acquire water from deeper groundwater
source, especially during summer (as demonstrated by
deuterium isotopes). The depth of rooting is critical to the
ability of vegetation to transfer water and ore metals. A
global rooting depth survey suggests that deep roots,
especially sinkers, are ubiquitous with 10 m plus depths
reached and confirmed in several climatic settings. Uptake of
trace elements by plants is facilitated by production of
ligands by plant roots (Romheld, 1991). Australian native
plants are known for producing cyanide, oxalate and citrate.
It is likely that other compounds are produced which have
the ability to solubilise trace elements.
Bioturbation by burrowing insects (termites, ants,
earthworms) and other organism is a powerful mechanism to
bring particles from depth to the surface. On the basis of
optical luminescence in quartz, Pillans (ANU, pers comm)
estimates that the soil to a depth of 20 m can be recycled
over a period of about a million years or less.
The role of microbes in transferring metals upwards is
restricted, but they affect most of the processes responsible
for metal transfer. Microbial metabolism affects the kinetics
of many hydrochemical processes, especially redox and
sulfide oxidation.
Considering all the above possible mechanism, it is difficult to
see anything other than vegetation and mechanical bioturbation
that will directly bring elements to any significant degree in
semi-arid to arid areas where the watertable is deep. The
limitations of all groundwater-transfer mechanisms is the upper
limit to which groundwater rises, which is the watertable and the
capillary fringe, except for the case of seismic pumping. In the
Australian landscape, groundwater occurs commonly more than
5 m below surface except in lower landform sites. Therefore
other mechanisms are necessary to transfer metals from the water
table upwards into the soil. Surrogate organic volatiles with
indirect signatures of mineral deposits have potential to bring
elements up through the vadose zone. It is these areas that future
research should be directed, in the thick and arid Australian
regolith.
CONCLUDING COMMENTS
There have been some important developments over the years in
exploration geochemistry which have become routine tools.
However, many of the advances reported over the last few years
are considered to be largely incremental. They relate to
enhancing known techniques, and understanding regolith
architecture. The focus for the future research is to develop an
understanding of mechanisms that are responsible for dispersing
ore and pathfinder indicator elements through transported cover
in well controlled dynamic 3D regolith-landform frameworks.
This will allow us to be more predictive and less empirical. The
major challenge is to identify plausible mechanisms that will
raise diagnostic labile elements through the vadose zone.
Although there has been some important progress in
understanding the physical/chemical processes in some
environments, soil gas and biological processes that may in fact
be the dominant contributing process.
42 Adelaide, SA, 19 - 22 September 2004
ACKNOWLEDGEMENTS
This presentation is drawn from a selection of recent and current
projects of LEME that relate to exploration through the regolith.
Material comes from recent LEME monographs, and from
individual colleagues, for which we are grateful for their help.
Some of the source material relates to interim releases, and does
not fully lend itself to referencing in the classical sense. Some
interim releases are available on our website
www.crcleme.org.au. We express thanks to Frank Reith, Bear
McPhail, Brad Pillans, John Keeling, Cajetan Phang, Mel
Lintern, Ray Smith, David Gray, Mat Cornelius, Rob Hough,
Charles Butt, Graham Carr, Geoff Denton, Ken McQueen, Steve
Hill, Patrice de Caritat, Dirke Kirste, Lisa Worrall, Richard Lane,
Rob Fitzpatrick, Marian Skwarnecki, Merhooz Aspandair, and
indeed many others.
REFERENCES
Anand, R R, 2001. Evolution, classification and use of ferruginous
materials in gold exploration, Yilgarn Craton, Western Australia,
Geochemistry Exploration Environment Analysis, Vol 1, pp 221-236.
Anand, R R, 2003. Importance of regolith for gold exploration in the
Yandal gold province, Yilgarn Craton, in Yandal Gold Province:
Geoscience and Exploration Success (Eds: K S Ely and G N Phillips)
219 p (CSIRO Exploration and Mining: Melbourne).
Anand, R R, Fraser, S F, Jones, M R, Shu Li, Munday, T J, Phang, C,
Robertson, I D M, Scott, K M, Vasconcelos, P, Wildman, J E and
Wilford, J, 2002. Geochemical exploration in regolith-dominated
terrain, north Queensland, Final Report for the CRCLEME/AMIRA
Project 417, Open File Report 120, (CRC LEME), 130 p plus
Appendices.
Anand, R R, Hough, R, Phang, C, Norman, M and Gleuher, M, 2004.
Regolith mineral hosts for gold and other base metals, Minerals
Exploration Seminar Abstracts (CRC LEME: Perth).
Anand, R R and Paine, M, 2002. Regolith geology of the Yilgarn Craton,
Western Australia: implications for exploration, Australian Journal
of Earth Science 49:3-163.
Anand, R R, Smith, R E, Phang, C, Wildman, J E, Robertson, I D M and
Munday, T J, 1998. Geochemical exploration in complex lateritic
environments of the Yilgarn Craton, Western Australia, Open File
Report 58, 297 p (CRC LEME: Perth).
Anand, R R and Williamson, A, 2000. Regolith evolution and
geochemical dispersion in residual and transported regolith Calista
deposit, Mt McClure district, in Yandal Greenstone Belt, Regolith,
Geology and Mineralisation (Eds: G N Phillips and R A Anand)
Australian Institute of Geoscientists Bulletin, 31:333-349.
Butt, C R M, Lintern, M J and Anand, R R, 2000. Evolution of regoliths
and landscapes in deeply weathered terrain implications for
geochemical exploration, Ore Geology Reviews, 16:167-183.
Cameron, E M, Hamilton, S M, Leybourne, M I, Hall, G E M and
McClenaghan, 2004. Finding deeply buried deposits using
geochemistry, Geochemistry Exploration Environment Analysis,
4:7-32.
Carey, M L and Dusci, M E, 1999. WMC Resources exploration success
in lake terrains applications of element dispersion, Kambalda, in
Proceedings Third Australian Regolith Conference Regolith 98, New
Approaches to an Old Continent, (Eds: G Taylor and C Pain), pp
135-148 (CRC LEME: Perth).
Cornelius, M, 2004. Regional laterite geochemistry of the central Yilgarn,
Abstract Volume Minerals Exploration Seminar 2004 (CRC
LEME: Perth).
Cornelius, M, Smith, R E and Cox, A, 2001. Laterite geochemistry for
regional exploration surveys a review of sampling strategies,
Geochemistry Exploration Environment Analysis, Vol 1, pp 211-220.
Craig, M, 2001. Regolith mapping for geochemical exploration in the
Yilgarn Craton, Western Australia, Geochemistry Exploration
Environment Analysis, Vol 1, pp 383-390.
Davey, A J and Heinson, G, 2004. Electromagnetic profiling of
palaeochannels in the Girilambone Region NSW, in Proceedings of
Exploration Field Workshop Cobar Region (Eds: K McQueen and K
M Scott) (CRC LEME: Perth).
PACRIM 2004
 ADVANCES IN REGOLITH RESEARCH A CRC LEME PERSPECTIVE
Drown, C, 2003. The Barnes gold discovery, MESA Journal, 28:4-9
(Primary Industries and Resource Development South Australia).
Gray, D J, 2001. Hydrogeochemistry in the Yilgarn Craton, Geochemistry
Exploration Environment Analysis, Vol 1, pp 253-264.
Gray, D J, Lintern, M J and Longman, G D, 1990. Chemistry of gold in
some Western Australian soils, CSIRO Australia, Division of
Exploration Geoscience Report 126R, 62 p.
Gray, D J, Sergeev, N B, Britt, A F and Porto, C G, 2001. Supergene
mobilisation of gold and other elements in the Yilgarn Craton final
report, CRCLEME/AMIRA P504 supergene dispersion of gold.
CSIRO Exploration and Mining Restricted Report 757R 79 p.
Gray, D J, Wildman, J E and Longman, G D, 1999. Selective and partial
extraction analyses of transported overburden for gold exploration in
the Yilgarn Craton, Western Australia, Journal of Geochemical
Exploration, 67:51-66.
Hamilton, S M, 1998. Electrochemical mass-transport in overburden: a
new model to account for the formation of selective leach
geochemical anomalies in glacial till, Journal of Geochemical
Exploration, 63:155-172.
Hough, R, Phang, C and Anand, R R, 2003. Mineral hosts for gold: A
study of transported overburden at the Enterprise pit, Mount Gibson
gold deposit, in Proceedings CRC LEME Regional Regolith
Symposia: Advances in Regolith (Ed: I C Roach).
Idnurm, M and Senior, B R, 1978. Palaeomagnetic ages of Late
Cretaceous and Tertiary weathered profiles in the Eromanga Basin,
Queensland, Palaeogeography Palaeoclimatology Palaeoecology,
24:263-277.
Keeling, J, Mauger, A J, Scott, K M and Hartley, K, 2003. Alteration
mineralogy and acid-sulphate weathering at Moonta copper mines,
South Australia, in Advances in Regolith (Ed: I C Roach), pp
230-233 (CRC LEME).
Klusman, R W, 1993. Soil Gas and Related Methods for Natural
Resources Exploration, 483 p (John Willey and Sons: New York).
Lane, R and Worrall, L, 2002. Interpretation of airborne electromagnetic
data: Summary of the Challenger workshop, Geoscience Australia
Record 2002/02.
Le Gleuher, M, 2003. Mineral hosts for gold and trace elements in the
regolith, Boddington and Mt Percy Gold deposits, Western Australia,
CRC LEME Restricted Report 196R, 95 p.
Lintern, M J, 1999. Study of the distribution of gold in soils at Mt Hope,
Western Australia, CRC LEME Report 65.
Lintern, M J, 2001. Exploration for gold using calcrete lessons from the
Yilgarn Craton, Western Australia, Geochemistry Exploration
Environment Analysis,Vol 1, pp 237-252.
Lintern, M J, 2004. The South Australian Regolith Project final report
summary and synthesis, CRC LEME Report 156.
Madden, J, Anand, R R and McNaughton, N, 1998. Geochemical
dispersion into transported overburden, Deep South Gold Deposit,
Mt Gibson, Western Australia, in Proceedings Third Australian
Regolith Conference, Abstract: Regolith 98, New Approaches to an
Old Continent (Eds: A F Britt and L Bettenay) p 49.
Mazzucchelli, R H and James, C H, 1966. Arsenic as a guide to gold
mineralisation in laterite covered areas of Western Australia.
Transactions of The Australasian Institute of Mining and Metallurgy,
75(B):286-294.
McQueen, K, 2004. Element fractionation and mineral hosts in the
regolith, in Abstract volume Minerals Exploration Seminar 2004.
(CRC LEME: Perth).
McQueen, K G, Pillans, B J and Smith, M I, 2002. Constraining the
weathering history of the Cobar region, western NSW, in Geoscience
2002: Expanding Horizons. Abstracts of the 16th Australian
Geological Convention: (Ed: V P Preiss) p 426 (Geological Society
of Australia).
Pain, C F, Chan, R A, Craig, M A, Gibson, D L, Kilgour, P and Wilford, J
R, 2001. RTMAP Database Field Book and Users Guide,
CRCLEME, Report 13, Perth.
Pidgeon, R T, Brander, T H and Lippolt, H J, in press. (U+Th)-4He dating
of laterite from the Darling Range, Western Australia, Australian
Journal of Earth Sciences.
Pillans, B, 2004. Geochronology of the Australian Regolith, in
Regolith-Landscape Evolution Across Australia: A compilation of
regolith-landscape case histories and landscape evolution models
(Eds: R R Anand and P de Broekert), 12 p,
http://crcleme.org.au/RegLandEvol/Geochron_of_Aust_Regolith.pdf
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Pillans, B and Bateman, R, 2000. Palaeomagnetic dating of Phanerozoic
weathering imprints, Mount Percy mine, Kalgoorlie, Western
Australia, in Understanding Planet Earth: Searching for a
Sustainable Future: Abstracts of the 15th Australian Geological
Convention (Eds: C G Skilbeck and T C T Hubble), p 390
(Geological Society of Australia).
Pillans, B, Tonui, E and Idnurm, M, 1999. Palaeomagnetic dating of
weathered regolith at Northparkes mine, NSW, in New Approaches to
an Old Continent: Perth (Eds: G Taylor and C Pain), pp 237-242
(Cooperative Research Centre for Landscape Evolution and Mineral
Exploration).
Pontual, S and Merry, N, 1996. An exploration strategy to aid the
differentiation of residual from transported kaolinites using field
based spectral analysis, Restricted Report, Auspec International,
Melbourne.
Reith, F, 2003a. Evidence for a microbially mediated biogeochemical
cycle of gold a literature review, in Advances in Regolith (Ed: I C
Roach), pp 336-341 (CRC LEME).
Reith, F, 2003b. Relevance of micro-organisms for the formation of, and
exploration for, gold deposits in the Australian regolith, The AusIMM
Bulletin, Nov-Dec.
Robertson, I D M, 2001. Geochemical exploration around the Harmony
gold deposit, Peak Hill, Western Australia, Geochemistry
Exploration Environment Analysis, Vol 1, pp 277-288.
Robertson, I D M and Gray, D J, 2004. Harmony gold deposit, Baxter
Mining District, Western Australia, in CRC LEME Monograph
Regolith Expression of Australian Ore Systems, (Eds, Comp: C R M
Butt, M Cornelius, K M Scott and I D M Robertson),
http://www.crcleme.org.au/
Robertson, I D M, King, J D and Anand, R R, 2001. Regolith geology
and geochemical exploration around the Stellar and Quasar gold
deposits, Mt Magnet, Western Australia, Geochemistry Exploration
Environment Analysis Vol 1, pp 353-364.
Rogers, S, 2004. Role of biota in mineral transformations and transport,
Abstract volume Minerals Exploration Seminar 2004, (CRC
LEME: Perth).
Romheld, V, 1991. The role of phytosiderophores in acquisition of iron
and other micronutrients in graminacious species: an ecological
approach, Plant and Soil, 130:127-134.
Schmidt, P W and Clark, D A, 2000. Palaeomagnetism, apparent polar
wander path and palaeolatitude, in Billion-Year Earth History of
Australia and Neighbours in Gondwanalan (Ed: J J Veevers), pp
12-17 (Gemoc Press: Sydney).
Schmidt, P W and Embleton, B J J, 1976. Palaeomagnetic results from
sediments of the Perth Basin, Western Australia, and their bearing on
the timing of regional lateritisation, Palaeogeography
Palaeoclimatology Palaeoecology Vol 19, pp 257-273.
Scott, K M, Chambers, D I, Ransted and T, Kairaitis, 2003. Wyoming
gold deposit, in Regolith Expressions of Australian Ore Deposits
(CRC LEME), www.crcrleme.org.au
Sheard, M J and Robertson, I D M, 2004. Regolith characterisation and
geochemistry as an aid to mineral exploration in the Harris
Greenstone Belt, Central Gawler Craton, South Australia, CRC
LEME Report 155.
Skwarneck, M S, Fitzpatrick, R W and Davies, P J, 2002. Geochemical
dispersion at the Mount Torrens Pb-Zn prospect, South Australia,
with particular emphasis on acid sulphate soils, CRC LEME
Restricted Report 174, 68 p.
Smith, R E, in press. Recognising, in cover, detritus from the Gossan Hill
VHMS deposit, Australia, Abstract, SEG 2004, Predictive Mineral
Discovery Under Cover, Perth.
Smith, R E, Anand, R R and Alley, N F, 2000. Use and implications of
palaeoweathering surfaces in mineral exploration in Australia, Ore
Geology Reviews, 16:185-204.
Smith, R E, Campbell, N A and Litchfield, R, 1984. Multivariate
statistical techniques applied to pisolitic laterite geochemistry at
Golden Grove, Western Australia, Journal of Geochemical
Exploration, 22:193-216.
Taylor, G and Shirtliff, G, 2003. Weathering: cyclical or continuous? An
Australian perspective, Australian Journal of Earth Science, 50:1-9.
Tonui, E, de Caritat, P and Leyh, W, 2003. Geochemical signature of
mineralisation in weathered sediments and bedrock, Thunderdome
prospect, Broken Hill region western New South Wales: implications
for mineral exploration under cover, Geochemistry Exploration
Environment Analysis Vol 3, pp 263-280.
43
R D GEE and R R ANAND

Welch, S and McPhail, D C, 2003. Mobility of major and trace elements
during biologically mediated weathering of granite, in Advances in
Regolith (Ed: I C Roach), pp 437-440 (CRC LEME).
Wright, J H, Eshuys, E and Anand, R R, 1999. Bronzewing the role of
regolith-landform control and regolith geochemistry in the discovery
of a large gold deposit, in Proceedings Third Australian Regolith
Conference: Regolith 98, New Approaches to an Old Continent (Eds:
G Taylor and C Pain), pp 173-181, (CRC LEME: Perth).

44 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

Overview of Mineral Project Risk Issues and Role of
Mineral Industry Professionals (Especially Consultants) in
Risk Reduction Strategies
M J Lawrence1
ABSTRACT
The various risk components to be addressed by explorers and project
developers (and their investors) to minimise and/or manage the risk of
Pacrim mineral investment are summarised to provide an overview of the
issues involved and to set the scene. It then highlights the role of
individual Mineral Industry Professionals (especially independent
consultants) who are at the heart of any risk reduction or risk
management strategy. The involvement of accredited
qualified/experienced (competent) and reputable professionals is essential
to produce realistic and reliable results that gain and maintain investor
confidence. Today, high ethical standards as well as technical competence
are sought by government legislation/licensing regimes and initiatives of
professional bodies like AusIMM, MICA, Professions Australia, Council
of Professions, etc. All attempt to regulate and guarantee reputable
individual professional behaviour (particularly in the context of currently
perceived high corruption levels in many Pacrim countries). Brief
comment will discuss the usefulness of accreditation and ethics codes
plus other specific codes (eg JORC/VALMIN and their international
analogues in Canada, South Africa, etc) in reducing risks.
INTRODUCTION
Most of mankinds endeavours carry some quantum of risk. It is
not so much about limiting or removing these natural phenomena
(although it is natural to try to do so) but about managing it,
usually by trying to reduce the associated uncertainty that is its
co-conspirator against success. All within a timeframe that is
never long enough.
It is relevant, however, to be aware of the proportional
risk-reward relationship since the rational assumption of risk
represents a business opportunity. No business can be entirely
risk-free thats the opportunity.
It is useful to outline, in the context of this paper, that the
author considers risk (in money terms) not solely as the product
of the probability of an events occurrence (or not) and the
magnitude of financial (or social/environmental) detriment of
that specific event subsequently occurring (or not). Sandman
(1993) calls this the Hazard component to which he adds
another called Outrage. It is his useful attempt to specify the
bio-physical/emotional risk component which can result from the
degree of coercion and morality involved, as well as the ultimate
impact (how catastrophic?).
1. FAusIMM, Managing Director, Minval Associates Pty Limited,
Croydon NSW 2132. E-mail: lawrenceminval@msn.com.au
2. Civil Society is a term that describes the increasingly influential,
community-based organisations which claim to (and often do)
represent the publics conscience, eg those that presently form the
outspoken political pressure groups in the indigenous land rights,
global warming/clean environment and sustainable development
debates.
3. See 1998 Back from the Brink: Reshaping Minerals Tertiary
Education, Report by the Minerals Council of Australia. Also see
Catalano, S, 2000. The risky degree incorporation of risk
management into the undergraduate degree, in Proceedings AusIMM
Student and Young Professionals Conference 2000 (who promoted
the teaching of OH&S principles and practice at university).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Those involved in mineral project risk minimisation and/or
management broadly includes the explorers and project
developers (owners and their investors). Each has specific
imperatives and responsibilities. Owners means directors,
management and employees, but involves suppliers, contractors
and consultants in this context. Investors could be individuals,
government-run or private companies (listed or not on a stock
exchange), banks or investment funds. It is all about good
governance and the duty of care corporate or otherwise.
Technical risk is associated with each input to a Feasibility
Study or Exploration Program (explaining the tenements
prospectivity) and getting them right clearly underpins any
projects attractiveness. They tend to be reasonably well
addressed, but beware cutting edge innovation which magnifies
rewards and risks.
Obviously, to varying degrees, all have to take account of the
local socio-political system and regulatory environment
applicable to the particular Pacrim country within which they are
operating, but todays investment regime is global. This means
that the socio-political systems and regulatory environments of
both the host and the home country (or countries) must be
considered; as well as all those that importer of the mines
products. This mix of local and international stakeholders can be
large, complicating the risk management task.
Every country also has a unique physical and
cultural/religious environmental setting that can further
complicate a projects risk profile. Communities have concerns
over any change and financial benefits do not suit everyone,
everywhere.
Often these host and source countries have a very sophisticated
Civil Society2 that is adept at media management and well
organised protests, as well as having considerable political
influence at a local, national and even international level. Hence,
mineral project developers who ignore the concerns of civil
society do so at great risk.
In any event, minimisation of conflict at the interface between
exploration and mining and the affected community is critical
today. The minerals industry has genuinely changed, MCA Chief
Executive, Mitchell Hooke, said at MCAs 2003 Sustainable
Development Conference in Brisbane:
Effective community engagement has been
underscored as a fundamental driver of the
business case for sustainable development. The
emergence of relationships over regulation
and partnerships over process has been
recognised as a highly successful means of
delivering solutions and outcomes that add real
value to both the industry and the community.
New graduates must become more aware of these issues and
the need to acquire these life skills, too, which are not taught at
university (see Lawrence, 2001a, a presentation to AusIMMs
2001 Youth Conference in Brisbane)3.
Ethical investing protocols are also increasingly being applied
by individuals, companies and fund managers, as they decide
between competing investment opportunities, irrespective of
where the project is located. They apply not only to mining but to
45
M J LAWRENCE
other business activities, such as exploiting timber and using low
cost labour. Nevertheless, it is fair to say that these comments
particularly apply to countries of the Pacrim, since they have
been exploited in the past and they are now perceived to have
high environmental and cultural value to the rest of the World
that are under current threat.
The same philosophic shift applies today to workplace safety
the occupational health and safely revolution is well advanced in
Pacrim countries, being justified on humanitarian and
economic/operational4 grounds. Research work by the minerals
industry Safety and Health Centre (University of Queensland)
generated a national Mineral Industry risk assessment guideline
(NMISHRAG, 2004) to assist in the design and execution of
health and safety risk assessments (available at
www.mishc.uq.edu.au)5.
As with most risks, it is always better to identify and attempt to
manage them at the start, then have them unexpectedly require
attention subsequently. It is within this multi-faceted context that
mineral project exploration and development risk is discussed
more specifically below. It is a complex subject and many papers
at many conferences6 have attempted to address these issues
before, so this paper will of necessity be merely an introductory an
overview to set the scene and explore less obvious areas of risk.
RISK CATEGORIES
There are numerous ways to try and catalogue the various risks
faced by mineral project developers. Use could be made of the
detailed listing of risks in Standards Australia (1999) publication
Australian/New Zealand Risk Management Standard (AS/NZS)
4360. Edwards and Bezzina (2003) provides a good outline of its
approach and usefulness. In it, reference was made also to the
numerous AS/NZS Handbooks7 that provide specific guidelines
on applying AS/NZS 4360 and also indicate how it can be
specifically applied to various industries.
Risks can be categorised as being legal (title security,
organisational, contracts, rule of law/corruption, intellectual
property); technical (geological, mining, metallurgical, human
resources, procurement, construction and maintenance);
commercial (economic conditions, marketing, commodity
prices/hedging and forex); environmental (impact/pollution
limits and community opposition); political (policy issues, eg
taxation and capital repatriation); safety and regulatory
(systems/reporting, health, operational conditions, royalties and
charges); and force majeure (natural disasters, industrial/civil
unrest, terrorism/war).
Some use the term sovereign risk as primary risk filter. It
combines the socio-political and regulatory uncertainty of a
country in which a mineral deposit occurs, including some force
majeure-type risks (like strikes, riots and war). In essence, this
risk encapsulates the likelihood of expropriation and the
4. It is not only about controlling production costs and throughputs, but
avoiding huge fines or imprisonment and loss of reputation which
magnify any operational inefficiencies.
5. See King, R, 1999. A legal perspective of the impact of the new
mining safety and health legislation, in Proceedings 1999
Queensland Mining Industry Health and Safety Conference,
Yeppoon, 22-25 August (noting that mine workers did not consider
that safety issues were their concern and that mine management
lacked respect from the workers for its commitment to safety).
6. Particularly see numerous relevant papers in Proceedings Mining Risk
Management Conference 2003, 488 p (The Australasian Institute of
Mining and Metallurgy: Melbourne).
7. For example: HB141 (Risk Financing); HB142 (Basic Introduction to
Managing Risk); HB250 (BHP Billiton and Qantas case study
guides); HB231 (Information Security Risk Management); and
HB203 (Managing Environmental Risk).
46 Adelaide, SA, 19 - 22 September 2004
repudiation of contracts. For those who rely upon external
sources of finance, it is major risk to address, because of its
inherent unpredictability. Most of the other major mineral project
risks (especially technical risk) can be basically solved by money
and research time. Emotion drives sovereign risk and often
reflects a clash of cultures and value systems.
Since 1997, The Canadian Fraser Institute has conducted an
annual survey of metal mining and exploration companies to
assess how their perception of a countrys mineral endowment
vis-B-vis that countrys public policy (including taxation and
regulation) their Policy Potential Index affects exploration
investment decisions. This Index reflects the effects on
exploration funds allocation of regulatory duplication and
inconsistencies; uncertainty in the administration, interpretation
and enforcement of existing regulations generally; environmental
requirements; taxation policy; uncertainty concerning native land
claims and protected areas; existence and quality of
infrastructure; socio-economic agreements, political stability and
labour issues; and the existence and quality of geological
databases.
The 2003 - 2004 survey results, ranking a countrys
attractiveness (in allocating exploration funds), represent the
opinions of 159 exploration managers from 1021 companies
operating around the world whose views were sought (they spent
$US 642 M in 2002). Some 53 jurisdictions (including the
Canadian provinces/territories, the Australian states and 14
selected US states) were assessed in its 2003 - 2004 survey. To
provide some Pacrim context, Australia scored sixth (ranked third
on technical potential), New Zealand 22nd (technically ranked 50th)
Chile second (technically ranked fourth), Indonesia 50th
(technically ranked sixth) and the Philippines 51st (technically
ranked 20th). Visit www.fraserinstitute.ca to obtain a copy.
Technical risks are another critical criterion as to whether or
not a project is developed. However, they are commonly well
addressed and will not be dwelled upon here. McCarthy (2003)
provided a good summary of the risks involved and their
management. He identified the key risks as:
mine design and scheduling 32 per cent
resources/reserves (geology) 17 per cent
metallurgical testwork (scale-up) 15 per cent
process design/equipment selection 12 per cent
geotechnical analysis nine per cent
cost estimation seven per cent
mining equipment selection four per cent
hydrology four per cent
Another way is to look at risk is to consider the sequence in
which an exploration prospect moves through being a
development project to an operating mine (always keeping in
mind that there can be jurisdictional differences across state and
national borders). Here, the risks associated with having an
equitable interest/agreement recognised in law; and the speed
and certainty of approval processes, land access and tenure are
seen to be paramount. Without a legal right (or title) there is
nothing further to do. The main sources of risk involve the
actions of the exploring/developing parties (particularly the
operator and the contractors); the land owner(s); those in the
affected communities; and the relevant government departments
responsible for the various licences, permits, etc (especially those
involving environmental monitoring and approvals). Some
include within this category the risks associated with law and
order and civil unrest (and attitude to illegal miners); unexpected
changes to the financial/administrative regulatory regime
(including taxation, other national and local governmental
imposts and royalties, capital repatriation rules and corruption);
even the force majeure risks of terrorism, war and expropriation).
PACRIM 2004
 OVERVIEW OF MINERAL PROJECT RISK ISSUES AND ROLE OF MINERAL INDUSTRY PROFESSIONALS
Then, there are the physical or geographic risks related to the
projects location, eg lack of water, floods, fire, wind storms,
volcanoes, earthquakes, tidal impacts, etc. Related to these
natural disaster risks are the geotechnical ones which have safety
and operational risks/liabilities attached. Note that where
transport links to access the site are long, they may become even
more important, if movement of supplies and/or sold product are
impeded. Some risks may be unavoidable, but good information
and effective insurance policies can reduce their impact
(although risk is ultimately paid for so some risks are more
economically retained).
The risks associated with the estimation of Resources/
Reserves (and liability for directors and Competent Persons see
Phillips, 2000) are considerable in their own right and have been
the subject of separate conferences in the past and in AusIMM
Monograph 238.
The following outline of the main risks inherent in a mineral
project valuation and review illustrates the types of relevant risks.
The authors consultancy, Minval Associates Pty Limited
(MINVAL), specifically brings them to the attention of its clients
so that they are well aware of the risks and uncertainties involved.
Firstly, the overall economic factors are considered as part of
general the risk in any MINVAL valuation or review. Such things
as inflation, currency fluctuations, interest rates, sharemarket
sentiment, political or industrial unrest, all may affect the ability
of a party to fund its share of a project or to raise additional
working capital (either as debt or equity). Of greater importance
is the constant lack of certainty about the future actions of
government (at local, State and Federal levels) whatever the
country involved. Next there are unexpected policy or regulatory
changes, involving such things as land access and any native title
process required to be observed (considering both the time delay
and the cost); increased environmental controls and standards;
changes to taxation, charges and royalties; and changes to fiscal
and monetary policy.
Thus, profitable returns from any conceptual mining
development scenario postulated cannot be guaranteed. When
coming to its conclusions, MINVAL has to rely upon
assumptions based on its own knowledge and on the content of
the clients technical database, as well as upon assumptions
about future events at that time. These especially concern
assumed commodity prices (commonly converted to the local
currency); the validity of the Reserves/Resources estimates, both
in tonnage/quality and spatial distribution; and the ability to
technically and economically mine and then metallurgically treat
the ore to obtain the recovery expected at the desired rate.
Although MINVAL generally believes its major assumptions
in its conceptual scenarios are reasonably based at the specific
time it prepares its valuations or reviews, the continued technical
success of any mineral project may be subject to future variations
beyond anyones control, as further exploration results are
obtained and development takes place at the various mining sites.
This is particularly the case if lower grades or established by
further drilling such that realisation of the assumed recovered
8. AusIMM Monograph 23, Mineral Resource and Ore Reserve
Estimation The AusIMM Guide to Good Practice (Ed: A C
Edwards), 2001. It looks at numerous risk areas from the collection
and treatment of sample data, integrity of the database and the
geostatistical model/modelling parameters used.
9. Even here, there are a subset of liability risks associated with the use
of non-genuine spare parts for such equipment from both a safety
and operational cost/performance view (warranties may be breached
and insurances compromised, too).
10. Minerals industry professionals encompasses an increasingly diverse
group of graduates, but at its core are the geologists, mining engineers,
metallurgists and environmental scientists.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
grade over the mines life is unlikely; or if higher contents of
deleterious impurities are found; or more dilution and mining
loss than predicted; or if mill recoveries are likely to be less than
expected after further metallurgical testing.
The ability to construct the project on time without cost
over-runs and/or to meet annual production targets
(quality/quantity) within the assumed operational time frame for
the mining site are also critical risk factors eg, due to equipment
availability and poor productivity9, lost time accidents, inclement
weather and natural disaster impacts, unexpectedly changed
geotechnical conditions (pit slopes or strip ratios). Associated
with this concept is that there are no unexpected delays or timing
changes of the predicted cash flows, nor adverse impacts on
them, eg changes to estimates and timing of operating and capital
(initial and sustaining) costs.
Ability to contain total operating and capital costs, the
estimated exploration and development expenditures and
replacement capital estimates at the levels necessary to justify
exploitation of a resource asset.
Protection of company databases is often considered in the
wider arena of document management and computer backup
regimes. In the context of project procurement and management
of contractors/consultants, attention is also specifically given to
managing this area of risk. Wood (2000) addresses this area in
detail from a legal viewpoint, since maintenance of good filing
allows timely retrieval of important information and documents
if disputes or disasters occur and maximises the ability to be a
defendant or claimant.
Information management is a critical underlying risk that must
be considered paramount even if it is rather a soft asset,
compared with buildings and equipment. Treatment processes
and patented technology or intellectual property in general;
licences, leases, permits and even title to exploit the minerals
themselves are all often initially left out of a risk assessment as
their protection is simply assumed.
THE SUSTAINABILITY ENVELOPE AND CIVIL
SOCIETY
Mineral Industry professionals10 tend to be better prepared to
deal with the hard technical issues rather than the soft human
issues, although they are becoming increasingly aware of the
importance of soft skills in dealing with Civil Society as the
debate still rages over how to create a sustainable future which
accommodates both industrial development and the environment,
particularly the emotion-filled concern with greenhouse gases,
global warming, etc. Today, project financiers embrace better
management of the environmental and social impacts through the
Equator Principles (Fuzi and Bateman, 2003).
Irrespective of the lack of universality in the acceptance of the
Kyoto Protocol by Pacrim countries the climate change issue will
not go away. Developing countries may prefer to place a priority
on project development, but even in developed countries like
Australia, its national governments rational decision not to ratify
the Kyoto Protocol, has been challenged. Various state
government initiatives (influenced by green politics) now
require companies to reduce greenhouse gas emissions and
engage in sustainable resource practices to mitigate climate
change risk. In an address in late 2003 to the Total Environment
Centre, NSW Premier Bob Carr argued the need for a de facto
compliance with the Kyoto Protocol by the States, mentioning
the civil suits being brought in the US in relation to failures to
reduce greenhouse emissions resulting in climate change. He
proposed the establishment of a new Department of Energy,
Utilities and Sustainability and a NSW Greenhouse Office by
2004. This type of socio-political risk is extremely difficult to
address because it has a high emotional component, but it is
essential to do so.
47
M J LAWRENCE
Civil society has been active, too. For example, Climate
Action Network Australia (CANA) distributed letters also in late
2003, warning of the regulatory and litigation risks to which
companies may be exposed if they fail to address the
consequences of any activities undertaken by them which
contribute to greenhouse gas emissions.
Note that compliance with todays standards will be no
defence to future actions. As with corporate governance issues,
having an eye to the future is always prudent. Also, the loss of
corporate or personal reputations is become a risk area in itself
today particularly when wanting to start new projects in new
jurisdictions.
RISK MANAGEMENT STRATEGIES
Risk management is the treatment of risk to the best economic
advantage of the parties involved11; and it involves the cultures,
processes and structures that are directed towards the optimal
potential opportunities and adverse effects (see AS/NZS 4360).
Risks are usually retained or transferred, too.
Any risk management strategy starts with identifying what are
all the reasonably foreseeable risks, uncertainties and hazards
(though not neglecting the opportunities, however) in the
particular context. A real plan must start with demonstrable
genuine enquiries about the nature any identified or perceived
risk (ask the right questions)12. They are then analysed to
ascertain the degree to which they can be practically addressed
and the degree which it is reasonable to expect to exercise
control over them; as well as assessing their consequences.
Implicit in this objective review is a cost/risk assessment that
looks at the likelihood of an event occurring (on the balance of
probabilities or frequency) and its expected severity and
consequential damage to the activity13. The envisaged result is
then balanced against the estimated cost of taking precautions
against the likelihood of the event occurring in terms of effort
(time and money) and effect on the ultimate purpose of the
activity (evaluate). Risks must be monitored as they can be
volatile and time dependent.
The experience of Solid Energy New Zealand Ltd is outlined
in some detail in Maslin (2003) and it warrants consideration for
its ideas on overall integrated risk management systems.
Shillabeer and Gypton (2003) particularly address the
management of technical risk (costs and operational
performance) and the specific questions to be asked in a review
of a Feasibility Study14 (also see McCarthy, 2003). Most serious
errors are concealed by a plausible context so use common
sense.
It is critical today to clearly identify the appropriate lines of
responsibility for various defined risks, since there are criminal
as well as civil consequences if many risky events eventuate.
There may be personal (a familiar theme of this paper) as well as
corporate consequences. Always seek independent peer review
regularly monitor.
All risk could be eliminated at infinite cost, so most are
mitigated because of realistic cost constraints. The Courts (the
ultimate arbiter) claim that only reasonable cost/benefit criteria
need apply, but!
11. Whilst this may be generally true, it is the reality that not everyone is
driven by simple economic advantage that complicates the risk
management process.
12. Put colloquially, dont just check that the goal posts are in the right
place, but ascertain how realistic it is for them to be kicked!
Continually delve and document evidence.
13. This is sometimes referred to as probability/impact analysis to
determine overall risk try to understand it.
14. Between 1980 and 2001 they found that cost over-runs averaged
22 per cent (within a wide range from -43 per cent to +50 per cent).
48 Adelaide, SA, 19 - 22 September 2004
Finally, to solve any problem requires updated, reliable input
data to the strategy development process and good people.
To get a firm handle on risk, quantitative rather
than purely qualitative risk analysis is needed.
Good analysis can only come from comprehensive
databases which are not as yet available. but
should be (personal communication, Quentin
Amos, HSBC Precious Metals [Australia] Ltd
see his paper also given at this Pacrim 04
Conference).
Such databases must be collected, preserved and protected
(sometimes insurance may not be enough for critical
documentation like lease documents and various agreements).
Also, the lawyers will testify to the importance of having a
reliable documentary trail if disputes or regulatory legal action
commences. Again, the words and definitions used in these
databases should have commonly accepted meanings to mitigate
risk.
INDIVIDUAL LIABILITY
Many participants in mineral developments are subject the legal
requirement to broadly act with reasonable care and diligence in
respect of their contribution to a companys activities. This
objectively tested duty of care arises out of common law, equity,
and statute. Ignorance and failure to enquire are no protection.
The quality/quantity of disclosure about a project involves a
chain of parties, all of whom must properly play their part for the
system to work and for investment risk to be minimised. The
participants in the chain, who are responsible for disclosing
and/or exercising varying degrees of due diligence when
confronted by incomplete, confusing or contradictory data and
information are:
exploration and mining companies (directors, officers and
employees, as well as technical and legal consultants and
contractors), who are responsible for the timeliness,
thoroughness and accuracy of the material data and
information disclosed orally or in written form (in press
releases, on the Internet, or in annual and quarterly reports)
so it is easily understandable, complete, accurate and true
and it is not misleading in any way;
promoters, brokers, analysts and investment advisers, who
are generally responsible for authenticating, interpreting and
disseminating this data and information to investors in the
public arena according to its suitability; and
investors themselves (whether a mutual fund/financial
institution or private investor), who must take some
responsibility for their investment decision, based on their
own financial circumstances.
According to Sheppard (2002):
Implementing formal risk management and
compliance programs taking positive steps to
limit exposures is one half of the protection
equation for companies and directors. Having
those programs in place, there are two further
steps that companies and directors can take to
maximise the protection available to them in the
event of allegations of wrongdoing; or a more
general investigation by regulatory authorities
to balance what are almost inevitably competing
interests.
She went on to explain that they involved Access and
Indemnity Deeds governing the directors access to documents
and records of the company both during and after the period of
their directorship; and the provision of an indemnity in respect of
costs or other expenses or losses incurred as a consequence of
having been a director.
PACRIM 2004
 OVERVIEW OF MINERAL PROJECT RISK ISSUES AND ROLE OF MINERAL INDUSTRY PROFESSIONALS
Such deeds often impose an obligation on the company to
arrange directors and officers liability insurance (and their
renewal, but it should be quite clear that directors will be
appraised in a timely way of the success (or otherwise) of
arranging this insurance cover, so that directors can reasonably
decide what action to take. This may involve them making their
own insurance arrangements or legally re-arranging their affairs
to limit liability.
Consultants and contractors should also hold appropriate
Professional Indemnity (PI) insurance and this subject was well
covered in the following papers: Lawrence (2002b);
Williamson-Noble and Lawrence (1994).
The trust-us-well-look-after-you approach of many large
organisations is unsatisfactory for todays professional because
parties expectations can differ substantially when problems
arise. Instead ensure appropriate detailed documentation exists
that outlines the personal risks and how they are to be mitigated.
It is sound practical advice to obtain independent legal advice on
the adequacy of such critical documents.
RELEVANCE OF DISPUTE RESOLUTION
SOLUTIONS
It may seem self-evident that a core consideration of most risk
management strategies is how to effectively and efficiently
resolve identified problems, but many fail to incorporate
adequate dispute resolution solutions into appropriate
documentation from the start. There is always traditional
litigation to rely upon, but the increase in the complex technical
component of disputes in the minerals industry (which judges
may have difficulty in fully appreciating) make it imperative that
alternative dispute resolution (ADR) model options are already
built-in: mediation, expert determination or arbitration. It is
widely accepted that ADR options are generally speedier,
cheaper and more confidential alternatives to litigation, when
they can be used.
At the heart of any ADR solution is the early inclusion of
appropriate contractual clauses that pre-plan and synchronise
between the parties involved (owners/lenders/contractors)
rational ADR mechanisms as options to the ultimate litigation
sanction; how to appoint the dispute resolver; and which rules
are to apply (especially in an international context). Its main
effect is to contribute to the reduction in uncertainty.
Professional bodies, like The Australasian Institute of Mining
and Metallurgy (AusIMM) and the Mineral Industry Consultants
Association of Australia (MICA) have implemented ADR
schemes that focus on providing their pool of highly skilled and
experienced, accredited technical members for that purpose. It is
an attempt to move the resolution of technically-based disputes
back where they belong in the technical arena away from the
increasing (and often unnecessary) involvement of lawyers.
Lawyers are essential to properly document the availability of
such solutions so that they are enforceable, but that should be
their sole role.
15. JORC (1999). Australasian Code for Reporting of Mineral Resources
and Ore Reserves (JORC Code), issued by the Joint Ore Reserves
Committee (JORC), comprising AusIMM, AIG and Minerals
Council of Australia, September, 16 p (The Australasian Institute of
Mining and Metallurgy: Melbourne).
16. VALMIN Code (1998). Code and Guidelines for Technical
Assessment and/or Valuation of Mineral and Petroleum Assets and
Mineral and Petroleum Securities for Independent Expert Reports,
February, 23 p (The Australasian Institute of Mining and Metallurgy:
Melbourne); incorporating Lawrence (1998). The revised VALMIN
Code and Guidelines (1998): an Aide Memoire to assist its
interpretation, The AusIMM Bulletin, 3:80-83.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Bremen (2004) recently addressed this issue of enforceability
in disputes involving international mining projects, particularly
in the developing world with each country having potentially
different legal systems. He outlined such matters as jurisdictional
issues and the need for symmetry in the contracts between the
various parties involved; whether to litigate or arbitrate and how
to ensure an equitable and enforceable result. One of the more
useful truisms that he outlined, since it has wider application to
many other risk management issues addressed in this paper, is
that it is unwise to simply take dispute resolution clauses off the
rack or adopt ones that were used in another mining projects
documentation. Expedient and apparent cost-saving solutions
often come back later with a vengeance because mineral deposits
are unique, due to their nature and location.
AusIMM and MICA, for example, also provides assistance to
individual members in the context of facilitation of dealings and
disputes between its members; between its members and
employers/clients; and between its members and the wider
community. For further details see Lawrence and Dunlop (2004)
and Lawrence (2002b).
ROLE OF INDEPENDENT CONSULTANTS
The importance of having regular independent third party
overviews of a project should be obvious, despite the apparent
initial cost. Lawrence (1997) thoroughly explained this aspect in
respect of the Busang/Bre-X scam (see below). However, an
extremely informative paper by Phillips (2000) outlines the
obligations and liability undertaken by consultants and others
who prepare key public documents (she uses Resource/Reserve
Statements as an example).
VALUE OF ACCREDITATION REGIMES
Professionals working in the minerals industry are generally
expected to have gained appropriate tertiary technical
competencies and to always act with probity. Proof of Corporate
Membership of an appropriate reputable, internationally
recognised organisation (eg AusIMM or MICA), demonstrates
this to ones employer (or client), peers and the wider
community, including government and regulators.
Professional bodies, like MICA/AusIMM, protect their
reputation and ensure their credibility by both ensuring that the
qualifications and experience of their members are appropriate
for the expertise claimed (or membership level granted); and that
their members abide by professional ethical standards and
relevant best practice codes, guidelines and policies. AusIMM
introduced (on 1 January 1996) a system of accreditation or
registration to fulfil the need to recognise those with the required
level of professional skill. This has evolved into the Chartered
Professional (CP) Scheme. A key element of the CP Scheme
(and the maintenance of MICA accreditation) is that all members
must maintain the currency of their expertise by undertaking a
required minimum amount of Continuing Professional
Development (CPD) programs.
Lawrence (1995) raised the issue of personal liability and the
value of defences relying upon being accredited by professional
bodies and following industry best practice codes (the Due
Diligence Defence). Lawrence (1999) addressed the question
Why belong to The AusIMM, by attempting to explain the
relevance of membership to the global minerals industry. Both
papers stressed that risk and uncertainty could be mitigated by
using competent (qualified and experienced) and reputable
professionals. If their membership was endorsed in additional
ways (eg AusIMMs Chartered Professional Status; or by being
an accredited MICA consultant) this provided some enhanced
protection.
49
M J LAWRENCE
Further, given the importance of reliable Resources/Reserves
estimates to project investment decisions, it is relevant to note
that the sample preparation and analysis should be performed by
laboratories that have relevant accreditation and use appropriate
quality control standards to mitigate this critical source of risk.
USEFULNESS OF CODES OF BEST PRACTICE
Note again Lawrence (1995) which dealt with the best practice
Codes defence for Pacrim practitioners; and Lawrence (1996) on
the value of the Due Diligence defence against personal liability
claims. The value of using best practice Codes, such as JORC
(1999)15 and VALMIN (1998)16, were dealt with by Lawrence
(1997). They were specifically explored in the context of the
infamous Busang gold fraud scandal in 1997 in Indonesia,
involving Bre-X Minerals Limited (Bre-X). It reminded market
participant yet again of the constant need for effective and truly
independent due diligence at all stages of a projects
development, since effective technical assessment is the
foundation of any projects feasibility/study or valuation and
subsequent financing. Using these Codes bolsters the use of the
Due Diligence Defence against a claim for breach of the duty of
care. They objectively document what a reasonable person of
ordinary caution would do in similar circumstances in that
commercial/technical setting.
Another value of best practice Codes is that they encourage the
use of nomenclature with globally accepted meanings that
itself mitigates risk. Effective communication is assisted when
words mean what they appear to say.
Such Codes need regular re-appraisal to ensure that they
continue to meet market expectations, but also to facilitate their
globalisation. The Reserves area is well advanced internationally
compared with an international valuation code through the
International Valuation Standards Committee (IVSC), see
McNamara (2001) and Lawrence (2002a). It has made less
progress due to the need to first obtain agreement on some issues
work with the accounting profession (International Accounting
Standards Board; IASB). See Lawrence (2001b).
CONCLUSIONS
It may be obvious to some, but good decisions require good input
data (that are also effectively communicated and which must be
retained and protected). Also, when allocating risk management
responsibility that party must have the authority to control or
influence that risk or obligation.
Technical professionals tend not to consider the soft issues
first. The emphasis is on project hardware and the better
understood technical, operating/capital cost and regulatory risks.
Less obvious issues impacting on project viability and/or
investment attractiveness are often relegated or forgotten.
Examples include the impacts of construction delays and slow
production ramp-up (eg due to unexpected local opposition,
procurement, human resource and contract management
problems); poorly understood and documented future mine
closure provisions (site clean-up/environmental remediation);
and those emotionally based problems that money just cant fix.
Any risk that can hinder debt repayment is of major concern to
project financiers.
There is too often little real cultural awareness of the
environment in which the project or will operate. Even the
pivotal and persuasive influence of Civil Society in creating a
sustainable global minerals industry is a relatively new and not
well understood, soft-risk concept with which project personnel
must effectively deal. Project participants have to do more than
just attend courses.
The minerals industry, whether here or abroad, is governed by
a complex regulatory and legal system that constantly changes,
with companies (and increasingly individuals) being held liable
50 Adelaide, SA, 19 - 22 September 2004
for any breaches. These breaches, whether direct or indirect, may
result in civil or even criminal proceedings that can place
personal assets at risk and can involve imprisonment. Owners,
company directors, managers and employees, together with
consultants, contractors and suppliers of equipment or
consumables are all at risk for decisions to act (or failure to act)
in managing operations or carrying out directions, as well as for
the actions of others under their control. Disputes over alleged
breaches may be dealt with in many ways, but adversarial
litigation is unfortunately the ultimate solution. General self
protection is usually a secondary consideration, until too late.
Hence, in an increasingly litigious world, it is prudent to seek
ways to manage not only the technical and financial risks
themselves that are inherent in any minerals industry business
endeavour (especially one in a Pacrim country) but to ensure
effective conflict management options are available, particularly
if it is likely to be personally directed.
The critical step in dealing with disputes is that one must be
able to avail oneself of the Due Diligence Defence. That is,
demonstrate (with hard evidence) in respect of any task
undertaken by a professional that all that could be done had been
done to avoid the problems that has arisen. More specifically,
that it can be reasonably claimed that any defect was caused by a
reasonable mistake; or resulted from reasonable reliance on
information supplied by another person, (preferably after
independent verification); or was caused by the act or error of
another person beyond ones control.
If personal professional liability is seen a priority, then the
myriad material risks and uncertainties inherent in exploration
and mining activities throughout the world are also likely to be
effectively managed and minimised.
But remember, little risk means little reward!
REFERENCES
Bremen, J, 2004. Dispute resolution in international mining projects: the
importance of being enforceable, E&MJ, June:24-26.
Edwards, G and Bezzina, M, 2003. Risk management a Standards
approach, in Proceedings Mining Risk Management, pp 421-424
(The Australasian Institute of Mining and Metallurgy: Melbourne).
Fuzi, G and Bateman, B, 2003. Project finance goes green (Equator
Principles), Clayton Utz Banking and Financial Services Insights
Newsletter, 29 October, 3 p.
IVSC, 2001. International Valuation Standards 2001, 1 July, 458 p
(IVSC: London).
Lawrence, M J, 1995. Personal liability Chartered Practising Status and
best practice defences for Pacrim practitioners, in Proceedings
Pacrim Congress 1995, pp 317-322 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
Lawrence, M J, 1996. The due diligence defence against personal
liability, in AusIMM Yearbook 1996/97, pp 64-69 (The Australasian
Institute of Mining and Metallurgy/Executive Media: Melbourne).
Lawrence, M J, 1999. Why belong to The AusIMM? Its relevance to the
global minerals industry, in Proceedings 32nd Annual Conference of
New Zealand Branch, Waitangi, pp 226-234 (The Australasian
Institute of Mining and Metallurgy: Auckland). Also Presidential
Address to AusIMM North Queensland Branch, Townsville, 8 p,
18 August, 1999; and AusIMM Western Tasmania Branch, Burnie,
21 September.
Lawrence, M J, 2000. Ethics, expert witnesses and professional best
practice codes an AusIMM perspective, in Proceedings Mining
Indonesia 00, The Mining Challenges into the Third Millennium,
Jakarta, 4 - 7 April, 15 p (IMA: Jakarta).
Lawrence, M J, 2001a. The transition from university to industry what
your mother and lecturer never told you, presented at AusIMM Youth
Congress 2001, 8 p (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Lawrence, M J, 2001b. International accounting and valuation standards:
a threat or an opportunity for AusIMMs VALMIN Code? AusIMM
Annual Review 2001-2002, pp 49-55 (The Australasian Institute of
Mining and Metallurgy/Executive Media: Melbourne).
PACRIM 2004
 OVERVIEW OF MINERAL PROJECT RISK ISSUES AND ROLE OF MINERAL INDUSTRY PROFESSIONALS
Lawrence, M J, 2002a. The Australasian VALMIN Code (1998): its
relevance to South African Valuation Code Initiatives, in
Proceedings Colloquium Valuation of Mineral Projects and
Properties: An African Perspective, Randburg, South Africa, 19 - 20
March, 12 p (South African Institute of Mining and Metallurgy:
Johannesburg).
Lawrence, M J, 2002b. ADR initiatives for MICA/AusIMM consulting
professionals, MICA Professional Indemnity and Liability Seminar,
Perth, 22 September, 7 p.
Lawrence, M J and Dunlop, J S F, 2003. Alternative Dispute Resolution
(ADR) Taskforce update, The AusIMM Bulletin, May/June:48-50.
Maslin, W J, 2003. Risk management practices at Solid Energy New
Zealand Ltd, in Proceedings Mining Risk Management, pp 136-146
(The Australasian Institute of Mining and Metallurgy: Melbourne).
McCarthy, P, 2003. Managing technical risk for mine feasibility studies,
in Proceedings Mining Risk Management, pp 21-27 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
McNamara, G, 2001. International valuation standards, in Mineral Asset
Valuation Issues for the Next Millennium 2001 (VALMIN 01), pp
51-54 (Australasian Institute of Mining and Metallurgy: Melbourne).
Phillips, R, 2000. The liability of company directors and competent
persons for resource/reserve disclosure, in Proceedings The Codes
Forum: The VALMIN and JORC Codes, After 2000: What Future for
Mining? MICA Seminar, 9 p (MICA: Sydney).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Sandman, P M, 2003. Responding to Community Outrage: Strategies for
Effective Communication (American Industrial Hygiene Association:
Fairfax, Vermont, USA).
Sheppard, S, 2002. Protecting directors: practical risk management, 2002
Practical Risk Management Seminar, (Clayton Utz Publications:
Sydney), 9 p.
Shillabeer, J and Gypton, C, 2003. Highlighting project risk following
completion of the feasibility study, in Proceedings Mining Risk
Management, pp 101-109 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Standards Australia, 1999. Australian/New Zealand Risk Management
Standard 4360 (Standards Australia International Ltd: Sydney).
Williamson-Noble, J C and Lawrence, M J, 1994. Individual personal
liability risk management in open pit mining, in Proceedings Fourth
Large Open Pit Mining Conference, pp 273-279 (The Australasian
Institute of Mining and Metallurgy: Perth).
Wood, J, 2000. Information management, implementing systems to
support innovation and risk management, Construction IT 2000,
Clayton Utz (Solicitors) Publications, 10-11 April, 10 p.
51
The Many Dimensions of Sustainability
I Lowe1
ABSTRACT
The scale and seriousness of Australias social and environmental
problems are no longer in doubt. We are consuming the opportunities of
future Australian generations by our lifestyle choices. And at the global
level, human activities are altering the very systems on which our species
depends. Doing nothing about such problems as climate change is not an
option. It is possible to move to a sustainable future but it will require
fundamental changes to our values and national social institutions. We
must recognise that we share a common future with the entire human
family and the other species with which we share the planet. It is not just
our humanitarian duty to improve the lot of the poorest people of the
world. It is also enlightened self-interest because a world of increasing
inequality will be a world of increasingly destructive tension.
INTRODUCTION
The evidence is clear. If civilisation is to survive, the next
century will have to be a time of transformation, not just in
technological capacity but also in our approach to the natural
world, and to each other. The second report in the United Nations
Environment Program (UNEP) series on the global
environmental outlook, GEO2000, said: The present approach is
not sustainable. Doing nothing is no longer an option. A
sustainable society would not be eroding its resource base,
causing serious environmental damage or producing
unacceptable social problems. It is clear that our present lifestyle
does not satisfy any of those main criteria. We are dissipating
resources future generations will need, damaging environmental
systems, and reducing social stability by widening the gap
between rich and poor. It is possible to move to a sustainable
future, but it will require fundamental changes to our values and
social institutions. While UNEP said that doing nothing about
the huge problems we face is not an option, it remains the most
common response of todays decision-makers.
RESOURCES
A sustainable society will not erode its resource base. Our most
serious medium-term resource problem is oil. Our entire
transport system is based around petroleum fuels which are
cheaper than any other liquid except tap water. While optimists
think the peak of world oil production may be fifteen years away,
pessimists now think it was in the year 2000. It is undeniable that
the age of plentiful cheap petroleum fuels will end soon. There
are technical alternatives, most obviously fuel cells using
hydrogen produced by splitting water with solar electricity, but
they are sure to be much more expensive. We should be planning
now for the post-petroleum age as the forward-looking oil
companies are.
There are other resource issues. Humans now use about half of
all the worlds available fresh water directly or indirectly. Our
forests, our fisheries, our agricultural soils and our grazing lands
have not been used sustainably. We have to reduce resource use
so we are using the income of natural systems; our present
approach is running down the capital. We must also improve our
scientific knowledge so we know sustainable use levels. We may
move to a system of tradable water rights, as this is an
economically efficient way to use a fixed amount of water. That
1. Emeritus Professor of Science, Technology and Society, Griffith
University, School of Science, Nathan Qld 4111.
E-mail: I.Lowe@griffith.edu.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
approach is morally justifiable only if we know that the overall
extraction rate is ecologically sustainable; otherwise we are
producing short-term economic benefits at the expense of future
generations.
ECONOMIC ISSUES
Two hundred years ago, the economic fad was global empires to
obtain cheap resources and open up markets. A hundred years
ago, the fashion was protection within each nation. Fifty years
ago, it was public funding of infrastructure projects by
government. Todays fad is a return to eighteenth century
economic empires, once again allowing free rein to global
movement of capital.
In local terms, Australia is already what Barry Jones called a
post-industrial economy. The traditional sectors of agriculture,
mining and manufacturing now account for less than a quarter of
the economy and fewer than 20 per cent of jobs. Most of us now
work in the services sector. Our relatively high wages and
salaries make our production costs high in labour-intensive fields
like textiles, clothing and footwear. Our low level of investment
in research and innovation makes us uncompetitive in
value-added manufactures. A sustainable economic future must
be based on reward structures that direct our limited resources
into areas in which we have some prospect of being competitive.
The present system might have been designed to ensure we
squander our resources.
SOCIAL COHESION
Fifty years ago, Australia was one of the most equal societies in
the world. The legal framework of the basic wage set a minimum
standard, while the highest salaries were only five to ten times
that minimum. A strong commitment to public education gave
opportunities to bright children from a range of disadvantaged
backgrounds. Most people used public transport to get around,
while the health system made provision on the basis of need.
Today, we are one of the most unequal of all the industrialised
nations. Some executives are given salary packages hundreds of
times the average household income, often with arrangements for
tax avoidance built in. Public health care and education have
been steadily run down, and Commonwealth ministers now
openly tell those who are miss out on hospital beds or university
places that they should be paying for private options. This is
cruel advice to the increasing fraction of the adult workforce that
is unemployed, under-employed or doing poorly paid casual
work.
Many Australian households now have great difficulty
affording the basic needs of food, power and clothing.
Low-income families do not have the luxury of buying
educational opportunities or privileged health care. While the
links are not simple, the widening divisions between rich and
poor are inevitably causing resentment and social tensions. If we
continue to ration life opportunities according to income, we
must develop policies that systematically reduce the inequality of
household incomes. This is a wise investment in future social
cohesion. Children growing up anywhere in Australia should
have the opportunity to realise their potential in all ways
physically, intellectually and emotionally.
We should also be responsible global citizens. It is not just our
humanitarian duty to try to improve the lot of the poorest people
of the world. It is also enlightened self-interest because a world
53
I LOWE
of increasing inequality will be a world of increasing tension. We
cannot be secure doing property deals on mobile telephones from
air-conditioned four-wheel-drive vehicles in a world where the
majority of people have never ridden in a car, made a telephone
call or owned any property. We should be much more generous
with aid, lifting our commitment from its present 0.26 per cent of
GDP to the international target of 0.7 per cent, and forgive the
debts of very poor countries. As the Australian delegation said at
the 1999 UNESCO World Conference on Science, we should
aim to make this not just a new century but a just new century.
ENVIRONMENTAL ISSUES
The scale and seriousness of environmental problems are no
longer in doubt. At the national level, two State of the
Environment reports have now been published. The 1996 report
showed that we have a beautiful and unique environment, with
many aspects in good condition by international standards, but
we also have very serious problems, most obviously loss of
biological diversity, degradation of inland waterways and
destruction of the productive capacity of rural land. Its final
section linked the environmental problems to lifestyle choices,
showing that a sustainable future will require integrating
environmental awareness into all social and economic decisions.
The second report, released in 2002, noted an improvement in
urban air quality but found that all the other critical
environmental problems are getting worse, because the pressures
on natural systems are still increasing. Each year the Australian
population grows by about 200 000, as the natural increase (the
excess of births over deaths) of about 120 000 is augmented by
migration. The material expectations of people also increase each
year; we use more energy, travel further in larger and less
efficient cars, live in larger houses, consume more resources and
produce more waste. The compounding effect of more people,
each on average demanding more, is putting ever-increasing
pressure on our natural systems.
Global studies draw the same conclusion. UNEP has now
produced three reports in its Global Environmental Outlook
series. They show some successes, such as the concerted
international effort to stop releasing the chemicals that deplete
the ozone layer and encouraging reductions in many countries
of urban air pollution. They also document what the third report
called environmental challenges increasing emissions of
greenhouse gases, over-exploitation of water, 1200 million
people without clean drinking water and twice that number
without sanitation, species being lost at an increasing rate,
fisheries in decline, land degradation, and problems caused by
increasing release of nitrogen into natural systems.
A new report by the International Geosphere-Biosphere
Programme, Global Change and the Earth System: A Planet
Under Pressure, paints a disturbing picture. It says human
activities are affecting global systems in complex, interactive
and apparently accelerating ways, so that we now have the
capacity to alter those natural systems in ways that threaten the
very processes and componentson which the human species
depends.
The most urgent issue is climate change. The politicians of the
developed world accepted the scientific arguments when they
negotiated the Kyoto agreement to slow down release of
greenhouse gases. Only the Bush regime in Washington and the
Howard government in Canberra have refused to ratify this
treaty. The Earth as a whole has warmed about 0.6 degrees
Centigrade in the last hundred years, with Australia warming
slightly more than the global average. The Earth is now warmer
than at any time since credible records began. As predicted by
climate scientists, there have been other changes associated with
the warming: shrinking of glaciers, thinning of polar ice, rising
54 Adelaide, SA, 19 - 22 September 2004
sea levels, changing rainfall patterns and more frequent extreme
events like droughts and severe storms.
Just as other serious environmental problems have been
tackled at the international level, we need a concerted global
response to climate change, not a head in the sand approach of
denial. We should set a target of substantially reducing fossil fuel
use to 40 per cent of present levels by 2050. Cutting our
emissions is our obligation to the world community. It will
require policy measures from all levels of government to
encourage cleaner energy supply and much more efficient
conversion of energy into the services we need. The obvious way
to fund the transition is to systematically phase out the huge
current subsidies of fossil fuel supply and use (NIEIR, 1996),
transferring those public funds to the expansion of renewable
energy supply technologies and efficiency gains.
HUMAN HEALTH AND WELL-BEING
We should recognise the link between healthy ecological systems
and human health. At its most basic, we rely on natural systems
to provide the essentials of life: oxygen, water and food. We also
need those systems to process our wastes. Some of the systemic
features that contribute to poor health result directly from the
degradation of the natural worlds capacity to provide those
essential services. More broadly, natural systems provide our
sense of place, our cultural identity and spiritual sustenance.
Again, we are healthier and more fulfilled when those needs are
satisfied. An investment in the health of our natural systems is
also an investment in the health of the community.
A SUSTAINABLE FUTURE?
In Resetting the Compass, Yencken and Wilkinson suggest a
guide for Australias Journey Towards Sustainability. Existing
policies will not achieve a transition to sustainability because
they do not address the growing pressures of increasing
population and rising material demands per person. So, they
conclude, we need to aim to stabilise the population as well as
being committed to dematerialisation. A German study argued
that Europe needs to reduce energy use by a factor of four and
materials use by a factor of ten. Several European nations have
now adopted those targets. More generally, we must recognise
that sustainability has many dimensions. The vision of a
sustainable future involves sustainable resource use, a
commitment to maintaining the ecological values of natural
systems, development of social cohesion, nurturing our cultural
traditions and finding durable economic activities. Balancing
those dimensions is a complex task that defies simple
approaches, like trusting the market or trying to keep everything
the way it is today. We have to take difficult decisions about
which things must change and which should be preserved. This
will only be possible if there is an open and transparent process,
involving the community and allowing time to work through the
costs and benefits of alternatives. Changing one thing in a
complex system always produces other changes, so no change is
ever universally beneficial; there are always losers as well as
winners. In a fair world, those who lose out from a change that
benefits the community as a whole should be compensated by the
rest of the community. We accepted that principle when we
decided that those using Sydney airport should pay a noise levy
to compensate those under the flight path.
Moving toward a sustainable future will require new
institutions. Concerted action is frustrated by division of
responsibility between different levels of government, by State
and local boundaries that have no social or ecological logic, and
by the silos of established departments. We need new structures
that will enable coordinated policies and actions, integrating
environmental goals with our social, economic and cultural
aspirations.
PACRIM 2004

CONCLUSION
This vision is utopian, but that has been said of all important
reform movements. Those who opposed slavery 200 years ago
were told that no economy could function without slave labour,
while the suffragettes were persecuted when they sought the vote
for women a hundred years ago. Closer to our time, 20 years ago
it was still utopian to be dreaming of Berlin without the Wall, or
South Africa without apartheid. Most social reforms we now take
for granted were initially denounced as utopian. They happened
because determined people worked for a better world. All around
the globe, people are striving to develop social and institutional
responses that will enable the transition to a sustainable future.
We have to change our values, recognising that we share the
Earth with all other species and hold it in trust for all future
generations. That is our moral duty.
REFERENCES
Australian Bureau of Statistics, 2002. Measuring Australias Progress,
ABS Cat No 1370.0 (Commonwealth of Australia: Canberra).
Council of Australian Governments, 1992. National Strategy for
Ecologically Sustainable Development (Commonwealth of Australia:
Canberra).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
THE MANY DIMENSIONS OF SUSTAINABILITY
Flannery, T, 1994. The Future Eaters, 423 p (Reed Books: Chatswood).
Global Scenarios Group, 2002. Great Transition (Stockholm
Environment Institute: Boston).
Hamilton, C, 2003. The Growth Fetish, 280 p (Allen and Unwin:
Sydney).
National Institute of Economic and Industry Research, 1996. Subsidies to
the use of Natural Resources, Environmental Economics Research
Paper No 2, Department of Environment, Sport and Territories,
Canberra.
State of the Environment Advisory Council, 1996. State of the
Environment Australia 1996 (CSIRO Publishing: Collingwood).
State of the Environment Advisory Council, 2001. Australia State of the
Environment 2001 (CSIRO Publishing: Collingwood).
Steffen, W, Jager, J, Matson, P, Moore, B, Oldfield, F, Richardson, K,
Sanderson, A, Schnellnhuber, H J, Turner, B L, Tyson, P D and
Wasson, R J, 2004. Global Change and the Earth System: A Planet
Under Pressure, 336 p (Springer-Verlag: Berlin).
United Nations Environment Program, 1999. Global Environmental
Outlook 2000 (Earthscan: London).
United Nations Environment Program, 2002. Global Environmental
Outlook 3, (Earthscan: London).
World Commission on Environment and Development, 1987. Our
Common Future (Oxford University Press: Oxford).
Yencken, D and Wilkinson, D, 2000. Resetting the Compass, 416 p
(CSIRO Publishing: Collingwood).
55
New Mining Projects Expectations and Outcomes
P L McCarthy1
ABSTRACT
When preparing an appraisal, valuation or feasibility study for a new
mining project it is difficult to imagine the technical, economic, social
and environmental context that the project will occupy in ten or 15 years
time.
This paper is about our ability to foresee the future, the choices
available now in matching the project to that future context, and about the
flexibility that exists in adapting the project to a changing environment.
The author draws upon his experience of project reviews, valuations
and feasibility studies in the mid to late 1980s in Australia and New
Zealand to explore how the project context has changed since key
decisions were made and what this tells us about decisions made today
for future projects.
INTRODUCTION
When preparing an appraisal, valuation, environmental impact
assessment or feasibility study for a new mining project it is
difficult to imagine the technical, economic, social and
environmental context that the project will occupy in ten or 15
years time. This paper is about our ability to foresee the future,
the choices available now in matching the project to that future
expectation and the flexibility that exists in the project, which
will enable it to be adapted to a changing environment.
Fundamental to our understanding of a new project is an
appreciation of its geological context. Whatever the initial
mineral resource and ore reserve position might be, we need to
assess whether, and to what extent, it is likely grow in the future.
We may also come to view the deposit in a different way, as at
Mount Charlotte, where a multi-vein underground operation
became a long-life, large-scale cave-and-fill mine. Or an open pit
may have potential to become an underground mine, although we
know nothing yet about the mineralisation below the proposed
pit floor. Some projects fail to deliver the scenario portrayed,
some run to plan, but others grow beyond the most optimistic
scenarios of the planners. To some regulators, non-government
organisations (NGOs) and community members this
above-plan performance, expressed as size or project life, may be
a worst nightmare scenario.
In the face of uncertainty about the scale and duration of the
project, development of a detailed closure plan at the time of
feasibility can be a futile exercise. The requirement should rather
be to demonstrate that the projects proponents are competent
and are committed to long-term stewardship of the property and
have the financial resources to deliver the required stakeholder
outcomes regardless of what surprises the orebody, technology or
market conditions might deliver.
REVIEW OF PROJECT OUTCOMES
In order to provide some quantitative support for the views
expressed in this paper, the author reviewed the outcomes of 56
Australasian gold mining projects for which prefeasibility or
feasibility study results, or confident descriptions of project
development, were announced during 1988 - 1989. This period
was chosen so that a 15-year post-study history was available.
The period fell between the October 1987 stock market crash and
1. FAusIMM, Managing Director, Australian Mining Consultants Pty
Ltd, 19/114 William Street, Melbourne Vic 3000.
E-mail: pmccarthy@amcconsultants.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
the introduction of a tax on gold profits in 1992. It should be
noted that there may have been some bias in the quality of
projects at the time, as some owners tried to bring production in
before the gold tax. Information was drawn from the authors
files, from the Register of Australian Mining (Resource
Information Unit) and the Gold Gazette (Resource Information
Unit).
Of the projects reviewed, 14 (25 per cent) failed to develop as
planned. Some, such as Bottle Creek, Disraeli, Horn Island,
Rothsay, Second Fortune, Timbarra and White Range failed
within months, or at best in less than two years, largely because
of an inadequate geological understanding of the deposit at the
time of the feasibility study. Some of these also had difficulty
with their environmental performance. Golden Cross failed
because of a geotechnical problem under a waste storage facility,
although the potential for this problem had been flagged before
the mine was constructed. Mineral Hill closed due to
metallurgical difficulties but was later successfully reopened.
Other projects such as Big Bell and Mount Todd open pits
operated for some time but suffered excessive costs due to rock
hardness that led to premature closure. Some projects were
announced but still have not been commissioned, notably
involving re-opening of underground gold mines in eastern
Australia, suggesting that the realities of developing these
projects were poorly understood at the time.
The failed projects often involved lengthy litigation and,
sometimes, public recrimination and usually led to
administration or receivership. Affected stakeholders in such
situations included investors, lenders, employees, local
communities, traditional landowners, governments and the
environment.
At the other end of the scale, 11 (20 per cent) of the gold
mining projects reviewed pleasantly surprised their owners by
doing substantially better than forecast, as follows:
Boddington was expanded from a 3 Mtpa mine producing
around 150 000 oz pa to become one of Australias largest
producers at over 400 000 oz pa and lasted 15 years. It has the
potential, by exploiting the deeper Wandoo mineralisation, of
returning to a leading position in the future.
The Granites was envisaged as a 0.6 Mtpa operation producing
90 000 oz pa. It had grown to a 210 000 oz producer by 1995
and continues today, mining the Callie deposit.
Hill 50 was to produce about 50 000 oz pa but by final
feasibility, 1.2 Mtpa and 100 000 oz pa was expected. The
project evolved through various configurations, delivering
substantially more gold, and is still in production.
Kanowna Belle had its origins in a heap leaching joint
venture that enjoyed exploration success. Since feasibility it
has conformed with or exceeded expectations and is now a
major gold producer.
Kundana was built as a 35 000 oz pa mine in 1988 but had
doubled its production by the mid 1990s and continued as a
successful underground mine.
Macraes was seen originally as an eight-year project
delivering 19 000 oz pa, until new studies in 1988 described
a 700 000 tpa operation for 48 000 oz pa. By 1997 it had
expanded to 117 000 oz pa and was being further expanded
to 4.5 Mtpa for an expected 180 000 oz pa.
57
P L McCARTHY
Martha Mine was to be a 55 000 oz pa operation for
completion in 2000. It runs today at 105 000 oz pa with an
expected life to 2006.
Misima closed in 2003 after producing 3.7 M oz, compared
with the 2.5 M oz scheduled in the feasibility study.
Mount Leyshon was planned as a 2.5 Mtpa operation
producing 160 000 oz pa over a six year life. It grew to a
226 000 oz pa operation and produced about 3 M oz rather
than the 650 000 oz in the feasibility study.
Plutonic was planned as a 1 Mtpa operation to produce
87 000 oz pa over four years from 1990. It grew to produce
274 000 oz pa by 1997 and continues as a major mine with
an extended life expectancy.
Wiluna grew from a 50 000 - 60 000 oz pa operation to
double that rate, and continues so today.
Although not discovered in the study period, Sunrise Dam is
another example of a project that performed spectacularly better
than its feasibility study. Commissioned in late March 1997 at
1 Mtpa for 100 000 oz pa, it rapidly grew to deliver 382 000 oz
in calendar 2002. A project such as Sunrise Dam that is
commissioned ahead of schedule and goes on to deliver
outstanding financial results is, unfortunately, rare. Of the
projects listed above, several experienced lean and challenging
periods of several years and have not necessarily provided
outstanding returns on investment.
The balance of 31 projects (55 per cent) performed more or
less as expected. Some had short lives of three or four years as
planned, and closed in the absence of exploration success.
Several took much longer to develop than planned, but then ran
according to earlier expectations. In general, large projects
designed by major mining companies delivered their study
outcomes. These include Cadia, Granny Smith, Kidston, Lihir
and Porgera.
It is reasonable to conclude that about half of gold mining
projects perform more or less as expected, and that stakeholder
expectations will be met. About one quarter of projects will fail
prematurely, usually under adverse financial circumstances, often
involving extended litigation, administration or receivership.
These projects have the potential to leave adverse environmental
and community legacies and to reflect badly on the industry as a
whole. A further quarter of projects will perform substantially
better than the owners expectations in terms of size or mine life.
Different stakeholders will have different views on whether this
is a good thing.
The amount of engineering effort required to design a small
project is not proportionately less than for a big project, but the
funding available for studies is much less. Consequently, the
failed projects tend to be the small projects. Overall, then, project
performance may be judged acceptable to good, because the very
successful projects tend to become large projects and compensate
many times over for the failures, while the other 50 per cent just
do what was expected.
Case study: Martha Mine
The author prepared the feasibility study for the Martha Mine in
1986 - 1987, reporting initially to the project manager on site and
later to the managing director of the firm having joint venture
management. Besides the author, who was responsible for the
mining component of the study, for co-ordinating, assembling
and summarising the various reports and for financial modelling,
a number of environmental, civil and geotechnical consultants
had substantial input. The plant and infrastructure engineering
was done by Edward L Bateman Pty Ltd.
The mine is located in the centre of Waihi, a small town of
approximately 4500 people located within the Hauraki District,
58 Adelaide, SA, 19 - 22 September 2004
approximately 150 km southwest of Auckland in New Zealand.
A large underground gold mine operated in the town from the
late 19th century until the mid 20th century. At the peak of
production in 1909, Waihi was the largest precious metals mine
in Australasia and one of the leading mines in the world. Because
of its location, the project was unique in the authors experience
for the amount of environmental consideration required and its
impact on the design of the mine.
At the time of the feasibility study Martha Hill, the former
mine site, was largely overgrown by pine trees, acacia and gorse.
The subsidence crater formed above the site of a primitive
sublevel caving operation had become the town pond known as
Mine Lake and was picturesquely landscaped, a haven for ducks.
A Cornish engine house stood alongside the former Number 5
pumping shaft. The operation involved open pit mining of the
halo of lower grade material in unmined veins surrounding the
old stopes.
Waihi had reinvented itself after the mines closed as a centre
for farming, horticulture, light engineering, manufacturing and
tourism. However, an electrical goods manufacturer (Pye) had
closed in the early 1980s causing a significant level of
unemployment. While many welcomed the proposed mining
venture and saw it as the key to ongoing prosperity, there was a
strong anti-mining lobby active throughout the Coromandel
Peninsular area.
The project company was the Waihi Gold Company, a joint
venture between Amax Gold Mines (New Zealand) Limited,
Welcome Gold Mines Limited, Auag Resources Limited and
Goodman Mining Limited. The resurgence of interest in Martha
Hill began in 1976 when the Waihi Mining and Development
joint venture obtained a prospecting licence. Various phases of
exploration and small-scale trial mining followed before the
intensive open-pit drilling campaign that supported the feasibility
study.
Expectations at Martha Mine
The feasibility study was undertaken in a context of extensive
consultation with all stakeholders. It was clear to the project
manager that the project would only proceed if it had general
community support. Additionally, the New Zealand regime for
regulation and approval set very high standards and was a very
public process.
The first examination of the drilling results revealed an
unacceptable information gap. The precious metals occur in a set
of near vertical quartz veins in andesitic volcanic rocks, thought
to represent the roots of a fossil geothermal system. The vein
outcrops had been trenched and sampled but the topography had
limited the location of drill rigs so that there was little or no
sampling information, other than the trenches, to represent the
first 12 - 18 months of open pit production. New drill pads were
cut and additional near-surface holes drilled to complete the data
set. Ultimately 13 500 m of diamond drilling were completed of
which over 10 000 m lay within the pit limits.
The perimeter of the open pit was defined by the width of a
buffer zone protecting houses, public buildings, streets and the
old Cornish engine house. It was clear from preliminary pit
optimisation (using MedSystem) that an economic pit limit
would be much larger than the practical limit. The overall pit
slopes, projected down from the practical pit limit, defined the
ultimate depth of the pit.
Geotechnical studies for the pit slopes were extensive and
necessarily conservative. The altered andesites and, to the east,
fine-grained ash flows or ignimbrites were overlain by a very
young (100 000 to 6000 years) layer of air-fallen volcanic ash.
The Waihi region is situated in a quiescent, but not aseismic,
region near the Taupo-White Island volcanic zone, which is
associated with subcrustal rifting.
PACRIM 2004

The vein system fingered out near the surface into the
multiple, narrow veins typical of an epithermal system. A
method of selective vein mining was proposed, using Caterpillar
245 excavators, initially with narrow buckets. A large amount of
number crunching using the surface trench data and drill data
gave an estimate of how mining dilution would vary with depth.
This estimate was applied to the resource model to give an ore
reserve grade for each mining bench, which was used in the
mining schedule. The overall ore reserve was 8.1 Mt at 2.7 g/t
Au and 24 g/t Ag.
With a 2 m annual rainfall, pit operations would potentially be
affected. Visits to the Huntly open pit coal operations yielded
delay data that was incorporated into the mining schedules.
When the mining schedule was used in a financial model, it
was clear that the investment return and the cash flow profile did
not meet the owners requirements. A period of intensive pit
scheduling followed, ultimately yielding a four-phase pit which
deferred waste movement as long as possible while accelerating
access to the central higher grade zone at depth. Phase-one
entailed mining a 250 600 m pit to a maximum depth of 110 m
below the topographic high. Phase-two was an extension
northward and to a depth of 140 m. Phase-three was a southward
extension. Phase-four took the pit to its ultimate depth of 200 m.
The final pit perimeter was to be 700 m 450 m.
The processing plant, waste storage facility and tailings dam
were located out of town on farmland purchased for the purpose.
It was clear that truck haulage of ore and waste through the town
would not be permitted; so all material was to be crushed and
transported by belt conveyor. The conveyor was installed in a
cutting to reduce noise, passing under three roads and a stream
and through a 180 m tunnel constructed in Union Hill to
maintain an acceptable gradient.
The crusher was to be located at the conveyor head adjacent to
the phase-four pit limit. It proved impossible to meet the
requirements for truck manoeuvring at the crusher, while
maintaining the required perimeter buffer zone, once the
phase-four pit limit was reached. This problem was never
resolved in the feasibility study; instead, it was assumed that
further detailed planning and understanding of the operation
would enable a solution to be developed. This deficiency could
have resulted in a fail for the feasibility study if it had been
considered significant in a due diligence review.
The feasibility study showed ore production at an annual rate
up to 750 000 tpa with waste rock mined at up to 3 Mtpa. The
oxidised nature of approximately 65 per cent of the orebody
would allow drill and blast mining to be minimised.
The project was the subject of a planning tribunal hearing
before it received final operating permits. The environmental
impacts were scrutinised, particularly with regard to potential
impacts on the town of Waihi and the Ohinemuri River.
Reality at Martha Mine
The project was commissioned on schedule in May 1988. The
dilution estimates proved to be very conservative, so that the
mine delivered above-plan ore grades for the first one to two
years. Year One production was 63 000 oz rather than the 55 000
oz scheduled. The author and project geologists were lauded for
this project performance at the time. However, the authors
conservative dilution estimate could have stopped the project,
and it was only retrieved by the creative scheduling of the
four-phase pit.
The assumption regarding drill and blast mining proved
optimistic, and waste blasting was required earlier than planned
on the northern benches. Sophisticated delay blasting with very
small charge weight per delay was required for an extended
period until the pit was well advanced. This affected operating
costs.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
NEW MINING PROJECTS EXPECTATIONS AND OUTCOMES
To optimise the capital and operating cost balance, the
overland conveyor width was limited to 1.0 m with a speed of
5.3 m/s (19 kph). This caused excessive noise that was addressed
initially by enclosing the conveyor line and then by replacing the
conveyor idlers with a different specification with quiet bearings.
In retrospect, a wider conveyor would have saved a lot of
unexpected trouble and cost.
The looming problem of manoeuvring space around the
crusher for the phase-three pit limit was addressed by studies of a
possible truck haulage tunnel to be developed between the pit
and the plant. This was deemed impractical and instead the
crusher area was realigned as part of a pit expansion study. The
crushing operation has been an ongoing management issue and
in retrospect options for relocating the crusher out of town, with
the use of innovative haulage systems, may have been viable.
Despite careful planning, a leak during commissioning from
the tailings line over the Ohinemuri River Bridge resulted in a
tailings spill containing cyanide into the river. This had no
observable effect on the river biota, but the company was fined
$NZ 30 000. The pipeline was enclosed to prevent a
reoccurrence. This received limited media attention at the time
outside Waihi, but in todays media climate would potentially be
a global news event that could threaten the operation.
A new pond was built in another part of town and has been
developed as Gilmour Reserve, a popular recreational area.
The mining contractor (Downer Mining Limited) elected to
use a Demag 115 t excavator for waste and a Hitachi 80 t
excavator for ore instead of the 60 t excavators envisaged in the
feasibility study. The contract costs came in within a few per cent
of the feasibility study estimates, but sorting old mine timber
from the ore stream was an unexpected additional cost and delay.
Some incipient upper bench failures required innovative
management. However, overall slope designs proved
conservative.
A bund was constructed around the southeast pit boundary to
minimise visual and noise impacts. This was less effective than
expected and a fence was added to the top of the bund during
early pit development. Noise was more of a problem than
anticipated around the pit and process plant. In retrospect, it
would have been cost effective to buy out more neighbouring
properties rather than to manage noise on an ongoing basis.
The waste and tailings disposal facility was very successful.
Rehabilitated pastured slopes were grazed within two years of
construction and the progressive rehabilitation minimised visual
impacts.
In December 2001, a subsidence crater affected a residence
built over historic mine workings on the Royal Lode and
potentially affected twelve adjacent properties. Although the
cause of this subsidence was unrelated to current mining
activities, it led to substantial cost and adverse publicity for the
then operator Normandy. Monitoring and safety management
systems were introduced.
The planned life was 12 years at an average rate of 55 000 oz
per year, with final production in 2000. However, by 1997/1998
the mine was producing at 85 000 oz per year and was being
expanded to 110 000 oz per year, with life extended to 2007. It is
currently running at about 105 000 oz per year with a life
expectancy to 2006. Recent exploration success on the Favona
lode suggests that an underground mine project from a separate
orebody may greatly extend this life. There will be surplus mill
capacity, so there is the possibility of trucking in ore from other
deposits in the area.
Lessons from Martha Mine
Based on the authors experience as outlined above, the
following conclusions seem reasonable:
59
P L McCARTHY
If the overall studies are robust, any deficiencies in planning
or prediction can be made good as operations proceed.
Gains in one area may offset losses in another. Individual
estimates may prove to be high or low but, provided they fall
within the estimation accuracy ranges, the expected
outcomes can be achieved.
Changes to the environment brought about by the project
may seem profound in anticipation, but seem benign in
retrospect.
No environmental management system design by humans
will be perfect. Environmental mishaps should be kept in
context and judged by their consequences, not by the rhetoric
generated at the time.
It is better to eliminate a potential environmental problem
through additional capital expenditure than to attempt to
manage it through operational constraints and expenditure.
Of course, a new project is likely to be capital constrained
and this option may not always be available.
A new mining project is likely to be held accountable for the
sins of the past or for the actions of neighbours, no matter
how exemplarily it is managed.
A large mine can be built in the most sensitive of
environments, provided stakeholders are willing to make a
commitment to work together in designing it.
A successful project is likely to grow larger and last longer
than can be demonstrated at the time of feasibility. The
consequences can be managed through ongoing consultation
with stakeholders.
Why expectations may not be realised
Having established that many projects develop in ways quite
different to those envisaged by their planners, it is reasonable to
ask why this happens. The obvious answer is that we do not have
enough information at the time of feasibility to draw the right
conclusions. Furthermore, it may be impossible to get that
information at reasonable cost, or at all, before committing to
construction. Gold projects in nuggetty vein environments such
as Norseman, Ballarat or Bendigo must be constructed, survive
and grow on around two years of ore reserves, because
underground development must be brought up close to the
mineralisation before it can be drilled out. The position of that
development can only be designed on a year to year basis as
those drilling results come in. Nevertheless, based on a hundred
years of mining history, we may believe with considerable
certainty that the resource is there.
We cannot know what pressure urban and rural development
will put on the mine environment a decade or more hence. Town
limits grow toward sites that were once satisfactorily remote.
Grazing properties are broken up into smallholdings whose
owners have lifestyle expectations that are incompatible (to their
thinking) with having a mine as a neighbour. The former
heap-leach project at Fosterville in Victoria is one project that
experienced this transition. There is scope to investigate such
development scenarios better in mine feasibility studies than is
done at present, using the advice of real estate professionals and
regional planners.
The regulatory environment may change. However, it is
possible to track trends in environmental and workplace
standards and to design to accommodate them when required.
Two current examples are the possibility that shift lengths may
be limited in the future to less than 12 hours and that the limits
on diesel emissions may require cleaner engines or more diluting
ventilation than currently designed for. Again, these possibilities
can be examined in a sensitivity analysis and contingency plans
developed. It is rare for regulatory changes to be unexpected or
to be brought in without a transition phase that enables
progressive compliance with the new standards.
60 Adelaide, SA, 19 - 22 September 2004
The political climate may change unexpectedly. In general,
such changes fall under the heading of country risk and must be
evaluated by project proponents and financiers, though
sometimes at a local political level they could influence project
planning. The author was responsible for a feasibility study for
the Ovacik gold project in western Turkey. The mine was built,
but local political groups opposed to economic imperialism
were able to develop and exploit environmental concerns and
delay commissioning for nearly a decade. These background
political forces were apparent to the study team (there had been
claims of cyanide poisoning from the exploration diamond
drilling activities), but were not perceived to be a threat to the
project at the time.
The communitys expectations may not be realised if people
believe that mine planners are infallible, or that a feasibility
study is an accurate description of everything that will occur over
the life of the project. It is important to explain our limitations
and good intentions while pointing out that an ongoing dialogue
will be required. The regulatory framework for environmental
impact assessments in most jurisdictions is deeply flawed
because it does not recognise this iterative process. Invariably, in
the authors experience, environmental approvals are sought in
parallel with the feasibility study process, potentially
compromising both reports. Process descriptions are provided
and promises made in the environmental effects statement based
on inadequate engineering promising outcomes that cannot be
delivered under changed circumstances.
Technological changes, generally favourable, will affect the
performance of a project over decades. Subjective experience
suggests that Australian unit costs in both open pit and
underground mining have fallen by around two per cent per year
in real terms for the last two decades. Subject to the trends in the
particular commodity price, lower costs may make alternative
mining methods possible. New machines may enable different
rates and methods of excavation, materials handling or ground
support. High pressure grinding rolls can reduce costs and
simplify plant layouts. Some trends are adverse; it would be
difficult to include hand-held (airleg) mining in a long-term
mine plan because the skilled operators are no longer generally
available. A mine planner must keep abreast of such changes
and, further, must be a futurologist who can make reasonable
predictions of where the trends are leading. New technologies
may be too immature for inclusion in the feasibility study, but the
planner should make it possible to adopt them in the future.
An important reason why expectations are not realised is
termed confirmation bias. Confirmation bias is the tendency to
seek or interpret evidence favourable to already existing beliefs
and to ignore or reinterpret evidence unfavourable to already
existing beliefs. In his book Why People Believe Weird Things,
Michael Shermer (2002) reports a number of studies of
confirmation bias, including confirmation bias in scientific
research, where selective interpretation of data has occurred. He
quotes psychologist Raymond Nickerson who concluded:
If one were to attempt to identify a single
problematic aspect of human reasoning that
deserves attention above all others, the
confirmation bias would have to be among the
candidates for considerationit appears to be
sufficiently strong and pervasive that one is led
to wonder whether the bias, by itself, might
account for a significant fraction of the disputes,
altercations and misunderstandings that occur
among individuals, groups and nations.
(Nickerson, 1998)
After limited geological knowledge, it is likely that
confirmation bias is the second main reason why mining projects
tend to differ from their feasibility study descriptions. In
PACRIM 2004

interpreting a mass of sometimes-contradictory data, the
feasibility study team is biased towards an interpretation of that
data favouring a positive study outcome and perpetuating the
existing beliefs of the various pre-feasibility studies that have
gone before. To the extent that a biased interpretation of the data
can be rationalised, grades are higher, costs are lower and
environmental impacts are less than a completely unbiased
interpretation would conclude. Confirmation bias does not
happen at a conscious level, and no dishonesty is implied.
A significant factor driving confirmation bias in feasibility
studies is the commercial incentive to make the project work.
Instead of asking is the project feasible? the proponents direct
the study team to make it feasible. Depending on the corporate
culture, study teams made up of internal company employees
may have little incentive to resist this pressure. External
consultants, valuing their firms long-term reputation, may be
more objective. But perhaps the most likely to succumb to the
pressure are design and construct firms that offer the feasibility
study as a loss-leader in order to win the engineering contract.
Employees in such teams can come under great pressure to
conclude that a mining project is a viable investment.
The confirmation bias may be better understood if one
imagines a project evaluation team, employed by an NGO,
having all the necessary professional skills but having an
ideological bias against the particular mining development. Such
a team is likely to identify many risks and inadequacies in the
project and to conclude that it should not proceed. Any attempt
to reconcile these views with those of the study team in an
adversarial environment would be like bringing together the
expert witnesses for two sides of litigation. In the end, there must
be a judge or jury that determines one argument is more credible
than another.
The process of having both views examined in this way would
reveal many valid shortcomings in the mine plan and would
result in a better project outcome. That is why major mining
companies employ internal or external peer review teams for
major projects. Such reviewers will minimise confirmation bias
only if they have a long and broad exposure to mine start-ups and
a healthy scepticism regarding the study process.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
NEW MINING PROJECTS EXPECTATIONS AND OUTCOMES
CONCLUSIONS
There are many success stories in the mining industry and a
majority of mining projects deliver project outcomes generally
consistent with their feasibility studies. That is not to say that
they provide satisfactory returns on investment, because they
operate in a variable commodity price environment and because,
if grade disappoints, the required product may be obtained by
mining and processing a larger tonnage than planned. However,
around 25 per cent of projects deliver outstanding results.
Another 25 per cent of projects fail to perform. It would be a
major step forward for the industry if these projects were
identified early and prevented from proceeding. To that end,
adequate time and funding should be provided for feasibility
studies, which should be prepared by well-experienced teams
subject to rigorous peer review.
While no-one can foretell the future, it is possible to take a
broad view of the likely changes in the environment of a project
that will maximise the chance of it adapting to that environment
and becoming one of the success stories. Potential problems
should be solved, if possible, by additional capital expenditure
rather than operationally. Of course, nothing can make up for a
disappointment in grade or tonnes.
It is possible to develop a mine successfully in the most
challenging environments, provided consultation is maintained
from the outset with all stakeholders. Project leaders should be
honest about their capabilities and limitations and not pretend
that outcomes are cut and dried. Above all, flexibility must be
maintained in planning and operations.
REFERENCES
Gold Gazette, published by Resource Information Unit, monthly issues
1987 to 2004 inclusive.
Nickerson, R S, 1998. Confirmation bias: A ubiquitous phenomenon in
many guises, Review of General Psychology 2, 2:175-220, 175.
Register of Australian Mining, published by Resource Information Unit,
1987 - 1988 to 2003 - 2004 inclusive.
Shermer, M, 2002. Why People Believe Weird Things, AWH Freeman
Owl Book (Henry Holt and Company: New York).
61
The International Mining Industry Linking the Upstream
Mineralisation With the Downstream Money
N Miskelly1
ABSTRACT
Just as Australia is part of the Pacific Rim, so the Pacific Rim should be
viewed as part of the whole world. That applies in particular to that
scarce commodity money.
There are five main factors of production in any mining development
mineralisation, materials, manpower, management and money. Lets
focus on the upstream input mineralisation, and the downstream input
money, especially in the area of communications.
Unknown to many, the Competent Person reporting on Mineral
Resources and Mineral Reserves has a direct and unavoidable impact on
the companys financial statements the profit and loss account and
balance sheet.
With respect to reporting of the major asset of most mining companies
the resources/reserves investors are getting a better deal now than at
any time in memory as regards flow and adequacy of information.
Australia has been a world leader in this regard. The outlook is for even
more informative and useful reporting.
The more rigorous reporting requirements for Exploration Results and
Mineral Resources/Reserves, which have become a worldwide
phenomenon over the last five years, will continue. This will be positive
for the mining industry as a whole and for Australia in particular.
However competition for the scarce investment dollar will also increase.
Put simply, mining has many competitors, unconnected with mining,
for the capture of the investment dollar. It behoves all participants to
recognise that there are three investment factors which influence strongly
the willingness of investors to put money into mining. These are
Confidence, Credibility and Consistency. Australias investment image
stands high in these three respects, but the competition will continue to
grow. Australias lead must be maintained. However, some investment
decisions should not necessarily be based on fundamental, quantifiable
facors. Some of these are outlined.
INTRODUCTION
There are five main factors of production in any mining
development mineralisation, materials, manpower, management
and money. Lets focus on the upstream input mineralisation,
and the downstream input money, especially in the area of
communications.
The transparent reporting of mineralisation is of utmost
importance. Compared with say ten or 20 years ago, the money
people investors can understand better now what the miners
are talking about. However, just as mineralisation is always
different, equally all investors are different no one model of an
investor fits all. Just because you have met one doesnt mean you
understand them all. For each buyer of a share, there is a seller.
Tim Goldsmith of PricewaterhouseCoopers has pointed out the
Competent Person reporting on Mineral Resources and Ore
Reserves has a direct and unavoidable impact on the companys
financial statements the profit and loss account and balance
sheet. This fact may surprise some geoscientists.
The financial/investment world as well as the investment
regulatory authorities are having an increasing influence on the
mineral reserves reporting standards and the reports as prepared
by the geoscientific profession. Studies in 2003 by
PricewaterhouseCoopers and KPMG highlighted what investors
and others see as best practice in reporting, including that of
resources and reserves. The International Accounting Standards
1. FAusIMM, Consultant, 2 Pembroke Avenue, Turramurra NSW 2074.
E-mail: seriatim@tpg.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Board has pointed out the growing significance of JORC-type
reporting standards and their possible future impact on financial
accounting. The Securities and Exchange Commission in USA
(SEC) is moving towards an accommodation with the US mining
corporations and professionals, but this will assuredly move
more in favour of more demanding reporting standards.
Mining is a fashion industry, often high risk, as it is subject to
many downstream factors which can cause funding and
investment volatility, for example commodity prices, exchange
rates and start-up performance.
However with respect to reporting of the major asset of most
mining companies the resources/reserves investors are getting
a better deal than at any time in history as regards flow and
adequacy of information. Australia has been an undoubted world
leader in this regard.
The outlook is for even more informative and useful reporting
as a result of significant developments, for example, the
inclusion of the requirement for a Competent Person to be
responsible for Exploration Results reporting in the 2004 JORC
Code. Other countries have or will require Exploration Results
reporting to be signed off by a Competent Person.
The more rigorous requirements for reporting of Exploration
Results and Mineral Resources/Reserves, which have become a
worldwide phenomenon over the last five years, will continue.
The clock cannot and will not be turned back. This will be
positive for the mining industry as a whole and for Australia in
particular. However competition for the scarce investment dollar
will also increase. All participants in the mining investment
chain will need to recognise this situation.
The role of all occupations in the mining industry has changed.
Not just at the technical level, but all the way up the chain to the
top.
To quote The Economist of 30 October - 5 November 1999:
Backed by a corporate bureaucracy, a chief
executive might have got by with good managers,
cost control, and a beady eye on the accounts.
Although such things still matter, a chief
executive now has to think for himself. He needs
to be an innovator, and an entrepreneur with a
global vision. He needs political skills, to steer a
course through the regulatory maze. He needs to
be a salesman and a preacher to woo consumers,
employees and investors.
HOW HAVE CERTAIN SECTORS OF MINING
FARED AS TO THEIR OWN INVESTMENT
PERFORMANCE?
In his paper Managing Technical Risk for Mine Feasibility
Studies at The AusIMM Mining Risk Management Conference,
Sydney, September 2003, Peter McCarthy made some quite
disturbing observations, (for investors) in supporting his
conclusion that the record of mine feasibility studies is poor.
Back in the 1970s, a study for the World Bank showed that in
the first year of operation after commissioning, 60 per cent of the
mines and 70 per cent of treatment plants surveyed achieved a
production rate of less than 70 per cent of design capacity. Most
achieved their design rate by the second or third year.
63
N MISKELLY
In the 1980s a study of 35 Australian gold mines showed that
68 per cent failed to deliver the planned head grade
(Burmeister,1988) while a review of nearly 50 North American
projects showed that only ten per cent achieved their commercial
aims, with 38 per cent failing within about one year (Harquail,
1991).
In the 1990s a study of start-up performance of nine Australian
underground base metals mines found that only 50 per cent
reached design throughput by year three and 25 per cent never
achieved it. While most mines were able to operate for at least
three months at a rate 15 per cent higher than design, on a longer
term average they operated at only 80 per cent of that peak rate
or 92 per cent of design. Three of the mines (33 per cent) failed
to achieve the estimated reserve grade and after five years each
was operating at about 70 per cent of its intended grade (Ward
and McCarthy, 1999).
Over the past decade, McCarthy, in annual surveys of
feasibility studies, concluded from 105 samples, that areas of
problems were:
Area of problem Frequency
Mine design and scheduling 32 per cent
Geology, resource and reserve estimation 17 per cent
Metallurgical testwork, sampling and scale-up 15 per cent
Process plant equipment design and selection 12 per cent
Others (four areas), totalling 24 per cent
One conclusion that could be drawn is that commodity prices
and exchange rates are not necessarily the major risk for mining
investors. Another conclusion is that some mine developers dont
really understand or contend very well with the risk which may
be associated with mining at the operations end of the business.
Maybe the answer lies in the fact that for a mining company
there may be only one mine in which the company places its
whole future. However, sometimes there may be several where
the risk can be diversified or spread (or performance hidden).
Developing a new mine is a sunk cost for a miner, whereas the
investor can diversify his portfolio by spreading his risk amongst
different stocks and by playing the cyclical market by selling out
of miners when they are not hot and returning when they have
cooled down.
Or maybe most investors do not have access to the above
mentioned figures. Or the companies manage to put forward the
best image on their lack of performance when it comes to
feasibility studies.
MINERAL REPORTING CODES AND THE
INVESTOR
Owing to the introduction of the JORC Code in 1989 both
long-term investors and short-term speculators are now provided
with much more meaningful and useful information about the
stuff in the ground than they were some 15 years ago. Had the
JORC Code not come into being, investors might well have done
worse than they did in the stockmarket, or worse still may have
sought investment avenues elsewhere than in mining, resulting in
a longer and deeper capital drought in the mining sector.
As an aside it is worthwhile recalling what some (but not all)
reserves reporting standards were like in Australia in the 1970s
and 1980s, before the 1989 JORC Code came into operation.
In a paper presented at the AusIMM Sydney Branch
Estimation and Statement of Mineral Reserves Conference,
October 1979, Joe Lord, then Director, Geological Survey of
Western Australia, instanced some categories of reserves
reported to them. He quoted some at the level of assurance below
the highest level, which included indicated, probable, drill
64 Adelaide, SA, 19 - 22 September 2004
indicated, assessed, conservative, expected, directors confident,
almost certain, geologically indicated, uncertified drill indicated
and reasonably assured.
In the lowest category of assurance an even greater variety of
terms was found, namely inferred, possible, drill inferred
estimated, blocked and inferred, anticipated, broad indications,
preliminary estimate, probable existence, likely, speculative
resources, less assured additional, surmised, etc.
Compare these with the seven categories of resources /reserves
currently in use almost worldwide, through CRIRSCO and its
standards (mineral resources, and its sub-categories measured,
indicated and inferred, and for mineral (ore) reserves, and its
sub-categories, proved and probable. Theses definitions either
through CRIRSCO links or through the United Nations
Economic Commission for Europe Framework Classification,
cover most of the globe, the main excepted areas being the
Sahara region. Could any investor have made a responsible
investment decision based on these descriptions?
The trend towards more meaningful and transparent reporting
of Exploration Results in JORC 2004 and in other CRIRSCO
linked countries should ensure much more meaningful,
informative and transparent reporting for investors and other
interested stakeholders in the reporting chain.
As a result, investor confidence in mining company reporting
of resources/reserves has gained confidence in the eyes of the
investor. Not perfect, but certainly more reassuring and certain
for the investor who after all, one way or another, provides the
finance for development of mining projects.
However, miners in Canada and USA, seeing those countries
now have mineral reserves reporting standards comparable to
Australia, can offer Australian investors mining stockmarket
investment opportunities outside of Australia, thus providing
geographic diversification as well as stockbroker research
support and information which may favour the larger North
American companies.
HOW ARE THE INVESTORS BEING SERVED?
WHAT ABOUT COMMUNICATIONS?
But are the investors being serviced and informed as well as they
should be by the mining companies? Evidence suggests there are
deficiencies.
PricewaterhouseCoopers in their Digging Deeper Managing
value and reporting in the mining industry global mining survey
in 2003 (this survey since updated in 2004) concluded, inter alia,
under the heading The Communications Challenge:
The survey indicated that mining companies need
to work more effectively to achieve their dialogue
goals with the investment community. More than
half of the companies surveyed believe they work
proactively to initiate or maintain, contact with
investment community, yet only 23 per cent of
investors and only nine per cent of analysts
characterise the mining sector as proactive in
this regard.
Why should mining companies improve their communications
performance?
According to PricewaterhouseCoopers:
Analysts and investors believe that a move
toward improved reporting on the part of mining
companies will produce significant benefits for
mining companies. Improved access to new
capital, increased share prices and increased
credibility of management are among the benefits
of better disclosure cited by the majority of
investors and analysts surveyed.
PACRIM 2004

In conclusion PricewaterhouseCoopers said:
If mining companies believe that they are
undervalued and the investment community
believes they cannot value companies properly,
herein lies both the problem and the key to the
solution. Bridging this gap demands better
dialogue. A collaborative exercise involving all
stakeholders to identify indicators of value is the
first step in aligning the importance that
companies, investors and analysts place on
different indicators. Only then can companies
focus their reporting on the indicators identified
by all stakeholders as important.
THE PRESSING NEED FOR BETTER DIALOGUE
AND COMMUNICATIONS IN ALL SECTORS OF
MINING
Tim Goldsmith, Mining Leader Australasia of Pricewaterhouse
Coopers who is also on the JORC Committee, at the
AusIMM/CMMI Congress at Cairns, 27 - 28 May 2002, drew
attention to the need for dialogue interaction in the mining
industry. Whether reserves or resources are utilised, does not
change the key requirement that those using or preparing
reserves and resources statements need to understand key
aspects. For example:
Geologist/engineers
Do they know how reserves or resources will be used?
Have they imparted information on variability of key
drivers such as grade or depth or dilution?
Do they know that major financial/business altering
decisions may be based on the analysis?
Accountants
Do they understand where the numbers have been
derived from and the assumptions that may have been
made in their determination?
Do they understand the degree of accuracy (or
inaccuracy) that may exist particularly with resources?
Do they understand the history of resource reconciliations
and what it may mean for the future?
Directors
Do they understand the processes undertaken by the
company to prepare the financial report and the
interaction between the geologists/engineers and
accountants?
Do they understand the potential sensitivity of results to
the input variables?
Goldsmith added that:
Financial reporting is extremely reliant on the
use of reserves and resources, however the
geologists/engineers/accountants and directors
do not always understand the interactions
between themselves which may lead to
misleading public reporting.
The purpose of this analysis is to reflect the
importance of reserves and resources in financial
statements, and in particular that their accuracy
is vital when considering the profit and loss of a
mining company. In essence if Ore Reserves and
Mineral Resources are misstated they can
misrepresent the profit and loss account and
balance sheet.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
THE INTERNATIONAL MINING INDUSTRY
The question that this begs is whether the people responsible
for preparing the Ore Reserve and Mineral Resources statements
understand that they may well be the most important determinant
of a companys profits. Experience suggests that there are major
expectation gaps:
The Competent Person/s who signs off an Ore Reserve and
Mineral Reserve Statement often has no idea that the
information will subsequently be utilised by accountants,
directors, auditors and investors. Worse still, they do not
realise that they may be responsible for the most important
determinant in a companys profit, share price determination,
or hedging strategy.
Furthermore, accountants, directors, auditors and investors
often consider Ore Reserves and Mineral Reserves to be a
fact rather than an estimate.
The operational team will prepare a life-of-mine forecast
based on geological information. Whilst there is often a high
degree of communication between geologists and operational
teams, this is not always the case. Where this does not occur,
inappropriate assumptions may be made by the operational
team.
The accountant preparing the financial statements doesnt
always understand some of the key assumptions that may
have been made in determining resources and reserves. From
(Goldsmiths) experience, there is many an occasion when
the accountant has treated the Ore Reserves and Mineral
Resources statements as a fact, and not considered the
possible upside and downside sensitivities.
Some auditors do not always place sufficient emphasis on the
method of determination of the Ore Reserves and Mineral
Reserves, or understand the underlying assumptions.
(Goldsmith hastens to add this does not apply to all auditors!)
Ultimately, the directors take responsibility for financial
statements when they pass a resolution to approve them. As
reserves and resources may be the most important
determination of the financial statements the directors are in
effect taking responsibility for the accuracy of the Ore
Reserves and Mineral Resources Statements.
However, not all investment decisions are based on
fundamental, quantifiable considerations.
SOME NON-TECHNICAL STOCKMARKET
GUIDANCE
Everybody likes to make money on the stockmarket, but it is
almost as important that you dont lose money. In other words if
you want to continue to make money in the longer term, you
have to survive in the short- and/ or medium-term. This is not
about how to make a lot of money, but how to avoid losing what
you have.
I would like to quote extensively from Australian Business
Magazine (ABM) of January 1992 and in particular a very
perceptive article called Survival Guide by Trevor Sykes. Sykes
writes under the name of Pierpont and is widely known and
applauded as one of Australias most closely read and influential
financial journalists, and author on matters financial and
investment, including, but not exclusively in the mining sector.
Although some of these hints do not apply particularly to
mining they are relevant to stock market investment generally and
are still as relevant today as they were in 1992. Maybe more so.
Pierpont wrote:
For those battered victims of the recession (in the
early 1990s) who are still seeking a little
guidance your correspondent has assembled a
few rules. They are not exactly conventional
wisdom on investment but you could do worse.
Plenty have.
65
N MISKELLY
1. Never sign a blank.
The most monumental argument in Pierponts career was
with an English hire car company who overcharged your
correspondent. Pierpont exploded on seeing the bill but as
he had signed the invoice in advance when the company
had taken the imprint of his platinum card, Pierpont had no
effective redress. Yet people sign all sorts of blanks every
day. An employer who signs blank cheques in advance
when going on a trip is simply asking his secretary to rob
him.
2. Remember Parkinsons Third Law.
This says that perfection of planned layout is achieved only
by institutions on the point of collapse. Prospering
companies are usually so dynamic that they outgrow their
premises before they ever get around to planning them. On
the other hand, any company that has time to plan the
perfect office layout may well have run out of more
profitable occupations for its staff. This is why marble in
the foyer is an ominous sign for investors. If there is a
fountain as well, phone your broker and sell the shares.
And if the chairman of the board has personalised number
plates on his Rolls Royce, dont walk to the phone. Run.
3. The trend is your friend.
All things being equal, which they never are, you are better
off following an investment trend until it is clearly broken.
If the shares in a company rise from $5 to $10, then drop to
$8.50 then the investor who stuck with the stock is still
$3.50 ahead. Sure, he didnt sell at the top but he has still
made 70 per cent on his opening stake and you dont need
too many winners like that to have an enviable portfolio.
The alternative is to try to pick turning points, but this is
terribly difficult. There are legends told about chaps who
picked the top or bottom of the market, but in his entire life
Pierpont has never known anyone who was able to do this
consistently.
4. Did the audit partner grow up with the company?
Auditors are not cold-blooded calculating machines
(mores the pity). They are warm-blooded humans who like
to be liked. An auditor who has been inspecting the books
of a company for a decade or more is bound to have
developed friendships within it. The company might be
perfectly sound but the cosy relationship may be such that
that the auditor is turning a blind eye to things he should
not. This is so where the auditor has been with the company
since it was very small.
5. Check the balance date.
Retailers often balance in January or July because it is too
difficult to do a full year or interim stocktake on New
Years Eve. Australian banks often balance in September,
but where there is no obvious reason for the balance date to
stray outside the traditional June/December pattern, its
worth asking why.
This advice goes in spades if a group of companies includes
one or more whose balance sheet is out of line. If Blue Sky
Mines balances in June and its associated company Madcap
Metals balances in July, the accounts of both companies
can be rigged by swapping assets between them just before
their balance dates.
6. Does the company do most of its dealings with associates?
The automatic suspicion is that the companies are playing
pass the parcel between each other and the prices they are
paying for each others assets could not be justified in the
open market.
66 Adelaide, SA, 19 - 22 September 2004
7. Are the deals too complex to understand?
In the course of his long career, Pierpont has had to unravel
some blindingly complex deals and in the process has
developed a deep prejudice against them. The best business
deals Pierpont has ever seen were fairly straightforward
transactions. The worst deals tend to be the most complex
and this is by no means coincidental. Sometimes a deal has
to be made complex to disguise the fact that the basic
transaction is a sham or a poor deal.
The same goes for overly complex financing packages or
structures. If a company has so many classes of options,
warrants, shares and cumulative, redeemable, convertible,
unintelligible unsecured notes that you cannot follow the
capital structure after two or three attempts, the most
probable explanation is that you are not meant to
understand it. Throw the sponge in and buy something else.
8. Are the market and the company saying the same thing?
If the company says it is doing just fine but the market is
dumping the shares as though they had leprosy, follow what
the market is doing and sell immediately. Anyone who
followed this advice would have got out of stocks such as
Adsteam, Bond Corp, Quintex, and Westmex when they
were halfway down, which was a lot better than hanging on
until they got to zero. Yes, the market is wrong sometimes,
but its right a lot more often.
9. Whos running the place?
The most valuable piece of advice in an annual report is the
names of the directors and executives. Have they been
associated with successful or solid companies in the past?
Have their names been tainted by association with scandal
or failure? An investor who checks out these details may
not need to know more about the company. Anyone who
has a long and active business career will certainly have
one or two mistakes or failures on his CV, but a good
businessmans winners will outnumber his losers. These are
the people to back. A company may be well and honestly
managed and still fail, but it has a much higher chance of
survival than one which is fraudulently or incompetently
managed.
It is Pierponts firm belief that incompetence in a
businessman becomes entrenched after the age of 35. Any
businessman who has a history of losers after that age will
never have a winner. Incompetence has arrived to stay.
10. Find a smart adviser.
If you can do this, you can ignore all the other Pierpont
Rules. The catch, of course, is that there are very, very few
people who are really smart. A lot of people look smart and
sound smart. But if you observe them carefully over time,
they so often prove to be superficial or are having a
temporary spell of good luck. In his long life, Pierpont has
made enough investment mistakes to rid him of any
delusions of infallibility. But having read and listened to a
lot of advice from stockbrokers, accountants, newsletter
tipsters and various other people whom the Securities Act
classifies as experts, Pierpont has come to the conclusion
that their batting average is not much better, and some
proved to be congenital idiots. As a starting point, why
should you take the advice of anyone who hasnt made as
much money as you? But finding someone who sounds
smart, and is richer than you, does not provide the whole
answer. He might be smart enough to rob you, or he might
be too busy tending his own fortune tend yours as well. So
you will not only need to find someone smart, but someone
who is smart, honest, and will be diligent in looking after
your affairs. One only has to state the definition to realise
that there are not many such individuals around.
PACRIM 2004
 THE INTERNATIONAL MINING INDUSTRY

Some mining specifics CONCLUSION

In addition to Pierponts views, to be specific about raising
money from the public requires several more insights into the
mining sector, especially at the exploration end.
All old mines ceased operations because of either:
1. the war Boer, WW1 or WW2;
2. the water inflow, or
3. the sulfides couldnt be treated.
No mine ever ran out of ore.
If an exploration geologist buys shares in a hot property the
share price will decrease by at least 50 per cent within the week.
Furthermore, only God knows where all the orebodies are.
Accountants are his little helpers in keeping this information
secret.
The more rigorous reporting requirements for Exploration
Results and Mineral Resources/Reserves which have become a
worldwide phenomenon over the last five years will continue.
The clock cannot and will not be turned back. This will be
positive for the mining industry as a whole and for Australia in
particular. However competition for the scarce investment dollar
will also increase. All participants in the mining investment
chain will need to recognise this.
Put simply, mining has many competitors, unconnected with
mining, for capture of the international investment dollar. It
behoves all participants to recognise that there are three
investment factors which influence strongly the willingness of
investors to put money into mining. These are Confidence,
Credibility and Consistency. Australias investment image stands
high in these three respects, but the competition will continue to
grow. Australias lead must be maintained.
It is not the strongest specie that survives, nor the
most intelligent but the ones most responsive to
change (Charles Darwin).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 67

Modern Solutions to Old Day Mining Problems
B Morrison1

ABSTRACT

The township of Waihi lies at the base of the Coromandel Ranges on the east coast of the Bay of
Plenty, New Zealand. Waihi is the most famous gold mining town in New Zealand, the history of this
pioneering period is apparent everywhere. It still has a rich goldmine in full production, hid modestly
behind leafy slopes close to the main street.
Underground mining of gold-silver bearing quartz lodes or reefs was carried out at Martha Hill in
the middle of the town over some 70 years from 1882 until 1952, with shafts and workings reaching a
total depth of approximately 600 m on 16 levels, to mine the steeply dipping orebodies. Between the
mine levels the orebodies were mined out in subvertical excavation blocks called stopes, which
followed the reefs along their lateral extent.
Overall the original Martha underground mine became established as one of the great mines of the
world producing 1217 tonnes of gold-silver bullion from processing 12.2 million tonnes of ore
mined.
The reefs at Martha Hill extended laterally for up to 1.6 kilometres. In the early years of mining the
stopes were backfilled with waste rock, but after 1914 this practice was generally abandoned and the
stopes were commonly left empty.
A detailed drilling program between 1980 and 1984 proved large reserves of gold and silver
bearing ore remaining in stock-work zones between the previously mined richer lodes. This led in
1988 to a new era of mining at Waihi with opening of the Martha Hill open pit.
Movement of the ground surface is common where underground mining has taken place. The
ground movement can vary from extensive, relatively uniform depressions, to sudden collapse
craters. Both these types of ground movement are occurring in the township of Waihi.
Some of the voids from collapsing stopes have migrated to the surface to form subsidence craters.
This is the main hazard at Waihi and was graphically demonstrated in December 2001 where a house
collapsed into a subsidence crater in the early hours of the morning.
Detailed investigations following the recent collapse revealed that 174 properties in the heart of
the township were affected by a subsidence hazard. Eighty-three of these, many of which were
privately owned, were deemed to be unacceptable for residential or commercial land use and had a
high probability of being affected by ground collapse.
The Council, the Government and the current mining company, Newmont Mining Corporation,
while not admitting any liability for the hazards posed by the historic underground mining, quickly
mobilised to remove residents from high hazard areas and jointly put together a community solution
where the affected property owners were able to receive a fair and reasonable price for there
properties.
The quickness with which the community solution was put together and implemented was a
testament to the commitment of all the parties to the ongoing welfare of the community.

1. Mayor, Hauraki District Council, William Street, Paeroa, New Zealand. E-mail: basil@hauraki-dc.govt.nz

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 69

Issues Confronting Major Expansions and New Developments
R Shaw1
ABSTRACT
The mining industry has expanded significantly in the Pacific Region
over the past two decades through a mix of new developments and
brownfields expansions of existing operations. Despite this overall
success there are also a number of well publicised examples of less than
successful minerals ventures. These problem developments highlight the
fact that success cannot be taken for granted and that rigorous planning
and attention to detail is essential.
The dominant growth has come through expansion of existing
businesses, notably alumina, iron ore and coal, achieved through both
efficiency gains on existing plant and step change capacity additions such
as new mines and/or expanded processing plant. This expansion has
largely been linked to increased market opportunities. The major issues
are the quality and extent of the host resource, achieving the market
targets, and the efficiency of the expansion effort. There can be also a
range of site specific issues which need to be addressed.
The development of new greenfields operations is often more complex
as it can bring a wide range of new issues. Depending on the nature of the
development, and the companies involved, these can extend from initial
financing through to obtaining a market position. Incorporating major
new technology and/or establishing infrastructure and attracting staff to
remote locations can also add significant challenges.
This paper outlines many of the critical factors which need to be
considered prior to proceeding with major expansions or greenfields
developments. These will be linked to a number of developments in the
region over the past decade
BACKGROUND
The mining industry has expanded significantly in the Pacific
Region over the past two decades through a mix of new
developments and brownfields expansions of existing operations.
Figure 1 taken from Haine (2004) shows recent data for
Australia. During that time it has continued to grow in
importance to the local economies of the region. In Australia it
constitutes over eight per cent of the GDP, and represents around
30 per cent of the total exports.
This growth in the industry has been achieved through a mix of
brownfields expansions of existing operations and greenfields
developments of new plants. The net outcome for the industry has
generally been positive but there have been a number of well
publicised developments that for one or other reasons have failed
to deliver the target results. At any one time there are a large
number of projects (Figure 2, Haine, 2004) under consideration all
of which need formal evaluation prior to their being undertaken.
Much of the training on project evaluation has centred on the
financial aspects. In practice the biggest challenges facing the
engineers involved in decision making are in developing credible
numbers to use in the financial calculations and in incorporating
realistic assessments of risk factors to judge the true robustness
of the project. There is a surprisingly small amount of truly hard
data in most project evaluations. Peoples judgement therefore
plays a critical role in making correct decisions.
In this context using complex mathematical tools such as Monte
Carlo simulations in analysis does not automatically give more
reliable predictions as they are heavily influenced by the quality of
input information. Running a series of calculations with changed
assumptions can be quite sobering in terms of judging the level of
certainty that a project will be financially successful.
1. General Manager Technology Support, Rio Tinto Technology,
Melbourne Vic.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Value of Committed Projects
30
Energy Minerals Minerals Processing
25
20
$billion
15
10
5
0
1995 1996 1997 1998 1999 2000 2001 2002 2003
FIG 1 - Australian resource project history.
This paper outlines the most common issues facing companies
and provides some project examples to highlight their
importance. These examples are largely based on Rio Tintos
experience because of more ready access to information but they
have many parallels within other companies in the industry.
GENERIC ISSUES IN PROJECT EVALUATION
Strategic fit with the company
The first part of any project assessment is to ensure that it has a
fit with the overall strategy. This seemingly obvious step can be
more difficult than expected especially if the company does not
have a clearly espoused view of the future. The vision for BHP
Billitons Aluminium Business (Salomen et al, 2002) is a good
clear example of how any project decisions can fit within a wider
context. Their stated aim is being in the top three in financial
performance in the industry through being the best upstream
business.
Their growth opportunities were seen as through embedded
growth, especially in the alumina refineries, and Greenfield
opportunities in smelting. The key is low-cost stranded power
and well located bauxite. Their projects at Worsley and Mozal
are clearly consistent with the strategy and taking this through to
decision making would effectively rule out projects that did not
have a fit with these criteria.
Tangible (engineering) factors
Orebody evaluation The delineation of the orebody has become
increasingly sophisticated over the past 20 years especially with
the use of 3D visualisation and related tools. One major
challenge is the linkage of the mineral and chemical composition
data to the ability to successfully mine and process the ore. This
linkage is obviously much simpler for brownfields expansions
where there is usually considerable existing knowledge of the
ore. For greenfields developments one simple rule of thumb is
that large-scale orebodies are more able to justify the effort
needed to overcome challenges in their mining and processing.
Processing to a marketable product Virtually any ore can
technically be taken through to a marketable product but the
costs associated grow rapidly with each additional processing
step required. This is particularly true of products which have
fairly low intrinsic value such as iron ore, bauxite and coal. The
71
R SHAW
FIG 2 - Current Australian projects (2003).
level and nature of the gangue minerals are often more important
in the processing challenges than the level of the target. Some
obvious examples of this in the local context are the role of
magnesium in nickel ores and of reactive silica in the form of
kaolin in bauxite. These do not necessarily prevent processing
but can add major costs through additional reagent consumption.
Capital cost for the new development There are a number of
systems available for estimating capital costs and the approach
often depends upon the client and/or engineering company used.
Local factors and lack of definition of the flow sheet are often
the most challenging aspects in achieving an accurate estimate.
New technologies therefore represent a special challenge and this
has to be recognised in judging the accuracy of any estimate.
Operating costs for the commissioned plant As with capex
there are well established systems for estimating opex. The
largest challenge is usually in correctly estimating the usage of
energy, materials and labour rather than their specific costs. One
good approach as a sanity check in this area is to compare the
estimates made against what is achieved in modern plants
carrying out similar functions albeit possibly for different
products.
Market and price for the product Either or both can strongly
influence the likely viability of any development. For high value
low volume commodities such as gold and platinum group
metals the price achieved is more likely to be the dominant
aspect. However for bulk commodities such as iron ore volume is
also critical and the large expansions over the past decade and
currently underway are in response to strong demand, most
notably from China, which underpins both the price and tonnage.
The entrance of a new significant level of supply can alter market
dynamics and the ability to predict future pricing rather than rely
on current figures is an important challenge in setting a reliable
price for the project output.
72 Adelaide, SA, 19 - 22 September 2004
Financial returns The data assembled from the above items
can be entered into a suitable financial model to determine the
economic potential of the project. This should also include
sensitivities around the key variable to judge the robustness of
the project in terms of the quality of the estimates.
Intangibles
Unfortunately having achieved a satisfactory NPV or IRR
outcome for the well defined project is not always sufficient in
the overall evaluation. A range of other factors can come into
play which may ultimately decide the success or failure. These
primarily relate to the capability of the company to deliver the
project in respect of available financial and people resources, and
external influences which invariably arise. Some of the more
prevalent of these items are as follows.
Resourcing the construction Minerals developments are
often in remote locations and in times of high economic activity
the availability of labour can be a restriction on construction and
can add markedly to costs. Similarly concurrent construction of
more than one plant such as would have occurred in the
Gladstone area if the magnesium plant had proceeded with the
new alumina refinery can severely strain facilities such as
housing, schools, etc.
Capability to commission and operate a new plant The early
stages of any new plant can be extremely challenging especially if
the company does not have established operational experience.
The common trend in project evaluation is to assume a relatively
smooth start-up and progression to design capacity. Unfortunately
this is often not the case and strategies to handle unforseen
problems need to be considered in the final decision on the
project.
PACRIM 2004
 ISSUES CONFRONTING MAJOR EXPANSIONS AND NEW DEVELOPMENTS
Managing potential delays Both internal and external items
can lead to unforseen delays especially prior to commencing the
project construction. External approvals such as EPA clearance
are the most obvious but issues with finance, local community
groups, equipment supply, partnerships and arranging customers
all can add unforseen delays. The timeframe from an initial
conceptual study such as based on an exploration find or
perceived market opportunity can be many years. Two examples
from within Rio Tinto are the recent Hail Creek development
based on a discovery from 1965 and the Comalco Alumina
Refinery which has been under consideration in varying forms
for over 20 years.
Financing the project The issue of project finance is an area
of ongoing debate within the industry. The major companies
have a significant advantage where they can fund developments
out of internal resources rather than have to work through
external funding agencies. Conditions imposed by financial
institutions can in themselves create barriers to success.
Company culture
The overall project assessment requires rigorous consideration of
factors such as those discussed above. An underlying issue faced
in most companies is that the evaluations are often led by staff
who are also champions of the project and are virtually in a
conflict of interest situation especially in the areas where there is
a degree of subjectivity. This aspect is of most significance in
projects which are marginal and therefore struggling to make the
cut off level for proceeding. In those cases it is difficult for
project teams who are themselves committed to making the
project happen to not assume favourable positions on items such
as technical risk, cost estimates and time to achieve design
capacity.
To counter this many companies use non involved groups to
carry out independent evaluation as part of the approval process.
These can either be internal staff, often supplemented by
consultants, or fully external groups who specialise in this area.
External financing is also likely to demand this type of evaluation.
INDUSTRY EXPERIENCE PERFORMANCE AND
CASE STUDIES
Strategic drivers support persistence and deep
pockets HIsmelt direct smelting
Rio Tinto is currently constructing the first commercial Direct
Iron Smelting plant based on HIsmelt Technology in Western
Australia. The development of the process dates back over 20
years and will have cost of the order of $A 1 billion by the time
the first commercial metal is produced. A recent paper (Dry et al,
2002) gives an excellent summary of the issues facing all of
these high technology developments and elucidates the three
major lessons learnt.
These are given as:
1. the time scale for developing this type of technology is
more like 20 years than three to five years;
2. for successful development there is no way of avoiding a
large expensive pilot plant; and
3. if the underlying motive is not both strategic and strong
enough to counterbalance the risk and cost of the exercise,
dont do it.
For Rio Tinto the existence of vast reserves of high phosphorus
iron ore within the Pilbara leases and the value adding
obligations to the Western Australian Government have been the
strategic reasons for persevering with the development. Even
with that having the internal financial and technical strength to
resource the project has also been critical.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Market led expansions the iron ore industry
A recent news article (Hiscock CNN, 2003) highlights the
importance of market pull for the major expansions in iron ore.
New production figures show global miners BHP
Billiton and Rio Tinto are relying increasingly on
Chinas surging economic growth to keep sales
on track. With its economy running at an annual
growth rate of eight per cent-plus and its car
industry targeting production of two million units
this year, China has become a massive buyer of
iron ore and coking coal the key metals used
to make steel.
Australias iron ore production has grown from almost zero to
a major industry fuelled initially by marketing to Japan and more
recently through the addition of sales to China. The first plants
were built on the back of the brave commitment to construct a
mine, processing plant, and the associated housing, rail, road and
port facilities within 18 months. The more recent expansions
over the past decade (Figure 3, Hadfield, 2004) have been less
risky through being closely linked to existing ore reserves and
known markets. Even in this situation the quality of decision
making is critical to achieving full benefit from the operations.
Iron Ore Exports
250
200
Million Tonnes
150
100
50
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
FIG 3 - Growth in Australian iron ore.
The major issue for these companies has been in matching
production capability, and product type, to the emerging market.
This in turn depends upon the availability of suitable ore reserves
and being able to access them. In cases where the ore changes
there is the need for ensuring market acceptance through a mix
of small scale internal and external testing followed by larger
scale trial shipments. In parallel with establishing the market is
the need to develop the production capability. The preferred
approach of companies such as Hamersley Iron is Operational
Efficiency first, Capex second (Smith, 2004).
One real difficulty facing companies in this type of decision
making is establishing the true capacity of equipment and
correctly identifying the bottlenecks which need addressing. The
tonnage being achieved through fixed major infrastructure items
such as the rail and port facilities has steadily increased with
changes in operating practices and technology. Judging the finite
limits, and matching the need for capital spend to increase
production in line with market demand, is an ongoing challenge.
At any one time companies such as Hamersley are likely to have
6 or more feasibility studies running in parallel.
Staging the risks during development Century
Zinc
The area where Century Zinc is located had been explored since
1887 and had been the scene of nearly 50 small mines over the
following century. The discovery of a large high-grade zinc ore
73
R SHAW

in what is one of the most remote and isolated parts of Australia

in 1990 led to over a decade of intensive activity to develop the Cumulative Gold Production v Prospectus
plant. The cost of the plant and associated infrastructure has been 2,000
in excess of $A 1 billion. The development has had to overcome
numerous challenges, many of which had the potential to stop 1,800

the project on the road to completion. (Williams, 1999) Some of Actual Forecast Prospectus
1,600
these only emerged as the project unfolded and an important part
of the way the project has been conducted has been the 1,400
identification and solution of these issues such that the project
moved steadily forward without putting excessive expenditure at 1,200

risk.

Koz
1,000
The dominant early issue to emerge was the difficulty in making
an acceptable product due to the fine mineralisation and the 800
presence of excessive silica. The silica level in the initial
concentrate was well above accepted levels and not suitable for 600

existing smelters. Two process routes were developed to remove 400

part of this silica and market development work undertaken to
achieve acceptance of the economically preferred fine grained, 200

slightly elevated silica concentrate which resulted. With that in

place, and hence being able to fix a flow sheet, work could 1997 1998 1999 2000
commence on all of the normal challenges facing such a project.
A preliminary feasibility study was then undertaken which
included building and operating a pilot plant to confirm the flow
sheet and generating sufficient product (5000 t) for evaluation FIG 4 - Lihir production performance.
within a target smelter. The study also highlighted areas where
there was most potential to reduce costs through innovative
approaches. The product trials at Budelco, who were very TABLE 1
attracted by the low iron content, were acceptable and a firm Lihir financial returns.
outlet for 50 per cent of the planned production was established
to underpin further development. Full year Second half Jan-May Prospectus
1999 1999 2000
The second major issue threatening the project was the
difficulty in obtaining environmental clearances and local Ore milled 243 275 261 238
community acceptance. The need to transfer the product by (t 000/mth)
slurry pipeline to the Gulf of Carpentaria caused particular Cash costs 217 190 224 238
concerns but the feasibility studies showed this to be the only ($US)
viable option if the project were to proceed. The project Gold produced 625 354 203 600
development was largely put on hold for the time taken to (oz)
achieve agreement with the Carpenteria Land Council (CLC). Profit/(loss) (7) (1) N/A
The importance of this step was highlighted in Pasminco Gold price 335 301 400 419
purchasing the project from Rio Tinto during this time subject to ($US/oz)
Rio Tinto achieving the CLCs agreement for the development to
proceed.
Community acceptance was achieved in 1997 and at that stage
the project moved into a much more defined state where the based entirely on a much lower realisation price for the gold
decisions needed focussed on much more tangible issues of produced that was predicted in 1995.
construction, equipment selection and the like. These in In essence the project proponents were reasonably accurate in
themselves had many challenges because of the location and size much of their data but the timing coincided with a low point in
of the projects but were much more manageable and did not the market. The short term volatility of many commodity prices
threaten the project. inevitably also adds risks to short term financial performance
The dominant lesson from this project development is the which can be a serious issue if there is insufficient financial
importance of overcoming the potential show stoppers as early as backing to ride out such hiccups. The project evaluation should
possible to ensure that the expenditure committed is kept in line be based on long term predictions on prices but must also include
with the development status of the project. sensitivity studies to ensure that fluctuations will not put the
company at risk.
The issue of metal price Lihir Gold
A good project opportunity needs to be well
The Lihir Project formally commenced development in 1995 executed
when a prospectus was prepared for a new company Lihir Gold.
Although several of the major companies held a stake in the Unfortunately the track record of the chemical and metallurgical
company the development was carried out based on a public industries is poor in managing projects to meet the basic success
float. A recent presentation (Lihir, 2000) gives a good criteria suggested by Morrow (1994) of:
summation of the subsequent performance of the company <10 per cent over budget;
against that prospectus.
starts up within three months of schedule; and
The project performed operationally above the levels used as
the basis for the prospectus in virtually all aspects (Figure 4, reaches 85 per cent capacity within 12 months.
Table 1). Despite this the financial performance was not so Typically only around 20 per cent of projects achieve all three
attractive, especially in the first two to three years of operation, of these.

74 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 ISSUES CONFRONTING MAJOR EXPANSIONS AND NEW DEVELOPMENTS
Chris Twigge-Molecey (2003) examined the factors
contributing to our industry spending over $US 20 billion in non
productive capital investment in 43 unsuccessful projects over
the past decade. His analysis indicates that over two-thirds of the
projects did not have proper phasing, over two-thirds had no
continuity in the execution teams from concept to start-up and
over 40 per cent had serious front end development issues. The
issue of capability to exercise the project also needs to be part of
the decision making process.
These failures exist across both brownfields and greenfields
projects and within well established companies. They are also by
no means restricted only to those projects dependant on new
technologies. The recurring theme in the analysis was the
importance of up front work with an underlying suggestion that
if the project is not robust enough to bear the cost of this
expenditure then there must be real doubts about its viability.
Given that the up front component is usually only a small portion
of the final installed cost this is a reasonable assessment.
The people component our resources are
decreasing
An issue highlighted by Twigge-Molecey (2003), and becoming
increasingly more evident in our own environment, is the
reduction in the number of skilled experienced people to provide
reliable information and make sound judgements especially on
the less tangible items. This is true both within the operating
companies and also within the supply companies who service the
industry.
Over the past 20 years our workforce has changed markedly
through staff losses by retirement and lay-offs, attrition and
voluntary departures. We have also struggled to maintain the
training of replacements. The technology area which underpins
many of our new developments has been particularly badly
effected with the closure and/or downsizing of many of the major
centres around the world. As an industry we will have to adapt
successfully to the ever growing competition from other areas
many of whom have the advantage of not operating in remote,
often unhospitable areas.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
CONCLUSION
Development of new projects is a complex activity and exposes
companies to significant risks. As engineers we tend to deal
much more effectively with the tangible issues but often the
largest challenges in the project are in the intangible decisions
where subjective judgements become important. Much of the
success or failure of any project therefore hinges on companies
having high calibre people supported by appropriate systems to
drive sound decision making and being prepared to apply the
resources to provide reliable data for the evaluations.
REFERENCES
Dry, R J, Batterham, R J, Bates, C P and Price, D P, 2002. Direct
Smelting: Why Have So Few Made It?, in Proceedings International
Conference on Iron Ore Processing, Kiruna, June.
Hadfield, S, 2004. Register of Australian Mining 2004/05 (RIU: Western
Australia).
Haine, I, 2004. Minerals and Energy, Australian Commodities,
11(2):325-330.
Hiscock, G, 2003. China driving iron ore sales, CNN.com Business,
24 July.
Lihir, 2000. JB Were Gold Conference (Paris), June.
Morrow, E and Yarossi, M, 1994. Managing Capital Projects. Where
Have we Been Where are we Going? Chemical Engineering,
October, pp 108-111.
Salomen, M, 2002. BHP Billiton Aluminium Briefing.
Smith, D J, 2004, Rio Tinto Iron Ore: Meeting the demand challenge in
Australia, AJM Conference.
Twigge-Molecey, C, 2003. Knowledge technology and profit, in
Proceedings Copper 2003-Cobre 2003, Vol 1 (MetSoc CIM).
Williams, I, 1999. Century Zinc Project in The Spirit of the Snowy Fifty
Years On (ATSE).
75
Recent Advances in Geophysical Tools
R J Smith1
ABSTRACT
This paper reviews recent advances in geophysical tools for mineral
exploration. It does not restrict comments to methods which are usually
applied in the tectonically active belt surrounding the Pacific Ocean, where
geophysics has often been secondary to geological prospecting (eg visual
recognition of surface alteration) and drainage geochemistry; but considers
geophysical tools applied to a broader range of both deposits and terrains.
For example, the subdued topography of much of Australia and
Canada has lent itself to the development and extensive application of
airborne methods. In Australia, deep weathering and transported cover
have necessitated the use of geophysical methods to not only detect
anomalies but also to map and understand basement geology. This same
weathering has caused specific problems, particularly for electrical
methods, but recent improvements in instrumentation, processing and
visualisation have largely overcome them.
A great deal of geophysical data is now being collected by government
agencies, often from the air, and made readily available to explorers. In
several Australian states it is free, often on line, and in almost all cases it
is relatively cheap.
These regional data sets are often augmented by more detailed local
surveys in areas of interest.
Although anomalies are still important much more is being done with
the high quality data which is now available The results are used for
geological mapping, to test geological hypotheses, to construct detailed
3D geological models and, together with fluid flow modeling, to predict
optimum locations for mineralisation.
INTRODUCTION
This paper will review some recent advances in geophysical tools
applied to mineral exploration. Exploration in the Pacific Rim
has tended to emphasise porphyry Au, Cu/Au or Cu/Mo deposits
in the tectonically active belt surrounding the Pacific Ocean. This
belt is often typified by steeply dissected terrain with well
defined drainage, and a significant part occurs in low magnetic
latitudes. For such targets, and in such terrain, geophysics has
often been secondary to geological prospecting (eg visual
recognition of surface alteration) and drainage geochemistry;
which have frequently been more cost effective. Nevertheless,
geophysical methods have often made significant contributions
and more can be achieved.
This paper will consider geophysical tools applied to a broader
range of both deposits and terrains.
For example, the subdued topography of much of Australia
and Canada has lent itself to the development and extensive
application of airborne methods. In Australia, deep weathering
and transported cover have necessitated the use of geophysical
methods to not only detect anomalies but also to map and
understand basement geology. This same weathering has caused
specific problems, particularly for electrical methods, but recent
improvements in instrumentation, processing and visualisation
have largely overcome them.
Some general comments will be made about the range of tools
available, all of which continue to develop incrementally. In
addition, some specific comments will address three major areas
where specific advances have occurred in recent years:
airborne gravity,
digital data sampling and processing, and
data inversion and visualisation.
1. FAusIMM, Greenfields Geophysics, PO Box 295, Mont Albert Vic
3127. E-mail: greengeo@bigpond.net.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Finally, some comments will be made about the improved
availability of regional data sets and how they are being used to
improve our understanding of 3D geology, and even to predict
favourable locations for the formation of mineral deposits.
The reader is referred to a recent review paper on
developments in geophysical tools for mineral exploration during
the last decade of the 20th century by Nabighian and Asten
(2002) for additional information on specific methods.
AIRBORNE METHODS
Magnetics and radiometrics
Airborne magnetic and radiometric methods are usually
considered to be mature technologies, however incremental
improvements continue, and even some new trends are emerging.
The two methods are frequently used together as basic tools in
almost all terrains.
High quality magnetic data is routinely collected in either
fixed wing or rotary wing aircraft, usually sampling at 0.1 second
intervals or better. In recent years both line spacing and terrain
clearance have been reduced to the extent that the detail now
collected from the air can often replace ground measurements, at
much lower cost.
The use of multiple sensors to measure magnetic field
gradients can assist in interpolation between flight lines and
improve resolution of small features (Hogg, 2004). Several
contractors now offer such a service.
Directional sensors measure the direction, as well as the
intensity, of the magnetic field and these have been applied in
areas of strong remanent magnetisation to improve data
interpretability (see Dransfield et al, 2003).
Radiometric data quality has improved dramatically with the
application of statistical methods to a broad radiometric
spectrum (commonly 0 to 3000 keV) which encompasses natural
radiation from potassium (40K) and the decay series of uranium
and thorium. Several noise reduction methods have been
developed and they are reviewed in a recent publication by
Dickson and Taylor (2003). The results are being used to classify
lithologies and even to attempt to identify zones of alteration
within them.
Electromagnetics (EM)
Airborne EM methods have also developed significantly. One
major advance was the introduction of TEMPEST, developed
jointly by CRCAMET and World Geoscience in the 1990s,
incorporating signal processing to improve data quality and
hence both depth penetration and interpretability. Other fixed
wing systems (eg Geotem and Megatem) have increased power
and depth penetration also.
Helicopter borne systems have a smaller footprint than the
equivalent fixed wing systems and they appear to achieve better
resolution. Several new instruments have appeared and a
particular trend has been the introduction of helicopter borne
time domain systems (AeroTEM and VTEM in Canada and
HoistEM in Australia) which offer improved resolution and
interpretability.
Figure 1 (A and B) shows the dimensions and configuration of
HoistEM and VTEM. They both use a horizontal transmitting
loop (coil) with a receiver loop at the centre, towed behind and
beneath a helicopter. Figure 2 shows HoistEM in flight in
Western Australia.
77
R J SMITH

A
B

FIG 1 - Diagrammatic representation of (A) HoistEM and (B) VTEM coil configurations (GPX, 2004 and Witherly et al, 2004).

FIG 2 - HoistEM in flight, Western Australia.

78 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


FIG 3 - Aerotem response over Powerline Deposit, Sudbury Basin, Canada (Aeroquest, 2004).
Most helicopter TEM systems use a similar geometry but differ
in transmitter waveform and power, sampling times and
processing options. They all appear to offer significantly better
resolution than the more powerful fixed wing systems and this is
often an important factor in interpretation. Figure 3, illustrates this
resolution in an area of surface infrastructure where power lines
might otherwise obscure the target response (Aeroquest, 2004).
In this example the Powerline deposit occurs between two
power line corridors, separated by approximately 350 m. The
responses from the power lines are resolved within about 75 m,
enabling the detection of a discrete conductor between them. This
anomaly was drilled and intersected sulfide mineralisation,
including 1.3 per cent Ni, 6.7 per cent Cu and 13.3 g/t TPMs over
approximately 13 m, subsequently named the Powerline Deposit.
Processing and visualisation of AEM data has also progressed
incrementally so that conductivity depth images (CDIs) are
used routinely by most practitioners.
Historically most helicopter borne AEM systems have
operated in the frequency domain. They have often suffered from
calibration problems and instrument drift so that quantitative
processing (eg to CDIs) has been difficult and unreliable. Recent
research has made progress in recalibrating historical data by
post processing to improve imaging and inversion (Ley-Cooper
and Macnae, 2004). If successful, this method should permit the
extraction of more and better information from existing historical
data, often now in the public domain.
Research is also ongoing in the development of calibration
methods for time domain AEM (Macnae and Davis, 2004).
Gravity
A major breakthrough in airborne geophysics has been the
introduction of airborne gravity gradiometry and the further
development of airborne gravity. While there are distinct
differences between the two types of measurements, both
measure variations in the gravity field and both respond to
density variations in the subsurface.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
RECENT ADVANCES IN GEOPHYSICAL TOOLS
Airborne gravity was first introduced by Carson in the 1980s,
but in general it has not offered sufficient sensitivity or resolution
to be widely used in mineral exploration until recently. In addition,
of course, the costs involved have often restricted its use to terrains
where ground measurements were not practicable. There are now
several contractors offering airborne gravity services at cost
effective rates, and with improved sensitivity and resolution.
Sander Geophysics and Fugro both operate from North America.
Canadian Microgravity has started operations in Australia and
recently completed a survey in the Northern Territory for a
consortium of companies and government agencies.
At present all airborne gravity systems measure the sum of the
aircraft acceleration (eg its response to turbulence) and the
acceleration due to gravity. In general, the aircraft acceleration is
also measured independently by GPS methods and subtracted to
give the acceleration due to gravity. Consequently the accuracy
of GPS limits the accuracy of the gravity measurement. Usually
the measurement is filtered by averaging to reduce the noise, but
with some loss of resolution. This reduction in noise with
smoothing, along a flight line, is illustrated in Figure 4, taken
from Argyle et al, 2000.
Surveys are usually flown on close spaced lines over a grid and
this permits additional filtering across adjacent lines. Further
effective noise reduction is possible by such grid filtering as
shown in Figure 5, also taken from Argyle et al, 2000. In this
figure each curve corresponds with a different filter length (in
seconds). Obviously, noise is reduced with increasing filter
length and further reduced by averaging across several lines,
each of which will contribute to loss of resolution.
An airborne gravity survey was recently (2003) undertaken in
the Northern Territory over an area on the western edge of
Arnhem Land adjoining Kakadu National Park (Figure 6, Gabell
and Tuckett, 2004; Duffett et al, 2004), commissioned by
Geoscience Australia, NT Geological Survey, Cameco Australia
Pty Ltd and Rio Tinto Exploration Pty Ltd. The survey was
conducted jointly by Fugro Airborne Surveys Pty Ltd and
Canadian Micro Gravity Pty Ltd.
79
R J SMITH

FIG 4 - Gravity noise versus filter cut-off (seconds) FIG 5 - Effect of line averaging on noise reduction
(Argyle et al, 2000). (Argyle et al, 2000).

FIG 6 - Location of West Arnhem survey, showing pre existing gravity stations on an approximate 11 km 11 km grid.

This survey was designed to replace a ground gravity survey
with stations at 2 km 2 km station spacing, due to difficulties
with ground access. Pre existing ground gravity data was on an
approximate 11 km 11 km grid (Figure 6) and the additional
work was planned to provide greater detail. An image of the pre
existing ground data is shown in Figure 7. The data used to
compile this image was not terrain corrected.
The survey was flown on east west lines at 2 km spacing with
north south tie lines at 20 km intervals. Although readings were
taken at two-second intervals, they were averaged over longer
periods to reduce noise. Figures 8, 9 and 10 show Bouguer
gravity images (including terrain corrections) from the airborne
survey with filter lengths of 107 seconds, 80 seconds and
60 seconds respectively. The 60 second filter, with a half width
of 2.1 km, was considered too close to the line spacing to work
well and it includes artifacts perpendicular to the line direction.
The other two images clearly show more detail than the pre
existing ground data with some loss of detail in the longer filter.
The airborne survey costs and data resolution were similar to the
ground survey originally considered but the survey was
completed in much less time and without ground access.

80 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RECENT ADVANCES IN GEOPHYSICAL TOOLS

FIG 7 - West Arnhem, pre-existing Bouguer gravity image FIG 9 - West Arnhem, Bouguer gravity image, from airborne data
(not terrain corrected). with terrain corrections and 80 second filter.

FIG 8 - West Arnhem, Bouguer gravity image, from airborne data FIG 10 - West Arnhem, Bouguer gravity image, from airborne data
with terrain corrections and 107 second filter. with terrain corrections and 60 second filter.

Airborne gravity has indeed advanced significantly and it can

now compete with ground measurements for regional surveys, but
further improvements are needed to provide the signal to noise,
and resolution, needed for most mineral exploration applications.
Gravity gradiometers measure the gradient of the gravity field
rather than the field intensity itself. Since the acceleration due to
gravity is a vector quantity, the gradient is a tensor with nine
components. In practice only five of these are independent and, if
they are measured, the remaining components can be calculated
explicitly. In addition, if measurements are made over a survey
grid, as is usually the case, a map of any one component can be
mathematically transformed into a map of each of the other
components. Although the various components may offer
different advantages for interpretation, we are mainly concerned
with the vertical gradient of the vertical component of the gravity
field (GDD). This component shows a simple high or low
anomaly over a point source of different rock density, similar to
conventional gravity maps, but with higher resolution (sharper FIG 11 - Diagrammatic representation of a Gravity Gradiometer
definition). (BHPB).
In its simplest form, a gravity gradiometer consists of two
gravity sensors a fixed distance apart (see Figure 11). The gravity In 1999 BHP Billiton introduced and commenced operations
gradient is the difference in measured gravity divided by the with the FALCON Airborne Gravity Gradiometer (AGG)
distance between them (a horizontal gradient in this case). system. This is a modified version of a Lockheed Martin system,
Gravity is usually measured in mgal (milligal) or ms-2 adapted for airborne operations. Three FALCON AGG systems
(micro-metre per second squared). Gravity gradient is simply the are now in service around the world and they are being used in a
gravity field divided by a distance and is usually expressed in Eo range of applications. Some of these will be addressed separately
(eotvos, where 1 eotvos = 1 ns-2, or one per nanosecond squared). by Harman et al (2004).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 81

R J SMITH
Subsequently, Bell Geospace Inc (BGI) has also adapted their
marine full tensor gradiometer (FTG) for airborne operations.
Both BHP Billiton and BGI use Lockheed Martin technology but
alternative systems are under development.
Both airborne gravity and airborne gravity gradiometry can be
used to upgrade and interpolate existing regional gravity
networks. This process of merging the disparate data sets is still
in its infancy, but some progress has been made (Lane, 2004a).
Eventually this should lead to more detailed, better quality
regional gravity sets, many of which are often readily available
from government agencies for use by mineral explorers.
Ideally, when high quality gradiometer data has been collected
and processed optimally, it should rival the resolution and
information content of detailed airborne magnetic data. Figure 12
below shows a grey scale image of terrain corrected GDD (the
vertical gravity gradient) over a section of the EKATITM
Diamond Mine property within the Slave Province (NWT,
Canada) from a Falcon survey flown by BHP Billiton. The
survey covering a total of 37 000 line km was flown at an altitude
of 80 m (draped over terrain) on lines 100 m apart, and
completed in 16 weeks.
Several previously known kimberlites are identified and these
are all clearly apparent in the image. Most of these occur under
lakes and there is no doubt that the lakes and lake sediments
contribute to the response. Nevertheless three new kimberlites
which had weak or no response to other methods, and were not
previously recognised, were identified by this survey.
The image shown in Figure 12 is quite remarkable. Rarely, if
ever, is it possible to see such an extensive area covered with
detailed gravity measurements, of any kind. The high quality
now available in airborne gravity gradiometry, together with the
speed and efficiency of data acquisition has made this possible.
In addition to the kimberlites, and many lakes, which are
evident as small dark areas in this image, there is clear evidence
of differences in basement lithology, and several N-S and
NNW-SSE dykes, particularly in the eastern half of the image.
FIG 12 - Terrain corrected airborne gravity gradient (GDD), EKATI
82 Adelaide, SA, 19 - 22 September 2004
Further improvements in data quality are anticipated and no
doubt the time will come when the joint interpretation of high
quality magnetic and gravity data sets will be used routinely to
interpret basement geology
A recent workshop on airborne gravity, held in conjunction
with the ASEG conference in Sydney (August 2004), reviewed
the current state of the art and speculated on future
developments. The proceedings are summarised in a Geoscience
Australia Record (Lane, 2004b).
GROUND AND BOREHOLE METHODS
GPS
Although not a specific geophysical tool, it seems appropriate to
mention the wide availability and reliability of GPS as a major
advance. As well as facilitating field operations in remote areas,
it has direct application to the location of geophysical
measurements, both on the ground and in the air. An obvious
application already mentioned is the monitoring of aircraft
accelerations in airborne gravity measurements. Further
improvements in GPS accuracy are planned for the next decade;
including both reliability and accuracy (Enge, 2004) and these
improvements will in turn generate new applications.
Electrical (and electromagnetic) methods
The main changes in electrical geophysical methods are due to
improved electronic instrumentation, and vast improvements in
digital data storage and processing. These will also be reviewed
separately by Ritchie (2004).
Compact, reliable sensors and multichannel receivers enable
the collection of more and better data for the same cost, and in
innovative ways. In particular, a number of multi-purpose
receivers are now available. These are in fact field computers
which can use a range of sensors, record data and process in real
time, or even record complete time series for later processing.
TM
.
PACRIM 2004

The use of multichannel receivers can speed up data collection
and can include remote sensors outside the survey area to
monitor background electrical noise (eg telluric noise). In
addition, the ability to deploy multiple sensors has led to the
development of 3D surveys.
Figure 13 illustrates an electrode array which can be used to
measure 3D resistivity and IP. The main receiving unit (CRU) is
located at the centre of the array, with multiple receivers (DAUs)
deployed over a grid, and a current injection point which can be
moved sequentially during the survey. The DAUs all operate
simultaneously and data is transmitted back to the central unit at
the end of the reading, for each transmitter location. A full-time
series can also be recorded from each sensor for later analysis
(see Ritchie, 2004).
FIG 13 - Electrode array for 3D electrical survey (GRS, 2004).
The simplest real time processing, averaging repeated readings
or stacking has been used for many years. This was superseded
to some extent by smart stacking, which rejected outliers,
sometimes sequentially, and then by more sophisticated signal
processing, which identifies and rejects noise from a variety of
sources before stacking.
FIG 14 - Removal of recorded tellurics from IP data (GRS, 2004).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
RECENT ADVANCES IN GEOPHYSICAL TOOLS
Data storage and processing
Advances in computer technology make it possible to store vast
amounts of data and process it rapidly. It is now possible to store
closely sampled, complete time series and to process them either
in real time or later. Coupled with advances in signal processing,
this has resulted in very significant improvements in data quality
which, in turn increase depth penetration and detectability of
subtle responses. Even more importantly perhaps, this facilitates
the effective application of quantitative interpretation methods
forward modelling and data inversion.
If a remote sensor has been used to record telluric noise during
the survey, it can be fitted to and subtracted from the field data.
Figure 14 (GRS, 2004) shows the recorded time series (pale
grey), the inferred natural field (mid grey) and the final corrected
signal (black). In this case the pronounced, low frequency telluric
signal has dominated the total received signal and the data would
be virtually unusable without its removal.
Sources of electrical noise can come from a variety of
sources including low frequency tellurics, high frequency sferics,
cultural noise (usually 50 Hz in Australia), electrode contact drift
and so on. These are discussed in some detail in Paine and
Copeland (2003). They have taken a different approach to
Ritchie (2004) and correct for the various noise sources by direct
analysis of the time series without a remote sensor. The various
noise sources are identified and removed separately by digital
processing.
Figure 15 (Paine and Copeland, 2003) shows pseudo-sections
of apparent resistivity (top panel), raw apparent chargeability
(second panel), corrected apparent chargeability (third panel) and
the telluric amplitude (bottom panel). The corrected apparent
chargeability shows a much more coherent pattern and is clearly
more interpretable than the raw data. Tellurics are a major source
of noise in the raw data but there are others. Some have been
removed by manually editing out obviously bad readings due to
electrode effects.
Both of these approaches utilise the time series to remove bad
data and improve final data quality. This is extremely important
if the new modelling and inversion tools are to be used correctly.
Inversion methods can usually produce a result consistent with
the data, but the result may be meaningless if the data is
corrupted by noise.
83
R J SMITH
FIG 15 - IP Pseudo-sections corrected for telluric noise (Paine and Copeland, 2003).
Of course, the noise from natural fields (eg tellurics) may be
utilised as signal if it can be recorded in a meaningful way.
Magnetotelluric (MT) and audio magnetotelluric (AMT)
methods do use these natural fields as energy sources. These
methods can be used for deep resistivity soundings but they
require intensive processing and rely heavily on computing
power. The ability to store and process time series is helping to
make these methods more practical and accessible.
INTERPRETATION
Data availability
In the last decade or so regional geophysical data has become
more readily available at low cost. In fact several state surveys
within Australia, and Geoscience Australia, often deliver high
quality, high resolution data at no cost. Globally, nations and
states or provinces are competing for exploration dollars. In
order to attract such activity (and dollars) they provide base
geological data (in increasing detail) and other geoscientific data
(geophysical and geochemical) to potential explorers. This
means that even sole operators, consultants or prospectors, can
generate exploration plays at low cost and without the support of
company resources.
84 Adelaide, SA, 19 - 22 September 2004
This availability of basic data (often on line) is of enormous
assistance to explorers, whether they have the resources of a
major company behind them or not. Interpretation is an
intellectual exercise which requires primarily experience and
access to data. Such data access is empowering the individual to
participate fully in future exploration.
Inversion 2D and 3D
There have been a number of developments in the inversion of
geophysical data, much of it pioneered at UBC, Vancouver, but
now emulated by several others. All of these methods produce a
model of one or more physical properties in the subsurface, from
field data. The model should be compatible with the data, within
a specified error margin, and it can also be constrained to
conform to any known parameters (eg depth of cover or known
contacts from drilling). Of course, the model is never a unique
solution and other possibilities may also be compatible with the
data, but it does provide a powerful interpretation tool.
One of the first developments was 2D inversion of, say, a line
of IP and resistivity data to produce a 2D model (a section) of the
resistivity and IP (usually chargeability or phase) of the
subsurface. This has since been extended to 3D inversion of
several lines of data, or of data collected in a 3D mode.
PACRIM 2004
 RECENT ADVANCES IN GEOPHYSICAL TOOLS

FIG 16 - Fluid flow simulations, Kewell Dome, Western Victoria (Rawling, T, 2004b). The dome trends NNW-SSE and this image is viewed
from the south west. Maximum fluid flow is indicated near the top of the dome which has now been removed by erosion, but significant
targets remain.

3D inversions of magnetic and gravity data have also been
developed and provide extremely powerful tools for visualising
the possible sources of a complex data set. 3D inversion of
magnetic data has proved particularly useful in low latitudes and
steep terrains, where the source location is often far from
straightforward. This is the case in many prospective areas
around the Pacific rim.
Geophysical inversion models are powerful tools, but they
supplement rather than replace interpretation. They should be
used with care, and with some understanding of the limitations
and ambiguity inherent in the inversion process.
Geological modelling
Many of the geophysical tools described above were initially
applied in the search for anomalies or bright spots which were
immediate and obvious targets. Although anomalies remain
important, geophysical interpretation has long progressed beyond
this and it is frequently used now to develop or validate
geological models. These in turn can be used in a predictive
sense to define prospective areas or even drill targets.
Two good examples were presented at the recent ASEG
conference in Sydney, in August, this year.
The first (Williams et al, 2004) used constrained 3D inversions
of gravity and magnetic data in a 150 km by 150 km region
around the giant Olympic Dam copper-uranium-gold deposit,
where basement is obscured by cover. This work used the
inversion methodology developed at UBC. The use of a reference
model, based on geological observations, guided the inversion
towards a geologically reasonable model, and allowed the testing
of geological hypotheses. The result was a map of alteration and
structure which is consistent with both the known geology and
the observed potential field data.
The second example (Guillen et al, 2004) used a different
approach, developed by BRGM (France). In this case the
software tools develop a 3D geological model from geological
observations (points at lithological contacts, structural
observations of interface orientations and a table of geometrical
relationships between lithological units) as a starting point.
Physical properties are assigned to the defined lithologies and the
model is varied to reproduce the observed data (ground gravity
data only in this case) within defined tolerances. This example
was applied to a 20 km by 20 km area around Broken Hill where
considerable data was available from a long history of mining
and exploration activity. Airborne gravity gradiometer data is
also available in this area now and it will be incorporated
following some modifications to the software.
Fluid flow modelling
Recent work at pmd*CRC, based at the University of Melbourne,
has used 3D models based on a combination of aircore drill data,
potential field data and geophysical inversion modelling as a
basis for numerical simulations of fluid flow. The approach used
was developed and refined around the Magdala gold mine at
Stawell, where extensive data was available from underground
workings and drilling. It was then extended into new areas and
used to guide the definition of additional drill targets. Some of
this work was reported earlier this year (Rawling et al, 2004a)
and work is continuing.
Figure 16 shows a 3D model of the Kewell basalt dome with
fluid flow vectors and zones of high fluid flow (high
mineralisation potential) identified. The top of this dome has
now been removed by erosion but areas of high potential on the
flanks of the dome remain and they are currently being
evaluated.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 85

R J SMITH
CONCLUSIONS
A number of new tools have appeared in the last decade. Many of
these are incremental improvements in existing technology,
closely linked to developments in electronic components and
computers. Signal processing methods often originating in
communications technology are now being routinely applied to
geophysical data.
A quantum leap in geophysical tools is the introduction of
airborne gravity gradiometry. This technology, which holds great
promise, will continue to develop, in conjunction with
interpretation tools.
With better tools, and particularly better quality data, realistic
and useful 3D geological models can now be derived. These
models can now be used in a predictive sense to define prospective
areas and even, supplemented by numerical fluid flow simulations,
to indicate targets for drill testing.
ACKNOWLEDGEMENTS
Several people generously contributed material used in this paper.
In particular the author would like to acknowledge contributions
from Richard Lane (Geoscience Australia), Greg Walker (BHP
Billiton), Terry Ritchie (Geophysical Resources and Services P/L),
Tim Rawling (pmd*CRC), John Paine (Scientific Computing and
Applications), Andrew Boyd (Integrated Geophysical Solutions
Pty Ltd), Ron Creagh and Pat Cunneen (GPX Airborne Pty Ltd),
Mark Duffett (NT Geological Survey) and Mario Bacchin
(Geoscience Australia). Data from the West Arnhem airborne
gravity survey was obtained from the NT Geological Survey web
site and used with the permission of both NT Geological Survey
and Geoscience Australia.
REFERENCES
Aeroquest, 2004.
http://www.aeroquestsurveys.com/pdf/AeroTEMDesignConsideratio
ns.pdf, July.
Argyle, M, Ferguson, S, Sander, L and Sander, S, 2000. AIRGrav results:
a comparison of airborne gravity data with GSC test site data, in The
Leading Edge, Vol 19, pp 1134-1138 (Society of Exploration
Geophysicists).
Dickson, B L and Taylor, G M, 2003. Quietening the noise: an evaluation
of noise reduction methods applied to aerial gamma-ray survey data,
Exploration Geophysics, 34(1&2):97-102 (Australian Society of
Exploration Geophysicists).
Dransfield, M, Christensen, A and Liu, G, 2003. Airborne vector
magnetics mapping of remanently magnetized banded iron
formations at Rocklea, Western Australia, Exploration Geophysics,
34(1&2):93-96 (Australian Society of Exploration Geophysicists).
Duffett, M, Bacchin, M and Lane, R, 2004. West Arnhem Land airborne
gravity survey, Geological Survey record 2004-001, pp 7-8 (Northern
Territory Government).
Enge, P, 2004. Retooling the Global Positioning System, Scientific
American, May:64-71 (Scientific American Inc).
Gabell, A R and Tuckett, H, 2004. Project 200380 West Arnhem Land
Northern Territory GT1A Airborne Gravity Survey Acquisition
and Processing Report, prepared by Fugro Airborne Surveys Pty Ltd
and Canadian Micro Gravity Pty Ltd for Geoscience Australia,
http://www.dme.nt.gov.au/ntgs/geophysics/regional_grav.html#2003
%20Gravity%20Program, July.
86 Adelaide, SA, 19 - 22 September 2004
GPX, 2004. GPX Airborne Pty Ltd, http://www.gpx.com.au, July.
GRS, 2004. GRS MIMDAS Telluric Cancellation, Geophysical
Resources and Services, Company Brochure, 2004.
Guillen, A, Courrioux, G, Calcagno, Ph, Lane, R, Lees, T and McInerney,
P (2004). Constrained gravity 3D litho-inversion applied to Broken
Hill, in Extended Abstracts, ASEG 17th Geophysical Conference and
Exhibition, Sydney (Australian Society of Exploration Geophysicists).
Harman, P, Isles, D and Anderson, C, 2004. The FalconTM airborne
gravity system as an exploration driver in Australia, in Proceedings
PACRIM 2004, p 221 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Hogg, S, 2004. Practicalities, pitfalls and new developments in airborne
magnetic gradiometry, First Break, 22:59-65 (European Association
of Geoscientists and Engineers).
Lane, R J L, 2004a. Integrating ground and airborne data into regional
gravity compilations, in Airborne Gravity 2004 Abstracts from the
ASEG-PESA Airborne Gravity 2004 Workshop, (Ed: R J L Lane)
Geoscience Australia Record 2004/18, pp 81-97.
Lane, R J L, 2004b. Airborne Gravity 2004 Abstracts from the
ASEG-PESA Airborne Gravity 2004 Workshop, (Ed: R J L Lane)
Geoscience Australia Record 2004/18 (Geoscience Australia in
press).
Ley-Cooper, Y and Macnae, J, 2004. Theoretical treatment of amplitude
calibration problems in HEM data, Extended Abstracts, ASEG 17th
Geophysical Conference and Exhibition, Sydney (Australian Society
of Exploration Geophysicists).
Macnae, J and Davis, A, 2004. Surface loop monitoring of airborne
electromagnetic systems; paper to be presented at the SEG
International Exposition and Seventy-Fourth Annual Meeting,
Denver, 1 - 15 October 2004 (Society of Exploration Geophysicists).
Nabighian, M and Asten, M, 2002. Metalliferous mining geophysics
State of the art in the last decade of the 20th century and the
beginning of the new millennium, Geophysics, 67(3):964-978
(Society of Exploration Geophysicists).
Paine, J and Copeland, A, 2003. Reduction of noise in induced
polarisation data using full time-series data, Exploration Geophysics,
34(4):225-228 (Australian Society of Exploration Geophysicists).
Rawling, T, Schaubs, P, Dugdale, J and Wilson, C, 2004a. Development
of new mineral targeting strategies using 3D modelling and
numerical fluid flow simulation techniques in western Victoria, in
Dynamic Earth: Past, Present and Future, Geological Society of
Australia Abstracts 73, p 113, 17th Australian Geological
Convention, February (Geological Society of Australia).
Rawling, T, 2004b. Personal communication. July.
Ritchie, R, 2004. Recent developments in electrical exploration, in
Proceedings PACRIM 2004, pp 251-256 (The Australasian Institute
of Mining and Metallurgy: Melbourne).
Williams, N C, Lane, R and Lyons, P, 2004. Regional constrained 3D
inversion of potential field data from the Olympic Cu-Au province,
South Australia, Preview, 109:30-33 (Australian Society of
Exploration Geophysicists).
Witherly, K, Irvine, R and Morrison, E, 2004. The Geotech VTEM time
domain helicopter EM system, ASEG 17th Geophysical Conference
and Exhibition, Sydney, 2004.
PACRIM 2004
Successful Management of Mining Companies What Have We
Learned From the 16th Century to the 21st Century?
A Trench1
ABSTRACT
This short paper outlines the authors views regarding the progress we
have made as an industry in managing mining assets over the past 500
years. It represents a personal view but one founded in fact-based
observations of mining companies and of the mining industry in general.
There is no doubt that Agricolas views as expressed in De Re
Metallica (1556) were visionary. There is also little doubt regarding
another issue that whilst a great deal has been learnt about the
management of mines since Agricolas time, observations suggest that
best practice management is the exception rather than the rule across our
industry. Although insights regarding the successful management of
mines have become apparent, we still witness unnecessary errors in our
industry that lead to the loss of shareholders funds.
By highlighting some aspects of current best practice, particularly in
the arena of operations management, it is hoped that this short paper
contributes a small step along the next 500-year-long learning curve.
The paper concludes with a brief description of the progress and
impact of continuous improvement initiatives at Kalgoorlie Consolidated
Gold Mines (KCGM). Located in the Kalgoorlie Goldfields, KCGM is
currently Australias largest gold mine producing over 800 000 ounces of
gold per annum. Improvement initiatives at KCGM look set to deliver
some $100 million in value over life-of-mine.
INTRODUCTION
The task set for this paper is a tough one: To synthesise the
learning of close to 500 years of managing mines into a short
conference paper. As a result, I propose to adopt a highlights
approach. The underlying assumption is that the highlights
contained herein are indicative of the state of the mining industry
as a whole. Personal experience from serving a number of large
and small companies in the metals and mining sector suggests
this assumption may be reasonable. The paper closes with a brief
look at one mining operation the KCGM SuperPit in
Kalgoorlie.
This paper briefly touches upon the following three areas.
What are the key lessons from the past, in particular
Agricolas De Re Metallica, that have significance to
management and ownership of mining assets in the modern
mining industry?
What key success factors characterise the successful
management of mining assets?
What does the future hold for our industry in respect of
management and ownership of mines? Put simply, where will
further improvement come from?
LESSONS FROM THE PAST AGRICOLA
Originally published in 1556, Agricolas De Re Metallica
represents the first book on mining to adopt a scientific approach
based upon field research and observations. It stands out as a
classic text to the present day. It is therefore insightful to revisit
the book to ascertain what Agricola had to say regarding the
management and ownership of mines.
1. MAusIMM, Manager Continuous Improvement, Kalgoorlie
Consolidated Gold Mines, Kalgoorlie WA 6433. Adjunct Professor,
Mineral Economics and Mine Management Western Australian
School of Mines, Curtin University of Technology. Director Heron
Resources Limited. Director Pioneer Nickel Limited.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Three lessons from De Re Metallica stand as true today as in
the 16th century.
Firstly, on mine management (described by Agricola in the
role of the mine owner), he recognised the importance of
gathering first-hand facts regarding the performance of a mine:
It is important that the owner who is diligent in
increasing his wealth should frequently himself
descend into the mine, and devote some time to
the study of the nature of the veins and stringers,
and should observe and consider all the methods
of working, both inside and outside the mine.
Secondly, Agricola recognised the importance of frontline
morale to the success of a mining venture:
sometimes he (the owner) should undertake
actual labour, nor thereby demeaning himself,
but in order to encourage his workmen by his
own diligence
On mine ownership, Agricola arguably pre-empted Markowitz
(1952) portfolio theory of ownership and returns by almost 400
years. He states:
When one man alone meets the expense for a
long time of a whole mine, if good fortune
bestows on him a vein abundant in metals, or in
other products, he becomes very wealthy. If, on
the contrary, the mine is poor and barren, in time
he will lose everything that he has expended on
it. But the man who, in common with others, has
laid out his money on several mines in a region
renowned for its wealth of metals, rarely spends
it in vain, for fortune usually responds to his
hopes in part.
The lessons from these statements stand true today.
Fact-based observations remain a key
management tool
One pitfall that faces present-day managers is being caught
unawares by failures buried within the complexity of modern
mining operations. Faced with a mining fleet that comprises
multiple loaders, trucks and drills, state-of-the-art processing
facilities with sensitive circuit characteristics and sophisticated
supply and administration systems, one might be excused for not
knowing quite where to start.
An important part of managing well is therefore to base ones
management actions upon fact-based observations. Many
management decisions are still made in the absence of a strong
representative fact-base. Reactive solutions to problems as they
arise are only a part of management. The diligent use of
representative fact-based observations to identify improvement
opportunities before problems arise is an under-utilised
technique. This is despite pioneering work using an
observation-based approach dating back close to 100 years (eg
Taylor, 1911).
Frontline morale is critical
Agricola noted that a mine owner should undertake actual labour
occasionally to encourage his workforce. Whilst modern day
87
A TRENCH
mine managers, male or female, need not regularly pick up an
air-leg or jump into the cab of a truck, building strong rapport
and mutual respect with frontline employees to raise morale is a
key aspect of successful contemporary mine management.
In a recent review of mine management at a gold operation,
the author conducted frontline interviews across all functional
areas. The objective was to gauge workforce morale, its
contribution to operational performance and to gather
preliminary feedback on a new system for capturing and
implementing employees ideas. Although all departments were
performing strongly against budget, there were noticeable
differences in the outlook of the frontline employees.
Staff in one department exhibited very high morale, typified by
the following commentary:
I really love my job We already generate lots of
ideas, but more feedback will really accelerate
things.
A second department exhibited solid morale synthesised as
follows:
Morale is OK. We need to see action on
implementing our ideas. Were not sure our
supervisors are really prepared for this (ideas
system implementation).
A third department clearly had work to do to embrace its
frontline staff:
We dont trust them (management). With all the
favouritism and politics, I dont think anything is
going to happen (to our ideas).
Not surprisingly, operational improvements achieved in the
first two departments were significantly greater than in the latter
department.
KEY SUCCESS FACTORS FOR MANAGING
MINING ASSETS
As mining managers, our task is to recommend and implement
changes to work practices that deliver tangible bottom-line
improvements. The culmination of successful mine management
practice should then impact positively upon share price
performance of minerals companies. Ultimately, share price
performance is the acid test for what works and what does not
when it comes to managing mining assets.
Comparisons at a company level are significant; they have the
potential to reveal those companies that have systematically
improved methods of management rather than relying upon
individuals to enact change.
But efficient management of mines forms only one aspect of
managing a successful minerals company: There are other
methods of value creation in addition to efficiency of mine
production. Indeed, traditional thinking when it comes to
creating value in the minerals sector often focuses upon three
arenas exploration (eg Woodall, 1984; White, 1997), project
development (eg Carter et al, 1996; Galloway, 2004) and mergers
and acquisition activity (eg Rudenno, 2004) as the sources of
differentiation between mining companies.
A study of share performance in the mining sector (Trench
2002) did not identify these factors as long-term sources of
competitive advantage for mining firms. Conversely, the study
suggested that companies that have developed and followed
renegade strategies achieve the best results. Renegade strategies
are based upon novel company-specific capabilities that
differentiate some companies from their peers. Perhaps
surprisingly, no mining company stood out as building its
strategy around worlds best practice mine management. Equally
surprising, companies that enjoyed lower production costs than
peers did not directly translate such advantage into higher
average shareholder returns.
88 Adelaide, SA, 19 - 22 September 2004
Some renegade strategies that generated above-normal returns
for shareholders in the period 1985 - 2001 included the following:
sophisticated hedging using more advanced risk management
than peers;
deriving income from mining royalties; and
holding non-managing interests in producing mining assets.
So whilst efficient management of mines should be a
successful ingredient in delivering above-normal investment
returns, the evidence that mining companies differentiate
themselves in this way is not yet clear based upon historical
performance.
LOOKING TO THE FUTURE
To be successful, mining companies need to deliver strong
investment returns in order to attract the capital required to
develop new mines and processing facilities. In addition to
creating value through developing new mineral discoveries and
realising synergies through industry consolidation, the efficient
management of operations provides a further pathway to create
shareholder value.
Mining companies are seeking to achieve such operational
efficiencies through deployment of a variety of management
tools. Indeed, the list of tools is becoming extensive. Techniques
include benchmarking, process mapping, management by
objectives, quality management (including Six Sigma) and
re-engineering to name but a few.
KCGM CASE STUDY
Kalgoorlie Consolidated Gold Mines (KCGM) produces some
800 000 ounces of gold per annum and is 50 per cent owned by
Barrick Gold and 50 per cent owned by Newmont; two of the
worlds largest gold miners.
KCGM has adopted a simple framework to capture value from
improvement opportunities. The framework involves the
application of a three-step problem-solving process to all
improvement opportunities (see Croser, 2004). Approaching
three years on from the commencement of the improvement
program, life-of-mine gains in the order of $100 million look
achievable. This value will be derived from production
enhancements driving revenue gains (~80 per cent) and from cost
efficiency gains (~20 per cent).
The problem-solving process is summarised as follows.
Stage 1: Get the facts, value the gaps
This phase diagnoses the size of the improvement potential based
upon fact-based analysis. Specifically, the subprocesses are to:
identify an area of potential focus;
subdivide the area of focus into its constituent elements;
diagnose the existing process and performance across each
subprocess; and
Size the Prize of the improvement potential in annual cash
flow and life-of-mine value.
Critical to success in Stage 1 is to focus the improvement
effort towards the highest value opportunities and to accurately
divide the problem into its constituent parts. Typically, the
highest value areas represent the bottlenecks to increased
production.
Stage 2: Load the silver bullets
Generate ideas for improvement targeted upon a focused area
of opportunity;
PACRIM 2004

evaluate and prioritise ideas based upon relative pay-off and
difficulty; and
approve Go-do ideas.
Critical to the success of Stage 2 is to confine the
problem-space sufficiently in order that improvement ideas,
often sourced from frontline employees, address the key parts of
the overall opportunity.
Stage 3: Make people want to do it
Prepare action plans for all approved ideas;
assign team member responsibility to ensure implementation
occurs;
monitor performance against targets (see for example Greif,
1991); and
provide visual, written and verbal feedback to all personnel
on all ideas submitted and improvements gained from the
ideas.
Critical success factors at this stage include the strong
communication of project results and active tracking through the
use of performance boards and through periodic management
updates and reviews.
CONCLUDING REMARKS
This short paper has suggested, but not proven, the following
hypotheses:
that lessons from Agricolas 1556 De Re Metallica remain
pertinent to modern-day mining management;
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
SUCCESSFUL MANAGEMENT OF MINING COMPANIES
that efficient mine management is one pathway for mining
companies to add significant shareholder value; and
that deploying management tools, in the case of KCGM,
structured problem solving, is one means through which to
unlock shareholder value in mining assets.
ACKNOWLEDGEMENTS
Brent Johnston, John McLuckie and John Shipp.
REFERENCES
Agricola, G, 1556. De Re Metallica. Translated from the First Latin
Edition by Herbert Clark Hoover and Lou Henry Hoover (1950,
Dover Publications: New York).
Carter, J, Van Dijk, M and Gibson, K, 1996. Capital investment: how not
to build the Titanic, McKinsey Quarterly, 4:146-159.
Croser, A, 2004. Continuous improvement at a large open pit operation,
in Proceedings The AusIMM New Leaders Conference, pp 91-98
(The Australasian Institute of Mining and Metallurgy: Melbourne).
Galloway, P, 2004. Project risks and returns, BHP Billiton address to the
CRU Aluminium Conference, New York, available at
www.bhpbilliton.com
Greif, M, 1991. The Visual Factory Building Participation Through
Shared Information (Productivity Press).
Markowitz, H M, 1952. Portfolio selection, Journal of Finance, 7:77-91.
Rudenno, V, 2004. Mineral Valuation Handbook, Second Edition
(Wrightbooks-Wiley).
Trench, A, 2002. Make your firm a business renegade, Executive
Excellence (Australian Edition), 19(5):20.
White, A H, 1997. Management of Mineral Exploration (Australian
Mineral Foundation).
Woodall, R, 1984. Success in mineral exploration the Joubin James
lectures, Geoscience Canada, 11(1):41-46.
89
Deposit Models and Regional Reviews

Gold Distribution and Geochemistry at Batu Hijau, Indonesia Geological J Arif, T Baker, C Clode, 93
and Economic Implications for Porphyry Copper-Gold Deposits B Harlan and
A Perdanakusumah

Gold Exploration of the Phuoc Son Tam Ky Suture, Central Vietnam M J Banks, R H Murfitt, 95
A Case Study N N Quynh, L V Hai

Geology of the Gold Deposits of the Yanacocha District, Northern Peru P D Bell, J G Gmez, 105
C E Loayza and R M Pinto

Common Geological Characteristics of Prominent Hill and Olympic Dam A Belperio and H Freeman 115
Implications for Iron Oxide Copper-Gold Exploration Models

The Sams Creek Peralkaline Granite Hosted Gold Deposit, Northwest R L Brathwaite and K Faure 127
Nelson, New Zealand A New Variant on Alkaline Intrusion-Related Gold
Deposits

Disseminated Gold Mineralisation in a Schist-Hosted Mesothermal D Craw, D J MacKenzie and 135

Deposit, Macraes Mine, Otago, New Zealand B Petrie

The Jianchaling Mine, Shaanxi Province, Peoples Republic of China M Erceg, K Fan and 143
A Carlin-Type Gold Deposit? T Leach

Distribution of Alteration and Mineralisation in Northern Carlin Trend Gold T M Leach 153
Deposits, Nevada

Recent Investigations of Submarine Hydrothermal Activity in Indonesia T F McConachy, 161

H Permana, R A Binns,
I Zulkarnain, J M Parr,
C J Yeats, N D Hananto,
B Priadi, S Burhanuddin and
E P Utomo

A Geological Review of the Tampakan Copper-Gold Deposit, Southern C Middleton, A Buenavista, 173
Mindanao, Philippines B Rohrlach, J Gonzalez,
L Subang and G Moreno

The Tintina Gold Province, Alaska and Yukon New World-Class Gold S H Mueller, R J Goldfarb, 189
Resources and Their Sustainable Development C J R Hart, J L Mair,
E E Marsh and
C S Rombach

Data Metallogenica Around the Pacific Rim and Across the World K N OSullivan and 199
A D T Goode
Gold Distribution and Geochemistry at Batu Hijau, Indonesia
Geological and Economic Implications for Porphyry Copper-Gold
Deposits
J Arif1,2, T Baker1, C Clode3, B Harlan4 and A Perdanakusumah2

ABSTRACT

Gold is an important by-product in the porphyry copper-gold deposits but the distribution and
chemistry of gold in such systems remains poorly understood (Titley, 1990; Ulrich et al, 1999).
Previous research on gold in porphyry copper-gold deposits were based on experimental (eg Simon
et al, 2000; Jugo et al, 1998; Gammons and William-Jones, 1997; Cygan and Candela, 1995),
metallurgical (eg Kesler et al, 2002; Ballantyne et al, 1997) and fluid inclusion studies (eg Ulrich
et al, 1999; Loucks and Mavrogenes, 1999; Werner et al, 2002). Here we report the results of
petrographic, flotation tests, microprobe and laser ablation inductively coupled plasma mass
spectrometry (LA-ICP-MS) studies of gold and associated copper sulfides within a paragenetic
framework from the world-class Batu Hijau (914 Mt @ 0.53 per cent Cu, 0.40 g/t Au) porphyry
copper-gold deposit, Indonesia. Unlike many porphyry copper-gold deposits early copper minerals
(bornite-chalcocite-digenite) are well preserved at Batu Hijau and late sulfidation
(chalcopyrite-pyrite) overprint is less developed, hence it provides an excellent insight into the entire
gold paragenesis of the porphyry system. More than 700 native gold grains were identified in over
100 polished thin sections. Almost all of the native gold grains occurred within the quartz veins,
attached to sulfide or as free gold along quartz or silicate grain boundaries. The native gold grains are
dominantly round in shape and mostly 2 - 12 mm in size. The majority of gold was deposited during
the formation of the early A veins and dominantly associated with bornite rather than
chalcopyrite.The petrographic and LA-ICP-MS study results indicate that in bornite-rich ores gold
mostly occurs within copper sulfide grains as invisible (ie within the sulfide structure) and as native
gold grains, whilst in chalcopyrite-rich ores gold mostly occurs as native gold grains with lesser
invisible gold. The petrographic observations also indicate a higher proportion of free gold (native
gold not attached to any sulfide) in chalcopyrite-rich ores compared to bornite rich ores. The pattern
of free gold distribution appears to correlate with the flotation test data, where the average gold
recovery value from chalcopyrite-rich ores is consistently lower than bornite-rich ores. Our data
suggests that porphyry copper-gold deposits with chalcopyrite-rich ores are more likely to have a
higher proportion of free gold that may require different ore processing strategies.

1. Economic Geology Research Unit, School of Earth Sciences, James Cook University, Townsville Qld 4811.
2. Mine Geology Department, Batu Hijau Mine, PT Newmont Nusa Tenggara, Jl Sriwijaya 258 Mataram, Lombok NTB 83125, Indonesia.
3. Mining Services, Newmont Mining Corportation, Malozemoff Technical Facility, Englewood CO 80112, USA.
4. PT Newmont Horas Nauli, Batangtoru, North Sumatra, Indonesia.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 93

Gold Exploration of the Phuoc Son Tam Ky Suture,
Central Vietnam A Case Study
M J Banks1, R H Murfitt2, Nguyen Ngoc Quynh1 and Le Van Hai1
ABSTRACT
The Truong Son Fold belt in Indochina hosts the significant Sepon
gold-copper district (Laos) and the similar gold-basemetal districts of
Phuoc Son and Bong Mieu (Vietnam).
With over seven years of exploration, Toronto-listed Olympus Pacific
Minerals Inc, is one of few foreign companies that have to date made a
substantial investment in mineral exploration in Vietnam. Much of this
work has been conducted along the Phuoc Son Tam Ky Suture, on the
south-eastern limb of the Truong Son Fold belt on the Vietnamese side of
the border with Laos.
These programs have produced a significant number of gold prospects
at varying stages of exploration development.
The Phuoc Son Tam Ky Suture is a major Paleozoic subduction zone
that extends from northern Laos, through Central Vietnam and out into
the South China Sea. It represents a Lower Mesozoic collision zone
between the Paleozoic-age Truong Son fold-belt terrain to the north, with
the crystalline terrain of the Pre-Cambrian Kontum Block to the south.
Since its formation, it has been a locus of deep faulting, continental
accretion, intensive intrusive activity and mineralisation.
This paper briefly discusses the geological setting, structure,
metamorphism, mineralisation, exploration history, and proposed mining
development of the Olympus Phuoc Son Gold Project, with particular
reference to the Phuoc Son Tam Ky suture.
INTRODUCTION
The Bong Mieu and Phuoc Son gold properties are being
explored by Olympus Pacific Minerals Inc (OPM), a Toronto
listed public company. Both projects are in Quang Nam province,
near the port city of Danang, in Central Viet Nam (Figure 1).
Both Bong Mieu and Phuoc Son gold projects are located
within the Phuoc Son Tam Ky Suture, which comprises an E-W
trending belt of deep-seated, high-angle faulting, allocthonous
ophiolite lenses and intrusives, where the southern boundary of
the Truong Son Foldbelt is juxtaposed against the northern
boundary of lower Proterozoic to Archaean crystalline
(upper-amphibolite to granulite facies) metamorphic rocks of the
Kon Tum Block (Figure 2).
Since the Bong Mieu project, 72 km SSE of Danang, is the
subject of a separate paper at this conference, it will not be
further addressed in this paper.
The Phuoc Son project, 85 km southwest of Danang (140 km
by road) and 6 km northwest of the Phuoc Son district town of
Kham Duc, is on the new Hanoi-Saigon national highway (Ho
Chi Minh highway) (Figure 1). The town has a population of
~5000, with excellent infrastructure, including a hospital,
mechanical workshops, abundant water and a 500 KV national
electricity grid.
This review will concentrate on the geology and exploration
results within the Phuoc Son section of the Phuoc Son Tam Ky
Suture.
1. Olympus Pacific Minerals (Vietnam) Ltd, 113/2 Nguyen Chi Tanh
Street, Danang, Vietnam.
2. MAusIMM, R H Murfitt and Associates, PO Box 46-225, Herne Bay
Auckland, New Zealand.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
MINING AND EXPLORATION HISTORY
There are unconfirmed accounts of gold mining in the Phuoc Son
district during the French colonial era, but the recorded
exploration of the district commenced in 1976 - 1985, when
geologists of the Vietnam government Geological Survey
(DGMV), conducted regional 1:200 000 scale mapping in the
district. In June 1995, Indochina Goldfields Ltd in cooperation
with the Vietnamese equivalent of a Geological Survey jointly
explored the Phuoc Son district for gold and associated metals. A
BLEG (bulk leach extractable gold) regional stream sediment
geochemical survey collected 507 samples collected (average
one sample per 5 km2). Of these, 61 were deemed anomalous
(threshold: 7 ppb Au; peak value: 592 ppb Au) Over 50 per cent
of the anomalous samples came from a single 240 km2 area of
contiguously anomalous catchments draining the Dak Sa Ridge.
In 1997 Indochina Goldfields withdrew from Vietnam and
Olympus Pacific Minerals purchased a majority interest in the
property. From 1997, artisan gold mining became widespread in
the area.
In 1998, two 50 km2 Exploration Licences were granted and
systematic exploration of the licences commenced with visible
gold being returned from widespread pan concentrate samples
and high gold values from float, outcrop and artisan mine
stockpile samples.
Since 1998, Olympus has continued with increasingly advanced
geological, geochemical, geophysical and drilling programs. Since
ore grade mineralisation was first drill-intersected at Bai Dat in
1999, potentially economic mineralisation is being delineated
progressively at more than 26 prospective localities within the
100 km2 project area.
GEOLOGY
Stratigraphy
The project area is dominated by probable Late Neoproterozoic
to Cambrian), regionally metamorphosed (lower greenschist
facies) shallow marine metasediments and intercalated meta
volcanics of the Nui Vu formation (Figure 3). This sequence is in
high-angle reverse fault contact with Upper Proterozoic Kham
Duc formation amphibolite facies metamorphics along the east
and south-east sides of the property (Figure 3). Lithologies
include; albite-actinolite schist, mica schist, phyllite, psammitic
schist, metabasite, marble, quartzite and chert. The depositional
environment is interpreted to have been a shallow marine
marginal basin.
Intrusives
Linear, Lower-Palaeozoic to Mesozoic, calc-alkaline magmatic
belts trend broadly east-west across Quang Nam province, along
the axis of the southern Truong Son Foldbelt (Figure 2). A
variety of post-tectonic granitoid complexes of upper Mesozoic
age occur along deep-seated high angle faults.
Basic to intermediate intrusives
Small lenses, dykes and stocks of amphibolite, gabbro, diorite,
quartz diorite, andesite and dacite porphyry occur throughout the
licence area. Pre-metamorphic, syntectonic and post tectonic
varieties are present.
95
M J BANKS et al

FIG 1 - Property location map.

96 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 GOLD EXPLORATION OF THE PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM

FIG 2 - Regional geology.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 97

M J BANKS et al

FIG 3 - Geology, gold occurrences and resource locations.

98 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 GOLD EXPLORATION OF THE PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM
Mesozonal granitoid plutons
Two coarse-grained granitoids occur in high terrain west and
southwest of the project area.
Epizonal granitoids
Small-scale (less than 1 sq km) stocks of medium-grained granite
to monzogranite and related dykes occuring within the tenement
near Khe Rin, Round Hill, K7 and Bo prospects (Figure 4),
outcrop poorly within the intense saprolitic weathering profile at
Phuoc Son.
Hiep duc ultramafics
Shear-bounded serpentinite, talc serpentite, greenschist and
metagabbro lenses occur within the Nui Vu sequence, particularly
in the Dak Sa valley. These are regarded as allocthonous thrust
wedges, of probable Silurian age.
Principal faults
Shear zones, common along contacts between competent and
ductile units, are often the focus of mineralisation at Phuoc Son.
The Poco River fault zone (PFZ)
A regional, 30 km deep (Pham et al, 1995) NNE-SSW trending
reverse-dextral fault system which extends for over 100 km. The
fault juxtaposes Late Proterozoic and Paleozoic metamorphics
and is inferred to represent rifting along the western side of the
Kontum block.
Cong Plong fault zone (CPFZ)
This 2 km wide, SW dipping, 35 - 40 km deep (Cau, 1991)
regional fault is prominent in satellite imagery. At Kham Duc,
olivine basalt flows of late Tertiary-early Quaternary age have
flowed from this structure. Indosinian-age dextral strike slip fault
movement is inferred to have been reversed to sinistral
reactivation during the post-Mesozoic Himalayan Orogeny.
Dak Sa fault zone
At Dak Sa, a low-angle NW dipping thrust fault zone extends
through the Dak Sa Prospect for over 4 km and is host to two
significant gold deposits, comprising quartz veins and hydraulic
quartz-sulfide breccias, at Bai Dat and Bai Go prospects.
Folding
Within the project area northerly plunging, open, isoclinal folds
predominate, but fault/shearing of fold axes may be common.
Saddle reef quartz veins are locally developed (eg at Bai Gio).
Deformation contrasts are marked between different
lithologies; competent units are often only slightly folded, whilst
ductile units may be tightly folded.
Geotectonics
Suturing and accretion of Late-Proterozoic to Cambro-
Ordovician miogeosynclinal marginal basins occurred from the
late Pre-Cambrian to Triassic. Ophiolites along the southern
boundary of the Truong Son Foldbelt are interpreted to be
remnants of Early Palaeozoic lower lithosphere (possibly
Silurian age), marking the position of an early Palaeozoic
NE-dipping subduction zone, (Tran et al, 1997). These ophiolites
are faulted within lower greenschist facies shallow marine
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
sediments and basic volcanics of probable Late Neoproterozoic
to Cambrian age.
Compressional regime, dextral NW-SE trending wrench
faulting of the Truong Son Foldbelt occurred during late
Palaeozoic-Early Mesozoic Indosinian Orogeny (Lepvrier et al,
1997). These faults and the older sutures were later reactivated
during the Burma/Eurasia and India/Eurasia collision events and
subjected to tensional tectonic regimes in the Triassic-Jurassic,
Cretaceous and early Quaternary (Himalayan Orogeny),
producing new faults during deformation in response to the
Burma-Eurasia and India-Eurasia collisions. The Indosinia
geoblock (including the Kon Tum massif and Truong Son
Foldbelt) is thought to have rotated 29 clockwise since the
mid-Cretaceous (Hutchison, 1989). At Phuoc Son this has
resulted in NW-SE trending sinistral wrench faulting, NE-SW
dextral wrench faults, and E-W normal faults; all in addition to
older Indosinian and Lower Paleozoic thrusts and reverse faults
consequent upon subduction tectonics in that era.
In the Phuoc Son area, the Phuoc Son Tam Ky suture and the
whole Truong Son Foldbelt is bounded by large NW-SE trending
wrench faults and flexed to the northwest, where it continues
northwesterly for more than 200 km across the Vietnam/Laos
border.
Metamorphism
The most significant Phanerozoic regional metamorphic event
began in the Silurian and reached a peak in the Permo-Triassic
(Indosinian) Orogeny.
Metamorphism of the Nui Vu formation is mainly of low to
moderate greenschist facies. Psammitic units comprise
fine-grained aggregates of: albite-actinolite, metabasites;
chlorite-albite-actinolite-epidote, whilst pelitic units comprise
mainly sericite-biotite-albite.
Although regional metamorphism does not appear to have
reached the almandine isograd, pale pink garnet occurs within
thermal aureole skarn and hornfels zones adjacent to
post-tectonic granitoid intrusives.
Schistocity is developed within micaceous pelitic lithologies
and talc serpentinites, but psammites, metabasites, marbles and
quartzites commonly show little or no schistocity.
MINERALISING ENVIRONMENTS
Exploration at Phuoc Son has revealed widespread hydrothermal
gold and base metal sulfid mineralisation related to the
emplacement of high-level acid-intermediate intrusives.
Potentially economic gold grades over potentially mineable
widths have been returned from surface sampling and diamond
drilling at numerous showings, most notably at Bai Dat, Bai Go,
Bai Gio, Khe Rin, Khe Do, and Tra Long (PSGC pers com).
Four distinctly different styles of mineralisation have been
recognised, as follows.
Low-temperature Au-Ag-Pb-Zn
At the southern end of the property (Dak Sa sector), the Bai Dat
and Bai Go deposits (1 km apart) comprise structurally
controlled, mineralogically similar (gold-silver-galena-
sphalerite-pyrite-pyrrhotitechalcopyrite) mineralisation zones,
within mesothermal quartz veins and hydraulic vein breccias
(Figure 5), formed in dilational zones within a pre-existing
2.5 km long, low-angle shear structure. The sulfides are locally
massive, but are more commonly confined to irregular brittle
fractures in quartz and as disseminations and in schist foliae
within vein wall and country rocks.
99
M J BANKS et al

FIG 4 - Geology and resource location, Dak Sa.

100 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 GOLD EXPLORATION OF THE PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM

FIG 5 - Representative NW-SE cross-section, Bai Dat orebody, Dak Sa.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 101

M J BANKS et al
TABLE 1
A summary of Mineral Resource estimates from Bai Dat and Bai Go deposits.
Category/deposit Tonnes
Measured and Indicated Resources
Bai Dat 165 000
Bai Go 153 000
Total Measured and Indicated Resources 318 000
Additional Inferred Resources
Bai Dat 115 000
Bai Go 73 000
Total additional Inferred Resources 188 000
(Sullivan and Kociumbus, 2004)
Although the Bai Dat and Bai Go mineralisation has not yet
been directly linked to a particular intrusive, trace element data
(eg anomalous tellurium) is consistent with these deposits having
been formed by distal deposition from metal-rich fluids of
intrusive origin.
High-temperature Au-Ag-Te-Bi
In the northern sector of the Phuoc Son project area (at Khe Rin
and Khe Do, Figure 6), gold mineralisation occurs in a
significantly higher temperature environment than at Dak Sa. At
Khe Do prospect, Au-Ag-Te-Bi mineralisation associated with
semi-massive pyrrhotite has been drill-intersected directly
beneath a fractured, but unmineralised marble unit. In one hole
(KRDH23), brecciated quartz with pyrrhotite mineralisation,
assaying up to 33.7 g/t Au, is in contact with a fine-grained
granite dyke. Since the gold-bearing marble footwall shear zones
contain abundant tellurides, bismuth and some scheelite, whilst
the granite contains abundant disseminated pyrrhotite, a genetic
connection with magmatic fluids associated with granitic
intrusion is inferred.
Au-Ag-Cu-Mo skarn
At Khe Rin prospect, clinopyroxene skarn and epidotised
retrograde skarns occur which host erratic bonanza grades of
high-fineness gold. The skarns are located within 300 m of an
epizonal monzogranite stock of about 0.6 km2 in area. The gold
mineralisation is often intergrown with molybdenite, but is not
associated with high bismuth assays or tellurium (as at Khe Do).
The skarn zones, which occur near the base of a metabasite unit,
replace thin carbonate horizons, and are often proximally
associated with altered monzogranite to syenite dykes.
Intrusive-hosted gold
Gold mineralisation within the apical zone and wallrocks of a
small epizonal granite stock (Round Hill) is inferred from soil and
rock geochemistry 2 km SW of the Khe Rin skarn mineralisation
and 1.8 km WSW of the Khe Do High-Temperature Au-Ag-Te-Bi
mineralisation (Figure 6). Round Hill is at an early stage of
exploration. Deeply weathered, but silicified and veined surface
rocks assay up to 23 g/t gold, but results from shallow angle
diamond drill holes were unavailable at the time of writing.
Round Hill drillhole one (RHDH01), collared within wall
rocks, intersected 60.96 m of densely fractured and veined
microdiorite and amphibolitic hornfels. RHDH02 collared within
the granite stock, intersected variably silicified, fractured and
veined, medium-grained granite, quartz diorite porphyry dikes
and enclaves of metabasite. Core from both holes exhibited early
fine-grained secondary magnetite-silica flooding, overprinted by
102 Adelaide, SA, 19 - 22 September 2004
Gold grade (g/t) Contained gold (oz)
19.38 103 000
8.87 44 000
14.32 147 000
15.30 57 000
7.10 17 000
12.12 74 000
later albite, actinolite, sericite and epidote alteration. Quartz
veins in both drillholes contain coarse pyrite and rare scheelite,
the intervening fractured wall rocks contain disseminated and
fracture controlled pyrite and pyrrhotite.
Other areas of quartz-veined and silicified granitoid intrusives
are known within the central sector at Phuoc Son (K7 Bo
prospects). Stockwork quartz veined granite float occurring
within gold-in-soil anomalies suggests fertile potential for future
discovery of intrusive-hosted gold.
A diagrammatic composite west-east section of Phuoc Son
geology, summarising mineralisation styles and environments is
presented in Figure 7.
PROSPECTIVE DEVELOPMENT
The observed mineralisation styles within the property are
believed to all be intrusive-related and closely analogous to
world-class (+10 million oz) intrusive-related gold deposits
elsewhere (such as: Telfer, Granites/Tanami and Pine Creek
deposits in Australia, Vasilkovskoye in Kazakhstan, and Pogo
and Donlin Creek in North America).
Project-scale geochemical, geophysical and geological/
prospecting surveys are in progress to systematically evaluate
26 discrete prospect areas within the licences and to advance
those deemed to have the characteristics of a large-scale gold
system.
In June of 2003, the Company filed a Mining License
Application in respect of the Bai Dat and Bai Go deposits.
Engineering, metallurgical and environmental studies to
determine the feasibility of underground development with pilot
scale production at Bai Dat and Bai Go are continuing.
CONCLUSIONS
1. The Phuoc Son Tam Ky Suture has been a locus of deep
faulting, continental accretion, intensive intrusive activity
and mineralisation since Pre-Cambrian Times.
2. The Phuoc Son gold project, which comprises a 100 km2
section of the Phuoc Son Tam Ky Suture, contains an
anomalously high frequency of significant gold showings
(3/km2) and a variety of mineralisation styles.
3. The observed mineralisation styles wihin the Phuoc Son
property are believed to all be intrusive-related and
closely analogous to world-class (+10 million oz)
intrusive-related gold deposits elsewhere, such as: Telfer,
Granites/Tanami and Pine Creek deposits in Australia,
Vasilkovskoye in Kazakhstan, and Pogo and Donlin Creek
in North America. (A diagrammatic composite west-east
section of Phuoc Son geology, summarising mineralisation
styles and environments, is presented in Figure 7.)
PACRIM 2004
 GOLD EXPLORATION OF THE PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM

FIG 6 - Northern sector, geology and location of significant prospects.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 103

M J BANKS et al
FIG 7 - Diagrammatic NW-SE section, gold mineralisation environments.
4. The high incidence of intrusive activity and associated
mineralisation is related to the proximity and deep-seated
nature of the Phuoc Son Tam Ky Suture and related fault
systems.
REFERENCES
Cao Dinh Trieu, 1991. Deep Structures and Characteristics of Seismic
Activity on the territory of Viet Nam, pp 140.
Hutchison, C S, 1989. Geological Evolution of SE Asia (Clarendon Press:
Oxford).
Lan, Ching-Ying, et al. First evidence for Archaean continental crust in
northern Vietnam and its implications for crustal and tectonic
evolution in Southeast Asia, Geology, 29(3):219-222.
Lang, J and Baker, T, 2001. Intrusion-related gold systems: the present
level of understanding, Mineralium Deposita, 36:477-489.
Le Dzuy Bach and Ngo Gia Thang, 1995. Phanerozoic ophiolites in
Indochina, Journal Geology, Series B, 5-6:212.
104 Adelaide, SA, 19 - 22 September 2004
Lepvrier, C, Maluski, H, Nguyen Van Vuong, Roques, D, Axente, V and
Rangin, C, 1997. 40Ar/39Ar Indosinian Age of NW-trending dextral
shear zones within the Truong Son Belt (Vietnam). Cretaceous to
Cenozoic overprinting, Tectonophysics, 283:105-127.
Pham Khoan, Quach Van Gung and Bui Kim Thanh, 1995. The faults and
intrusion of Vietnams territory found according to the interpretation
of geophysical data, Journal Geology, Series B, 5-6:395.
Sullivan, J and Kociumbas, M, 2004. A technical review of the Phuoc
Son gold project in Quang Nam Province, Vietnam for Olympus
Pacific Minerals Inc, Watts, Griffiths and Mcquat Ltd, Internal report
for Olympus Pacific Minerals.
Tran Van Tri, Tran Kim Thach and Truong Cam Bao, 1979. Geology of
Viet Nam (Research Institute of Geology and Mineral Resources:
Hanoi).
Yang Zun Yi, Chen Yu Qi and Wang Hongzhen, 1986. The Geology of
China, Oxford Monographs on Geology and Geophysics 3 (Clarendon
Press: Oxford).
PACRIM 2004
Geology of the Gold Deposits of the Yanacocha District,
Northern Peru
P D Bell1, J G Gmez2, C E Loayza2 and R M Pinto2

ABSTRACT
The Yanacocha District is located 20 km north of the city of Cajamarca,
in the Northern Andean Orogenic belt of northern Peru. Gold production
may have occurred as early as Incan times (Paverd, 2000), but modern
exploration that began in the 1960s targeted only copper and silver.
Newmont Mining Corporation, BRGM and Compaa de Minas
Buenaventura began exploring for gold in 1983, with Newmont as
operator. Initial drilling took place in 1985. By 1992 a reserve of 1.28
million ounces was reported (Teal et al, 2002). Production began at the
Carachugo deposit in 1993. Current reserves are 31.7 million ounces at
$US 325 per ounce at a grade of 1.03 ppm (Newmont Annual Report,
2003). The operating company, Mineral Yanacocha SRL (Newmont
51 per cent, Buenaventura 44 per cent and International Finance
Corporation five per cent) produced 2.85 million ounces in 2003, making
Yanacocha one of the largest gold mining operations in the world.
The district is made up of a series of epithermal, high sulfidation style
gold deposits and one gold-rich gravel deposit (La Quinua), aligned in a
NE trend. La Quinua is not discussed in detail in this paper. The bedrock
deposits are hosted in a Miocene volcanic package, the Yanacocha
Volcanic Complex that also trends NE. Regional NW striking Andean
faults intersect the NE striking Trans-Andean Chicama-Yanacocha
structural corridor (Turner, 1997), localising the district mineralisation.
These two structural orientations dominate the district, controlling
breccia and intrusion locations, and gold mineralisation. Discontinuous
EW oriented fracture zones are interpreted as extensional and locally are
important in controlling gold mineralisation.
Three main volcanic depositional phases are recognised. The oldest,
the Lower Andesite, sits uncomformably on Cretaceous basement rocks.
Above this is a sequence of pyroclastic rocks, including a lower
crystal-rich package and an upper lithic-rich package. Overlying the
pyroclastic rocks are multiple andesite flows, domes and minor
pyroclastic rocks. The entire volcanic pile is intruded by multiple phases
of andesite dykes and dacite to quartz dacite plugs and dykes. The latter FIG 1 - Yanacocha District location map.
are associated with porphyry style gold-copper mineralisation in the deep
parts of some deposits. Phreatic and phreatomagmatic breccias cut the
volcanic rocks as pipes and dykes. Compaa de Minas Buenaventura 44 per cent and International
Multi-stage high sulfidation style alteration affects the entire rock Finance Corporation five per cent) produced 2.85 million ounces
package, with massive silicification at the centre of the deposits, grading in 2003, making Yanacocha one of the largest gold mining
out through vuggy and granular silica, advanced argillic, argillic, operations in the world.
propylitic and finally fresh rock. In places this alteration assemblage is The deposits are located in the Andes of northern Peru, 20 km
overprinted by later intermediate sulfidation creamy chalcedonic silica. north of the city of Cajamarca (Figure 1). Current mining is
Gold occurs in all rock types, commonly with massive, vuggy and entirely by open pit methods with heap leach recovery.
granular silica, and especially as high grade with the creamy silica. Gold Production in 2003 was from the Cerro Yanacocha deposit and
mineralisation is frequently localised around the margins of less
permeable breccia pipes and andesite dome roots. Copper is not currently
the La Quinua fluvioglacial deposit. La Quinua is not discussed
recovered at Yanacocha, but copper mineralisation occurs with gold in the in detail in this report.
deeper parts of most of the deposits, as indicated by a limited number of Yanacocha is described as a high-sulfidation epithermal
deep drillholes. These holes indicate a change from high sulfidation deposit occurring in pyroclastic rocks. The success of Yanacocha
epithermal style to porphyry style mineralisation at depth. has made this type of target among the most sought after by
exploration companies worldwide. The important characteristics
INTRODUCTION of the geology of Yanacocha are reviewed below, along with
some specifics of exploration techniques used in the district.
The Yanacocha District is one of the worlds premier gold
camps, combining large reserves with substantial production and
low costs. Current reserves are 31.7 million ounces at a grade of HISTORY
1.03 ppm, using at $US 325 per ounce gold price (Newmont Ancient workings in the Carachugo area at Yanacocha suggest
Annual Report, 2003). The operating company, Mineral that the area may have been exploited by the Incas for gold
Yanacocha SRL (Newmont Mining Corporation 51 per cent, (Paverd, 2000). The next reported work was in the 1960s
including a British technical aid mission that detected anomalous
1. Newmont Mining Corporation, Yanacocha Peru, C/O- Minera silver and lead in stream sediment samples. Limited information
Yanacocha, Suite L1, 1700 Lincoln Street, Denver CO 80203, USA. is available about the first drill program in 1970 by Nippon
Mining, targeting porphyry copper deposits. Neither physical
2. Minera Yanacocha SRL, Yanacocha Peru, Av Camino Real 348,
Torre El Pilar Piso 10, Lima 27, Peru. core nor drill data are available from this effort.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 105

P D BELL et al

The current phase of work at Yanacocha began with the
acquisition of mineral rights totalling 10 700 hectares by
CEDIMIN, the Peruvian arm of French Bureau de Recherches
Geologiques et Minieres (BRGM). The area covered most of
what are now the Yanacocha deposits. CEDIMIN conducted a
work program from 1981 to 1983, including photo-geology,
mapping, soil geochemistry, geophysics and trenching. The best
gold value was 14 ppm but most values were less than 0.3 ppm.
Silver values were as high as 1000 ppm. The area was first
viewed as an epithermal silver target.
Newmont became involved in 1984, along with partners
BRGM and Peruvian Compaa de Minas Buenaventura. Surface
work continued in 1984, with the first drilling in 1985. A total of
25 reverse circulation holes and ten diamond drill holes were
completed. Results were disappointing for both gold and silver. It
was only the recognition that specific styles of silicification are
associated with gold that led to a re-evaluation of targets and
successful drilling in 1986. The drilling outlined what is now the
Cerro Yanacocha deposit. Rock chip sampling identified the
surface expressions of the Carachugo, San Jose and Maqui
Maqui deposits.
Drilling continued even through very difficult political and
economic conditions in Peru in the late 1980s, and by 1989 a
resource of ten million tonnes of 1.4 ppm was indicated. The first
reserve, 1.28 million ounces, was declared in 1992. The first gold
was poured in August 1993 with production from the Carachugo
deposit.
Ongoing exploration has led to continuous expansion of
reserves and production since start-up. Current reserves are 31.7
million ounces with 2.85 million ounces produced in 2003
(Newmont Annual Report, 2003). Production, with the exception
of La Quinua, has been entirely from a series of hard rock
deposits with similar geologic characteristics.
REGIONAL GEOLOGY
Deposits of the Yanacocha District occur within the Northern
Andean Orogenic belt, a fold and thrust belt that runs the entire
length of Peru. Rocks within the belt are Mesozoic sedimentary
rocks and Tertiary volcanic rocks. The Yanacocha mineral belt is
located along a regional-scale disruption in this regional belt.
Northwest orientations of folds and thrusts in Cretaceous
sedimentary rocks are deflected to nearly EW along the
intersection of an ENE trans-Andean structural zone (Turner,
1997). This trans-Andean zone, known as the
Chicama-Yanacocha structural corridor, trends over a length of
about 200 km, beginning at the Pacific Coast. It is 30 to 40 km
wide, and defined by displacement of the Peruvian coastline,
multiple parallel N50E faults, and the ENE alignment of the
Yanacocha deposits (Quiroz, 1997).
The oldest rocks in the Cajamarca region are Cretaceous
sedimentary rocks (Figure 2). A basal siliciclastic package is
overlain by platform carbonate rocks. No Yanacocha high
sulfidation mineralisation is known in sedimentary rocks, but
many other deposit style prospects in the region are hosted in
these rocks.

Tantahuatay

Cerro Corona
9250000N Hualgayoc
La Zanja

Amaro

Perol

Sipn
9230000N
Chailhuagn
La Carpa
Galeno

YANACOCHA DISTRICT

Michiquillay

9210000N
Chamis Aurora Patricia
Cajamarca
760000E

780000E
740000E

800000E

0 10 km

Quaternary Hydrothermal alteration

Huambos volcanic rocks Faults

Llama/Porculla volcanic rocks Folds

(including Yanacocha volcanic complex)
High sulfidation deposit
Intrusive rocks
Cu-Au or Cu porphyry deposit
Cretaceous Sedimentary rocks

FIG 2 - Simplified regional geological map of the Cajamarca region. Note the occurrence of high-sulfidation epithermal and porphyry
deposits in the region.

106 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 GEOLOGY OF THE GOLD DEPOSITS OF THE YANACOCHA DISTRICT, NORTHERN PERU
The basal Tertiary volcanic rocks in the Cajamarca region are
lava flows, volcanic debris flow conglomerates and volcaniclastic
strata of the Llama Formation. In the Cajamarca region the
Llama Formation has been dated as Paleocene (Noble et al,
1990). Llama Formation rocks occur to the south of the district.
Above the Llama are volcanic rocks of the Yanacocha Volcanic
Complex, host for the Yanacocha deposits (Turner, 1997). These
rocks correlate with the regional Porculla Formation. The
Yanacocha Volcanic Complex is an interlayered sequence of
andesitic lava flows and pyroclastic rocks that overlie the Llama
Formation along a transitional contact. Ten kilometres NE of the
district the Yanacocha Volcanic Complex is overlain by a
regionally extensive andesitic to dacitic ignimbrite, the Huambos
Formation (Fraylones Member). This unit has been dated at 8.4
to 8.8 Ma (Turner, 1997).
These sedimentary and volcanic rocks are cut by Tertiary
intrusions described further in the lithology section (Rivera,
1980; Wilson, 1985). In general, intrusive bodies occur as
circular to elongate plugs along a WNW trending belt that cuts
the eastern margin of the district. Intrusions are dated as
Palaeocene to Miocene (Laughlin et al, 1968; Noble et al, 1990;
MacFarlane and Petersen, 1990).
FIG 3 - Generalised stratigraphic column of the Yanacocha District.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
LITHOLOGY
Sedimentary rocks
Strongly folded quartzite and minor siltstone of the Lower
Cretaceous Farrat Formation, part of the regional
Goyllarisquizga Group, are the oldest exposed rocks in the
Yanacocha District (Wilson, 1985). Farrat Formation rocks are
exposed in the southern and western parts of the district. Middle
Cretaceous Yumagual Formation, part of the regional Puilluicana
Group, exposed in the NE part of the Yanacocha District (Rivera,
1980; Wilson, 1985). Yumagual Formation rocks consist of
prominent grey limestone intercalated with poorly outcropping
limey shale and siltstone beds.
Sedimentary rocks have not been encountered to date at depth
within the deposits. Fragments of sedimentary rock, especially
quartzite, are common in pyroclastic units and in breccias in the
deep parts of the Cerro Quilish deposit.
Volcanic rocks
The volcanic rocks of the Yanacocha district are subdivided into
three sequences. These are termed the Lower Andesite Sequence,
the Yanacocha Pyroclastic Sequence and the Upper Andesite
Sequence (Figures 3 and 4).
107
P D BELL et al
FIG 4 - Yanacocha District generalised lithological map showing the location of gold deposits (1) Maqui Maqui, (2) Carachugo-Chaquicocha,
(3) Antonio, (4) San Jose, (5) Cerro Yanacocha, (6) La Quinua-Tapado-Corimayo, (7) Quilish, (8) Cerro Negro.
The Lower Andesite Sequence includes the Basal Hornblende
Andesite Lahar unit, the Lower Biotite-rich Tuff unit and the
Lower Pyroxene-Hornblende Andesite unit. The Basal
Hornblende Andesite Lahar unit lies unconformably on folded
Cretaceous sedimentary rocks in the southern, western and
northern parts of the district. The unit includes andesitic lahars
and debris flows dated at 19.53 0.13 Ma (Longo, in press). The
Lower Biotite-rich Tuff unit includes andesitic lahars, biotite-rich
tuffs and pyroclastic breccias dated at 15.5 Ma 0.06 Ma
(Longo, in press) and has been recognised in the southern and
northern parts of the district. Above the Biotite-rich Tuff is the
Lower Pyroxene-Hornblende Andesite unit that includes
andesitic lava flows, crystal lithic tuffs and lahars rich in
pyroxene and hornblende crystals. It has been dated at 13.2 14
Ma (Longo, in press). The unit also includes distal-facies
stratified lahar and mass-flow breccias with reworked tuff
interbeds. These lahar breccias contain poorly sorted,
sub-rounded andesite clasts in a lithified mud matrix. The Lower
Pyroxene-Hornblende Andesite unit outcrops in the west part of
the district including at Cerro Quilish, and sits directly on the
folded limestone basement in the NE of the district.
The central part of the Yanacocha district is dominated by the
Yanacocha Pyroclastic Sequence, which consists of the Fine Tuff
unit, the Transitional Eutaxitic (Teut) unit and the Upper Lithic
Tuff (Ult) unit. These units are too strongly altered to yield
reliable age dates.
The lowermost Fine Tuff unit (Tft) is discontinuous, confined
within depositional basins formed during post Lower Andesite
sequence faulting. The unit is highly variable in thickness and
morphology. It consists of locally laminated to wispy textured
fine tuff with a fine crystal component, fine crystal-lithic tuff and
reworked volcanic deposits with local laminated lacustrine
sediments. The best examples of this unit are within the
Chaquicocha Sur and Yanacocha Sur deposits.
108 Adelaide, SA, 19 - 22 September 2004
The Transitional Eutaxitic (Teut) unit consists of intercalated
poorly welded crystal, crystal-lithic and lapilli tuffs. The unit is
variably eutaxitic and locally crystal-rich. Basement clasts of
quartzite and argillite occur locally but are never abundant. The
unit overlies the Tft within interpreted palaeo-depositional
basins, and directly overlies the Lower Andesite sequence in the
adjacent horst blocks. The Teut unit is thickest at approximately
300 m within deep palaeo-grabens at the Yanacocha and
Chaquicocha deposits.
The Upper Lithic Tuff (Ult) unit consists of hornblende
andesite lithic-crystal tuff with abundant previously altered tuff
clasts. This unit has a maximum thickness of 250 m. The best
examples of this unit are at the San Jose, El Tapado and
Corimayo deposits. At El Tapado the Ult is 250 m thick in a
graben within the La Quinua basin.
The Upper Andesite Sequence consists of the Maqui Maqui
ignimbrite (MMI), the Upper Pyroxene Hornblende Andesite
(UphA) unit, and the Upper Crystal-Lithic Ignimbrite.
The Maqui Maqui ignimbrite has a distinctly eutaxitic texture
with abundant flattened lapilli fragments, quartzite fragments,
porphyritic fragments and very rare fragments of previously
silicified rock. There are also laminated rocks and associated
vent and phreatic breccias interbedded within the unit. The
Maqui Maqui ignimbrite has been dated at 12.4 - 12.7 Ma
(Longo, in press) and is restricted to the north and the east of the
district. The MMI unit hosts the Antonio Norte and Maqui
Maqui Norte deposits.
The Upper Pyroxene Hornblende Andesite (UphA) unit
consists of lahar breccias, block-and-ash flows with sub-rounded
clasts and andesitic domes with adjacent porphyritic hornblende
andesite lava flows. This unit is dated at 11.6 to 12.2 Ma (Longo,
in press).
PACRIM 2004
 GEOLOGY OF THE GOLD DEPOSITS OF THE YANACOCHA DISTRICT, NORTHERN PERU
The Upper Crystal-Lithic Ignimbrite is dated at 11.2 - 12.2 Ma
(Longo, in press). This ignimbrite is characterised by large,
flattened primary igneous fragments. A subunit of Upper
Crystal-Lithic Ignimbrite, the Sacsha Breccia ignimbrite,
blankets the southeastern part of the district.
Breccias
Volcanic rocks in the district are cut by numerous breccias of
varying styles. In many cases these are intimately associated with
gold mineralisation. Breccia types commonly occur together, and
changes from one style to another are frequently gradational.
The most prominent and distinctive breccia type contains
rounded to sub-angular fragments in a matrix rich in broken and
ground plagioclase crystals. Fragments are heterolithic and
commonly reflect adjacent wall rocks. These breccias occur as
steeply dipping dykes and steeply plunging pipes up to several
hundred metres in diameter. The breccias are interpreted as
phreatomagmatic in origin, and are locally termed diatremes.
Also important are breccias with sub-angular to rounded
fragments in a strongly abraded, homogeneous sand-sized rock
fragment matrix. These breccias vary from matrix dominated and
supported to fragment supported. The breccias are commonly
monolithic, but heterolithic examples exist, including those with
fragments of previously formed breccia. These breccias also
occur as steeply dipping dykes, as individual pipes and within
composite pipes with phreatomagmatic breccias. These rocks,
referred to as sandy matrix breccias, are interpreted as phreatic in
origin.
Less common are breccias with sub-angular to sub-rounded
fragments in a porphyritic matrix of andesitic to dacitic
composition. The fragments are normally monolithic, with rare
heterolithic examples. These breccias occur at Cerro Yanacocha
as subvertical pipes above dacitic intrusions and are the transition
between intrusive rock and phreatomagmatic breccias.
Breccias containing angular to sub-angular fragments in a fine
grained quartz matrix, flooded by Fe-oxides, chalcedony, alunite
or sulfidic matrix are commonly monolithic, with a fragment
composition reflecting the surrounding wall rock. The matrix
may display a fluidised texture. These rocks occur as dykes
cross-cutting almost all rock types. These are interpreted as
hydrothermal breccias.
Cross-cutting relationships are complex, with multiple
generations of brecciation in all deposits. Within individual pipes
phreatomagmatic breccia is cut by phreatic breccia, which is in
turn cut by hydrothermal breccia. Breccias commonly contain
fragments of previous breccia, indicating multiple generations of
activity.
Intrusive rocks
Intrusive rocks cut the sedimentary rocks and the overlying
volcanic pile. Recognition of many of these intrusions is only a
recent development as overprinted hydrothermal alteration
makes their identification difficult. Intrusive rocks are recognised
as important controls on mineralisation, and as source rocks for
gold-copper porphyry mineralisation. Intrusive rocks in the
district are broadly subdivided based on the presence of quartz.
Quartz-free and quartz-poor porphyritic intrusions, andesitic in
composition, occur as dykes, plugs and flow dome complexes.
Generally these rocks are porphyries, with up to 25 per cent
phenocrysts of plagioclase and hornblende (Loayza, 2002).
Plagioclase phenocrysts are commonly glomeroporphyritic.
Quartz constitutes less than one per cent of the rock and is
microscopic. A wide variety of grain sizes occur, with local
names applied including fine grained Cp (Carachugo Porphyry)
dykes and coarse grained plugs and dykes, sills and flows called
Yp (Yanacocha Porphyry). A sample of Yp intrusive rock from
Cerro Yanacocha yielded an age of 12.10 Ma (Noble et al, 1990).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Quartz-bearing intrusions cut many of the deposits, including
Cerro Yanacocha and El Tapado. These intrusions are
quartz-feldspar porphyritic rocks, generally with a very fine
matrix. They consist of approximately 25 per cent phenocrysts,
including quartz (five to eight per cent), plagioclase (ten to 15
per cent), biotite and hornblende (Loayza, 2002). Overall quartz
content varies from trace to 20 per cent. Multiple phases of these
intrusive rocks are recognised, especially at Cerro Yanacocha
where phases rich in copper and in quartz veins are cut by
unmineralised phases. Quartz-bearing intrusive rocks are locally
referred to as Ypq (Yanacocha quartz porphyry) and dacite. A
sample of Ypq from Yanacocha Norte returned an age of 9.90
0.05 Ma (Turner, 1997).
The youngest intrusions have an andesitic composition, dark
green colour and with prominent fresh feldspar, hornblende and
biotite phenocrysts in a fine grained matrix. Quartz content is
highly variable, from zero to 18 per cent (Loayza, 2002). These
rocks occur as dykes, often cutting Yp, and are termed Ypb for
the presence of biotite. Turner (1997) dated a Ypb dyke from
Yanacocha Sur at 8.40 0.06 Ma.
STRUCTURE
Rocks in the Yanacocha district are cut by a dominant
N40-60E, N30-60W and E-W fault and fracture structural
set. The NE and EW structures are extensional, occurring as
open fractures and normal faults. NW striking faults are both
compressional and extensional, occurring as strike-slip shear
structures and normal faults. Frequently NE trending zones of
extension occur between parallel NW striking fault zones.
Locally there are secondary tensional NS striking fracture and
fault zones. Structures have experienced multiple reactivation
events.
Features of the NE structural trend include the alignment of
gold deposits, breccias and shallow-level intrusive rocks in the
district, the geometry and elongation of the alteration system and
a dominant joint fabric. Additionally, moderate to strong
magnetic anomalies have the same NE alignment.
Evidence of NW trending structures includes a pattern of horst
and graben blocks. The most prominent feature of this is the NW
trending Quaternary La Quinua Basin. Here secondary
fluvioglacial gravel, some containing gold, buried the hard rock
Corimayo and El Tapado deposits. The gold rich gravel forms the
La Quinua deposit. NW striking block faulting has offset
hydrothermal alteration systems, which occur at structural
elevations varying from 4200 m to 3300 m. The Chaquicocha,
Maqui Maqui, Cerro Yanacocha and Carachugo deposits show
control by NW faults and fractures (Gmez and Veliz, 2002).
Segmented EW and NS striking fracture zones are interpreted
as extensional and control high grade pods of mineralisation at
the Corimayo, Chaquicocha Sur, Cerro Negro Este and Cerro
Quilish deposits. Less well developed NS striking faults localise
mineralisation in Yanacocha Norte and Cerro Quilish Norte
deposits.
Structural intersections of NE and NW corridors are
interpreted to have formed important dilational channel ways for
ascending hydrothermal fluids, with gold deposits localised at
these intersections. This control is clearly visible at the
Corimayo, Chaquicocha and El Tapado deposits.
ALTERATION
Epithermal high-sulfidation style hydrothermal alteration in the
Yanacocha District is well studied, with predictable zoning
patterns (Figure 5). Complications arise from telescoped multiple
stages of high sulfidation alteration in some deposits, as well as
from the superposition of high sulfidation, gold copper porphyry
style alteration and later intermediate sulfidation alteration.
Alteration and mineralisation at Yanacocha are pervasive,
109
P D BELL et al
FIG 5 - Yanacocha District generalised alteration map.
affecting nearly every rock type in the district. Age dating
suggests continuous alteration and mineralisation for more than
five million years (Longo, in press).
Gold copper porphyry style alteration is present in the deeper
parts of many deposits. The known district porphyry deposits,
including the Kupfertal deposit, have distinct zoned alteration. At
Kupfertal the lithocap of a shallow copper-gold porphyry system
is exposed. Here microcrystalline to dense finely granular silica
and amorphous patches of pyrophyllite and/or alunite completely
obscure original protolith textures. This assemblage grades
downward into a stockwork of wormy textured quartz-sulfide
veins, changing to A-veins (Gustafson and Hunt, 1975) at further
depth, within an interpreted phyllic shell. Weak potassic
alteration with secondary biotite and chalcopyrite occurs in the
deepest studied portion of the system, 450 m below surface
(Pinto, 2002).
Epithermal high sulfidation alteration is similar in most
deposits in the district. Intense massive silica alteration, closely
associated with gold mineralisation, forms the core of each of the
systems. Massive silica alteration grades outward into a strongly
acid leached zone of vuggy and granular silica. The latter is
commonly texture destructive. Beyond the leached facies there is
advanced argillic alteration, including zones of alunite, clay and
weak silica, and this is normally the limit of economic grade
gold mineralisation. Advanced argillic zones grade outward to
strong clay rich argillic alteration zones, then on to propylitically
altered and fresh rock. Opaline silica frequently occurs close to
the surface, on the margins of alteration cells.
The scale of alteration zoning is highly variable, with strong
lithologic and elevation control on facies distribution causing sub
horizontal alteration zone geometry. Alteration zoning may occur
over kilometres horizontally, whereas in some areas, such as
Corimayo, strong massive silica alteration occurs only metres
vertically below fresh rock. Dykes and breccia bodies commonly
110 Adelaide, SA, 19 - 22 September 2004
are fresher than more porous surrounding pyroclastic rocks,
resulting in local argillic, propylitic and even fresh zones within
large silicified bodies.
Overprinting the high sulfidation alteration is an assemblage of
cream coloured chalcedonic silica, with or without well
crystallised barite in vugs. The cream colour is distinctive, and
due to the presence of Ti-oxides within the chalcedony
(Canchaya, 2002). Creamy silica is interpreted to be a result of
alteration by intermediate sulfidation state fluids. Recent
relogging of holes suggests that this style of alteration occurs in
all known deposits.
MINERALISATION
Typical of epithermal high sulfidation systems, the main
mineralisation of the Yanacocha deposits is localised in the
silicified core facies described above. At depth mineralisation is
usually related to higher temperature advanced argillic alteration
and potassic alteration that suggests proximity to gold copper
porphyry systems.
Several stages of mineralisation have been identified in the
Yanacocha District. The most important stages include: Stage 1,
a low-grade gold event with development of gold copper
porphyry systems at depth; Stage 2, the main gold-(copper)
stage; Stage 3, a late high-grade gold event; Stage 4, a late
copper-(gold) stage; and Stage 5, a late carbonate-sulfide stage.
Stage 1: The low-grade event is characterised by an pervasive
silicification, contemporaneous with the deposition of fine
disseminated pyrite and low-grade (less than 0.2 ppm) gold
(Harvey et al, 1999). At deeper levels this stage includes the
development of patchy textured silicification, grading to wormy
and A type veinlets, some banded, suggesting a transition from a
high sulfidation to a copper gold porphyry system (Pinto, 2002).
Fluid inclusion data, including temperatures that range from
PACRIM 2004
 GEOLOGY OF THE GOLD DEPOSITS OF THE YANACOCHA DISTRICT, NORTHERN PERU
200 to 500C and salinities higher than 43 per cent in some
samples support this interpretation (Reynolds, 1999; Thompson,
2002; Loayza, 2002). Secondary biotite from potassic alteration
at the Kupfertal porphyry copper prospect, using Ar39/Ar40,
yielded an age of 10.72 0.09 Ma (Longo, in press).
Stage 2: The main gold event post-dates the pervasive
silicification. Mineralisation is characterised by fine pyrite with
minor enargite and covellite. Sulfides occur as disseminations
and void and fracture fillings. In the oxidised portion of the
deposits mineralisation includes the presence of hydrothermal
breccias. Gold in this stage occurs as submicron grains usually
closely associated with Fe-oxides (Bersch, 1999; Turner, 1997).
Stage 3: A high-grade (greater than 1 ppm) gold event is
recognised by the occurrence of coarse gold associated with
blocky barite or by cross-cutting creamy chalcedonic silica. The
creamy silica cross-cuts previously silicified pyroclastic rocks,
phreatic breccias and occurs as the matrix in some hydrothermal
breccias. Stage 3 style mineralisation is occurs in all deposits,
and is especially important at the Chaquicocha Alta, El Tapado
and Corimayo deposits.
Stage 4: Late copper-(gold) mineralisation is closely
associated with dacitic intrusive rocks and phreatomagmatic
breccias. It is characterised by presence of enargite, covellite and
pyrite with advanced argillic silica-alunite alteration at shallow
levels and pyrophyllite-diaspore alteration at depth. Alunite
related to this stage yielded a radiometric age of 9.12 0.32 Ma
(Longo, in press). This stage is recognised at the Cerro
Yanacocha deposit.
Stage 5: Represented by sparsely distributed veinlets of
rhodochrosite-dolomite and base metal sulfides, is interpreted as
representing a transition from acidic fluids to a more neutral pH
fluid. This suggests as a local change in the sulfidation state of
the system. This latest stage has been observed at the Cerro
Yanacocha deposit.
Mineralisation controls vary from one deposit to another, but
most include structural and lithological controls, including dome
margins and multiphase breccias (Figure 7). At the district scale
the location of deposits is controlled by NE and NW structural
intersections. At deposit scale the main structural controls are the
NE, NW and extensional EW faults. Structural zones of
weakness controlled the emplacement of multiple generations of
breccias and intrusive rocks along NE and NW trends. These
multiple events are associated with multiple stages of alteration
and gold mineralisation.
Lithologic control is very important in most deposits.
Mineralisation occurs mainly in favourable pyroclastic rocks.
These more porous and permeable rocks localised hydrothermal
fluids that produced alteration and mineralisation. Examples of
this type of control occur at the San Jose, Corimayo, Cerro
Yanacocha and Antonio Norte deposits.
Dome and diatreme margins control the location of gold,
especially high grade (greater than 1 ppm), in many deposits. An
example of this is at the Yanacocha Sur deposit where the highest
gold grades are at the contact of the favourable pyroclastic rocks
(Ult) with a clay-altered andesitic intrusive. This setting is
duplicated at the Corimayo deposit where the high-grade gold
mineralisation is at the contact between strongly silicified
pyroclastic rock and both an argillic altered phreatomagmatic
pipe and an argillic altered to fresh andesitic dome. The
interpretation is that the barrier formed by impermeable rock
promoted local fluid flow changes that favoured the precipitation
of gold.
GEOCHEMISTRY
Geochemical sampling and the study of geochemical zonation
has been integral to Yanacocha exploration from early in the
program. A 1985 rock chip sampling program for gold-silver
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
identified the first gold anomaly clusters that were later identified
as the Carachugo, Maqui Maqui, San Jos, Cerro Yanacocha,
Cerro Quilish and Cerro Negro deposits. These anomalies occur
over silicified oxidised rocks. Programs in the 1990s improved
understanding of the relationship between gold and trace
elements, as extensive surface and drillhole data was collected
and analysed.
On the surface, where the deposits crop out, there is a positive
correlation between gold and silver, arsenic, barium, bismuth and
mercury. Frequently however, gold deposits are buried, covered
with a leached cap of granular silica. In such cases gold is
depleted at the surface but enrichments of mercury, antimony and
arsenic occur. The geochemical pattern over capping andesite or
dacite in a flow dome is more complicated. Patterns of gold,
silver, arsenic, bismuth, mercury, lead and antimony anomalies
occur but with much lower values than in exposed pyroclastic
rock. The centre of a dome may show no anomalous values
(Warscheid et al, 2002). Recognition of alteration style and
geological setting is important in interpretation and application
of geochemistry in exploration.
Vertical element zonation is apparent within the deposits
(Figure 6). A surface oxidised zone of barium, bismuth and
silver, with mercury, arsenic and antimony changes to copper and
arsenic with some lead, Zn, and silver in the sulfide zones
(Gmez, 2002). At the deepest studied levels Mo becomes more
significant. This zonation pattern is influenced by supergene
effects, especially remobilisation of copper in the oxide zone.
Further complication is added by the overprinting of multiple
events, especially at the Cerro Yanacocha deposit.
GEOPHYSICS
Many geophysical techniques are employed at Yanacocha. The
most useful approach is to look for silicified rocks, usually
related to gold mineralisation with resistivity methods (Goldie
and Rutti, 2002). High resistivity zones correlate with
silicification related to gold mineralisation. Resistivity data are
obtained from induced polarisation (IP) pole-dipole, time domain
electromagnetic (TDEM) and controlled source audio
magnetotelluric (CSAMT) surveys.
IP surveys were the first resistivity method applied in the
district and are very useful for mapping near-surface resistors.
Additionally, chargeability data from this method provide an idea
of the depth of the oxide/sulfide boundary. Limited vertical
penetration is the main disadvantage of this method, along with
an inability to resolve multiple resistive layers. Despite these
shortcomings the method is useful. A 100 m resistivity IP depth
slice compilation map matches almost all outcropping deposits in
the district, with the highs corresponding to the deposit
footprints.
As exploration has moved to deeper targets, other resistivity
methods have been used to locate deeper (+200 m depth)
resistors. TDEM and CSAMT are now widely used in the
district. TDEM is an inexpensive reconnaissance method. It can
map covered resistors and resolve multiple layers. The
disadvantages of this method include poor definition near the
surface and a lack of chargeability data. CSAMT provides better
definition near surface and better depth of investigation. The
disadvantages include high cost, slow surveys and problems with
local noise.
A successful example of the application of the techniques
mentioned above is the discovery of the Corimayo deposit, a
blind deposit covered by gravels and fresh dacitic lava flows. A
TDEM resistivity anomaly at depth, reflecting silicified
pyroclastic rock, led to the discovery of this deposit below fresh
rock.
111
P D BELL et al

FIG 6 - Sketch showing geochemical patterns in some deposits at the Yanacocha District.

FIG 7 - Mineralisation controls at deposit examples in the Yanacocha District.

112 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 GEOLOGY OF THE GOLD DEPOSITS OF THE YANACOCHA DISTRICT, NORTHERN PERU
CONCLUSIONS
The Yanacocha District is one of the premier gold districts of the
world. Total gold content is high; the known deposits sit close to
the surface and metallurgy of the oxidised ore is straightforward.
The District setting, at the intersection of regional structural
trends, with favourable porous pyroclastic rocks, is not unique.
The most striking features are the pervasive nature of the
alteration and mineralisation, affecting nearly every rock type,
the complexity of the breccias and the long life, more than five
million years, of the alteration system (Longo, in press). It is this
long lived system that is critical to the enormous gold
accumulation of the District.
Rapid reserve and production growth in some cases outpaced
understanding of very complex and challenging geology. Recent
advances have greatly aided understanding of the geology,
assisting both exploration and mine geology.
In February 2004 the 15 millionth ounce of gold was poured at
Yanacocha. Combined with end of year reserves and resources of
35 million ounces the district has a known oxide gold
endowment of 50 million ounces. Current drill indicated sulfide
mineralisation has the potential to substantially add to this.
ACKNOWLEDGEMENTS
We gratefully acknowledge the management of Minera
Yanacocha SRL and the parent partners, Newmont Mining
Corporation and Compaa de Minas Buenaventura for their
support and for permission to publish this paper. In particular we
thank Lewis Teal for initiating the effort. Additional thanks to
Brian Arkell and Colin Price for geologic discussion, support and
editing. This paper reports ideas developed by many Yanacocha
geologists and we thank them for their contributions. Rossana
Cabrera is thanked for her assistance with the figures.
REFERENCES
Bersch, M, 1999. Yanacocha high grade study, Minera Yanacocha
internal report, Cajamarca, Peru, 157 p.
Canchaya, S, 2002. Estudio microscpico de siete muestras, Minera
Yanacocha internal report, Cajamarca, Per, 41 p.
Goldie, M and Rutti, M, 2002. Applications of resistivity methods in gold
exploration at Yanacocha, Newmont internal report, 52 p.
Gmez, J, 2002. Corimayo, depsito aurfero de alta ley en un sistema
epitermal tipo cido sulfato. Distrito minero de Yanacocha,
Cajamarca, Per Tesis para ttulo profesional de Ingeniero Gelogo,
Universidad Nacional de Ingeniera, Lima, Per, pp 6975.
Gmez, J and Veliz, J, 2002. Control estructural de los depsitos en el
distrito Minero de Yanacocha, norte del Per. XI Congreso Peruano
de Geologa, Lima, Per, 7 p.
Gustafson, L B and Hunt, J P, 1975. The porphyry copper deposit at El
Salvador, Chile, Economic Geology, 70:857-912.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Harvey, B, Myers, S and Klein, T, 1999. Yanacocha Gold District,
Northern Peru, in Proceedings PACRIM 99, pp 445-459 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
Laughlin, A, Damon, P and Watson, B, 1968. Potassium-argon dates from
Toquepala and Michiquillay, Peru, Economic Geology, 63:166-168.
Loayza, C, 2002. Geologic study of Cerro Yanacocha gold-silver deposit,
Yanacocha district, northern Peru, Masters thesis, 94 p, University of
Nevada, Reno, USA.
Longo, T, in press. Volcanic stratigraphy and the temporal relationship of
volcanism to gold and copper mineralization, Yanacocha Mining
district, Peru, PhD thesis, Oregon State University.
Macfarlane, A W and Petersen, U, 1990. Pb isotopes of the Hualgayoc
area, northern Peru, Economic Geology, 85:1303-1327.
Moore, S and Saderholm, E, 2002. Yanacocha district stratigraphic
summary, Minera Yanacocha internal report, Cajamarca, Peru, 54 p.
Noble, D C, Mckee, E H, Mourier, T and Megard, F, 1990. Cenozoic
stratigraphy, magmatic activity, compressive deformation and uplift
in northern Peru, Geological Society of America Bulletin,
102:1105-1113.
Paverd, A, 2000. The Discovery of the Yanacocha Gold Deposit, Segundo
Simposium Internacional del Oro, Lima -Per, 11 p.
Pinto, R, 2002. Transicin de un sistema de alta sulfuracin a un sistema
porfirtico de alto nivel en Kupfertal, distrito minero de Yanacocha,
Cajamarca, Per: Tesis para ttulo profesional de Ingeniero Gelogo,
Universidad Nacional Mayor de San Marcos, Lima, Per, 96 p.
Quiroz, A, 1997. El corredor estructural Chicama Yanacocha y su
importancia en la metalogenia en el norte del Per; in Resmenes
extendidos IX Congreso Peruano de Geologa, pp 149-154,
(Sociedad Geolgica del Per: Lima).
Reynolds, J, 2000, Unpublished memorandum about fluid inclusion
studies in Kupfertal and other areas in Yanacocha, Cajamarca, Peru,
1 p.
Rivera, L, 1980. Mapa geolgico del cuadrngulo de Cajamarca. Sector
Energa y Minas, Instituto Geolgico Minero y Metalrgico,
Repblica del Per, Boletn N 31, 67 p.
Teal, L, Harvey, B, Williams, C and Goldie, M, 2002. Geologic overview
of the Yanacocha district gold deposits, northern Peru, Society of
Economic Geologists; Global Exploration 2002, Integrated methods
for discovery, 2002, abstracts pp 43-44.
Thompson, T B, 2002, Kupfertal project mineralogical and fluid
inclusion notes, Minera Yanacocha internal report, Cajamarca, Peru,
1 p.
Turner, S, 1997. The Yanacocha epithermal gold deposits, northern Peru:
high sulfidation mineralization in a flow dome setting, PhD Thesis,
Colorado School of Mines, Colorado, USA, 342 p.
Warscheid, W, Espinoza, J, Flores, E and Trujillo, J, 2002, Geochemical
characterization study for intrusive dome margins, Montura Grid,
Chaquicocha Sur, Minera Yanacocha internal report, Cajamarca,
Peru, 41 p.
Wilson, J, 1985. Mapa geolgico del cuadrngulo de Cajamarca. Sector
Energa y Minas, Instituto Geolgico Minero y Metalrgico,
Repblica del Per. Boletn N 38, 104 p.
113
Common Geological Characteristics of Prominent Hill and
Olympic Dam Implications for Iron Oxide Copper-Gold
Exploration Models
A Belperio1 and H Freeman1
ABSTRACT
The Prominent Hill copper-gold deposit is located approximately 150 km
north west of Olympic Dam, on the southern margin of the Mount Woods
Inlier in South Australia. Its discovery in November 2001 resulted from
the application of a geological model developed from geological,
geophysical and petrophysical data pertaining to the Olympic Dam
deposit. The discovery has dispelled the notion the Olympic Dam may be
unique, and that another deposit like it is unlikely to be found.
The Prominent Hill mineralisation is hosted by haematite-matrix
hydrothermal breccias, which have suffered extensive iron-sericite-silica
alteration and overprinting. Copper, gold, barium, fluorine, uranium,
cerium and lanthanum are all intrinsically associated with the
hydrothermal iron. Breccias intrude an east-west striking, fault-controlled
sedimentary succession, which is flanked to both the south and north by
volcanic complexes. Many similarities between Prominent Hill and
Olympic Dam are documented including geological setting, mineralogy,
breccia styles, and alteration. Differences relate largely to the nature of
the immediate host rocks.
The similarities and difference provide key data for improving
exploration models for this style of iron oxide copper-gold deposit, both
within the Gawler Craton as well as globally.
INTRODUCTION
Since the discovery of the super-giant Olympic Dam orebody in
1975, iron oxide copper-gold (IOCG) deposits have been
attractive exploration targets for mineral exploration companies
worldwide. Numerous exploration models have been used, based
primarily on a spectrum of associated iron ranging from
haematite to magnetite end members. A number of IOCG
discoveries have resulted (Haynes, 2002), but the discovery of
another haematite-dominant end member of this family has had
limited success and the suggestion that Olympic Dam was
somehow unique has been increasingly invoked.
The exploration model that lead to the discovery of Prominent
Hill by Minotaur Resources was specifically targeted at the
haematite end member of the IOCG family of deposits (Belperio,
2001), where deposit characteristics indicate higher grades and
greater potential for world-class reserve status viz:
Olympic Dam (haematite system) 717 Mt @ 1.7% Cu,
0.6 g/t Au, 3.6 g/t Ag,
0.5 kg/t U308
Candelaria (magnetite-haematite) 470 Mt @ 0.95% Cu,
0.2 g/t Au, 3.1 g/t Ag
Salobo (magnetite system) 450 Mt @ 1.15% Cu,
0.5 g/t Au
Although a resource is yet to be announced for Prominent Hill,
the extent and tenor of mineralisation recorded to date with
significant down-hole intercepts such as 209 m @ 1.54 per cent
Cu and 0.93 g/t Au (DP005), 57 m @ 7.7 g/t Au (DP010) and
41 m @ 6.06 per cent Cu and 0.31 g/t Au (RP017) (Minotaur
Resources, 2002, 2003) underscores the significance of the
discovery.
1. FAusIMM, Chief Geologist, Minotaur Resources Ltd, 247 Greenhill
Road, Dulwich SA 5065.
1. Minotaur Resources Ltd, 247 Greenhill Road, Dulwich SA 5065.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
The Prominent Hill discovery was a vindication of both the
exploration strategy and models utilised by Minotaur.
Confirmation that Prominent Hill is directly comparable in its
style and genetic history to Olympic Dam would dispel the
notion that Olympic Dam is a unique or singular occurrence.
More importantly the deposit can provide key comparative
information to that available from Olympic Dam on the
geological setting, timing, processes, mineralogy and alteration,
thus allowing significant refinement of global IOCG exploration
models.
REGIONAL GEOLOGICAL SETTING
The Gawler Craton is an Archaean to Mesoproterozoic terrane
occupying 600 000 square kilometres of southern Australia
(Figure 1). Outcrop is limited and geotectonic interpretations rely
heavily on drillhole and geophysical data sets (Daly, Fanning and
Fairclough, 1998). The Craton hosts the world-class Olympic
Dam iron oxide Cu-Au orebody, the newly discovered Prominent
Hill deposit, and a number of other sub-economic discoveries of
IOCG mineralisation (Acropolis, Wirrda Well, Gunson,
Manxman). Each of these deposits has been related to the same
major thermal event (the Hiltaba Suite Gawler Range Volcanics
magmatic event) that affected the central and eastern part of the
Gawler Craton from about 1600 - 1580 Ma (Flint, 1993; Daly,
Fanning and Fairclough, 1998; Ferris, Schwarz and Heithersay,
2002).
Three major areas of IOCG alteration and mineralisation are
recognised along the eastern part of the Gawler Craton; the
Olympic Dam area, the Moonta-Wallaroo district, and the Mt
Woods Inlier. The three areas are interpreted to represent separate
footprints of crustal-scale thermal anomalies and include both
magnetite-biotite and haematite-sericite alteration assemblages
(Skirrow et al, 2002) formed at different temperatures and
pressures (crustal levels).
KEY CHARACTERISTICS OF PROMINENT HILL
Details of the Prominent Hill deposit have been progressively
released since November 2001 through Company ASX
announcements, annual reports, conference presentations and
publications (Minotaur Resources, 2002, 2003; Carter, Belperio
and Freeman, 2003; Hart and Freeman, 2003).
Geophysically, the deposit is characterised by a discrete gravity
anomaly (Figure 2) the size of which varies depending on survey
station spacing and the methodology used to remove regional
gradients. Its size is approximately 2.5 mgals (at 1 km station
spacing), increasing to 4.5 mgals (400 200 m spacing) and
6 mgals (100 100 m spacing). The source of an adjacent intense
magnetic anomaly is a package of magnetite- chlorite-tremolite-
phlogopite altered metasomatic rocks with skarn affinities, and
intercalated highly altered, porphyritic intermediate intrusive and
chlorite-matrix tectonic breccias. The gravity response reflects the
overall package of hydrothermal iron altered rocks including
mineralised haematite-matrix breccias. The mineralised portion of
the breccias is not directly detectable by gravity or by other
geophysical techniques (Hart and Freeman, 2003), being masked
115
A BELPERIO and H FREEMAN

FIG 1 - Tectonic map of the Gawler Craton, South Australia (simplified after Daly, Fanning and Faiclough, 1998),
showing location of Prominent Hill and Olympic Dam.

116 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 COMMON GEOLOGICAL CHARACTERISTICS OF PROMINENT HILL AND OLYMPIC DAM

A B

FIG 2 - Geophysical responses of the Prominent Hill breccia complex. (A) Residual gravity; (B) 1VD gravity;
(C) 1VD gravity contours on TMI image with traces of diamond drillholes in yellow.

by complex dense host rock geology and other alteration features
(sediments, volcanics, faulted skarn blocks). Northwest- and
northeast- trending crustal penetrating structures are visible in
magnetic and filtered gravity imagery in close proximity to
Prominent Hill, and lower order equivalent structural control is
evident in grade variations at the deposit level. Hiltaba-age
bimodal igneous intrusions and sub-volcanic activity are present in
close proximity to the deposit.
Mineralisation comprises copper-gold-silver-uranium-cerium
-lanthanum intimately associated with haematite-matrix breccias.
The deposit occurs below ~100 m of cover within a broader
breccia complex that intrudes a sequence of low metamorphic
grade mafic to felsic volcanics (c.1590 Ma Lower Gawler Range
Volcanics basalt, andesite, rhyolite) and associated
volcaniclastic sediments (Mentor Formation). The volcanics and
volcaniclastics occupy an arcuate graben (ENE to ESE) that was
subsequently subjected to intense and pervasive hydrothermal
iron alteration, replacement and brecciation (Figure 3). The
breccias display a spectrum from clast-supported simple jig-saw
breccia types to matrix-supported multiple brecciated,
heterolithic varieties. Drilling to the north of the main prospect
area has intersected lower amphibolite to granulite facies
metamorphic rocks more typical of the main Mt Woods Block.
The centre of the linear ENE-trending gravity anomaly
coincides with a large area of massive, barren haematite silica
flooded volcanics that records a strong IP chargeability anomaly.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 117

A BELPERIO and H FREEMAN
This is flanked to the north by an area totalling approximately
2 0.5 km of haematite-rich breccia sheets. Layering within the
breccias suggests that sheets are sub-vertical, strike ENE to ESE,
and were emplaced along structural and lithological contacts. A
halo of lesser haematite alteration and host-rock brecciation
extends up to a further 0.5 km away. Iron alteration is almost
exclusively haematitic. Magnetite occurs within late dolerite
dykes (Gairdner Dykes, ca. 800 Ma). Sericite, silica, fluorite,
barite and carbonate are common, either as breccia matrix
components with haematite, or as late stage veins with or without
haematite. The haematite breccias and associated alteration and
mineralisation are believed to have developed through repetitive
hydrothermal venting, with fluids emanating from the same
linear volcanic axis that generated the host volcanics and graben.
Lithologies
Key lithologies include mafic to intermediate volcanic units and
intrusive dykes, volcaniclastic sediments, dolomitic carbonates,
hydrothermal breccias, intensely iron-silica altered variants
(steely haematite), and lower intensity earthy haematite altered
to in situ crackle-veined variants (Figure 3).
FIG 3 - Geological interpretation of the Prominent Hill breccia complex.
118 Adelaide, SA, 19 - 22 September 2004
Volcanics of mafic, intermediate and acid composition (basalt,
basaltic andesite, andesite, dacite) are ubiquitous and form the
base of a volcano-sedimentary succession. Coarser grained
varieties may represent sub-volcanic dykes or sills. Late-stage
fine-grained volcanic dykes (andesitic) are present within the
breccia complex. Fragmental volcanic lithologies include
haematitic agglomerates, conglomerates, lapilli tuffs and
pepperites. The volcanics exhibit numerous low-temperature
alteration assemblages involving albite, sericite, haematite (after
magnetite, martite), chlorite and carbonate. More felsic volcanic
lithologies (dacite to rhyodacite) occur at the western end of the
prospect.
A package of low metamorphic grade sediments occupies the
ENE to ESE trending graben affected by the hydrothermal event.
A south to north succession of volcanics greywacke
argillite dolomite (footwall to hangingwall) is observed across
the deposit. Together with the volcanics, the sediments define the
original host graben. Sandstone and greywacke units, in contact
with volcanics to the south, thin significantly in the western
domain, replaced by broader sequences of argillite and dolomite.
PACRIM 2004
 COMMON GEOLOGICAL CHARACTERISTICS OF PROMINENT HILL AND OLYMPIC DAM
Dolomitic units are Ca-Mg- carbonate-rich, poorly bedded to
massive and white, red-grey or tan in colour. They exhibit
haematite (-chlorite) crackle-veining and variable silicification,
sericite alteration and quartz veining. Argillites are generally fine
grained, pale red-brown sericite-quartz-rich lithologies. They are
well laminated throughout, although commonly exhibit
disruption of original layering, and tensional fracturing and
veining by haematite and sericite (Figure 5h). Other textures are
indicative of slumping, suggesting that the sediments were only
partially lithified during alteration/mineralisation. This also
indicates deposition during an evolving rifting/volcanic event
and associated volcano-sedimentary graben development.
Greywacke units are generally poorly sorted, poorly bedded, fine
to coarse grained, sericitic and quartzose. They are very common
as clasts within the mineralised haematite breccias.
Haematite breccias are generally confined to the sequence of
greywacke and sandstone and the volcanics to the immediate
south. They are highly variable in composition and texture but
invariably contain haematite as a matrix component. Matrix
varieties include dark grey, blue-grey, purple, red-brown,
jet-black or steel grey haematite. Quartz, copper sulfides
(chalcocite, bornite, chalcopyrite), carbonate and fluorite are also
common matrix components. The haematite breccias can be
subdivided further based on variations in grain size of the matrix
components (in particular haematite).
Disseminated copper mineralisation is generally most
abundant in (dark grey to black) crystalline matrix haematite
breccia. They are high-energy fluidised breccias characterised
by a relatively coarse grained haematite (-sericite-quartz-
fluorite-carbonate) matrix and commonly exhibit a wide variety
of clast types, including greywacke, altered andesite to basalt,
haematite breccia and haematite rock of unknown precursor.
Textures vary from milled to compositional layered. Embayment
of haematised or sericitised clasts is common. They have a high
matrix percentage and are correspondingly high in Fe, REE, Co,
U, Cu and Au.
Earthy haematite matrix breccias consist of a fine-grained,
granular, dark brown to red haematite matrix. They are
interpreted to reflect a relatively low-temperature, highly
oxidised regime in comparison to those breccias with a coarse,
crystalline haematite matrix. Compositional layering is less
common. They lack high contents of Cu-sulfides and examples
of layered breccia with alternating granular earthy and dark grey
crystalline haematite matrix show a decrease in sulfide
mineralisation within the earthy haematite layers. Free (visible)
gold and native copper may be present in earthy matrix breccias
west of the main copper mineralised zone. They have higher clast
component (lower matrix percentage) and correspondingly lower
levels of Fe, REE and U.
Steely haematite replacement of breccia occurs primarily
within the eastern part of the deposit (coincident with the centre
of the eastern gravity anomaly, Figure 2C), but also occurs
sporadically within or adjacent to breccia sheets. These zones are
characterised by a dense, non-porous matrix of very fine-grained
steel grey haematite and silica with densities in excess of 4.0
gcm-3 and an intense hardness attributed to the ultra fine
silicification. Limited Cu-sulfide intersected in these breccias to
date consists of coarse blebs of bornite disseminated within the
matrix. As at Olympic Dam, high-grade gold mineralisation is
more common in these silicified zones. The steely nature is due
to extremely fine-grained haematite intergrown with
cryptocrystalline silica. Clasts consist of chert and intensely
silicified rock fragments of unknown precursor. Geochemically
the steely haematite zone is distinctive and high in Fe (>40 per
cent), As, Ba and REE, and very low in Co, Cu, K, Mg, Na, Ni,
Ti, Zn and Y.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Steely haematite replacement of volcanics occurs along an
ENE axis that coincides with a linear gravity shoulder (Figure
2C) that corresponds with the southern margin of the graben
(Figure 3). Original volcanic textures are sporadically preserved
within this iron-silica alteration zone, indicating both primary
and brecciated volcanic rocks (basalt to andesite precursor)
containing possible sedimentary fragments (shale or siltstone),
sediments, pseudo-breccias and true breccias. These are intensely
overprinted by steely-grey massive haematite veins (to a few
metres thickness), post-dated again by specular haematite and
barite veinlets (generally <30 mm thickness). Only minor sulfide
is present (pyrite and chalcopyrite) usually within fine (<1 mm)
late-stage veinlets, but a zone of enhanced gold mineralisation a
few tens of metres thick is commonly encountered around the
outer margin of the alteration envelope.
Alteration and veining
Metasomatitic rock assemblages north of the main mineralised
zone are varied but include abundant chlorite, magnetite,
phlogopite, actinolite-tremolite, pyrite (up to 20 per cent),
serpentine, talc, albite, former scapolite, sericite, and haematite
alteration. The metasomatites regularly contain low-grade copper
(~0.1 per cent Cu as chalcopyrite) but may attain higher grades
where density of late stage carbonate- (quartz-chalcopyrite-
haematite-fluorite) veins is higher. Chlorite is abundant within
this zone as a matrix component of tectonic breccias and in
altered intermediate intrusive. Chlorite alteration also occurs
within the footwall volcanics along with haematite and sericite,
but its abundance decreases towards the centre of the deposit.
The main alteration products associated with Cu-Au
mineralisation at Prominent Hill are haematite, silica/quartz and
sericite. The zone affected by intense haematite alteration is
characterised by extreme sodium depletion and absence of
magnetite. Haematite is ubiquitous within Cu and Au
mineralised zones. Martite (former magnetite) grains are
common on the peripheries of the deposit.
Sericite is most common within the sequence of greywacke
and argillite host rocks where it is a co-dominant constituent with
quartz. It occurs in lesser concentrations with albite, carbonate,
chlorite and haematite in the footwall volcanics, occurring as
infill in vesicles and replacement in phenocrysts after feldspar as
well as the groundmass. It also constitutes a major component of
some mineralised breccias, both within clasts and within a
haematitic matrix. Some volcanic dykes and/or sills within the
breccia complex have undergone intense sericitisation.
Late stage vein types are widely varied in morphology and
mineralogical composition. Veining is generally lower frequency
in Cu-mineralised haematite breccias compared to host units.
Common vein constituents include: carbonate (calcite and
dolomite), copper sulfide, barite, quartz, fluorite and specular
haematite. Carbonates at Prominent Hill are Ca and Mg rich
whereas the most common carbonate at Olympic Dam is siderite.
Chalcocite, bornite and chalcopyrite all occur as discreet late
vein phases, commonly with haematite or carbonate.
Mineralisation
High grades of mixed Cu-Au mineralisation occur almost
exclusively within the breccias where copper and gold occur as
granules within the haematite-dominated matrix. Gold, in
addition to its widespread association with copper in the
haematite breccias, also occurs in higher-grade zones around the
haematite-silica core, and in haematite crackle-veined haloes in
dolomite, argillite and greywacke host rocks adjacent to breccia
(Figure 4). The mineralised breccias reach the top of bedrock in
the eastern part of the deposit, but to the west increasing
interbeds of host sediments and lower grades of mineralisation
are encountered.
119
A BELPERIO and H FREEMAN

0 500 Gold at Prominent Hill is present in greater proportion than

N Olympic Dam, forming about 30 per cent of the in-ground value
Scale (m)
at PH compared with ten per cent at OD, and is found in three
M a gn distinct settings (Figure 4). The main hydrothermal breccia body
etic D
o main
contains a consistent level of gold over hundreds of metres of
Increasing Breccia F2 drill intercepts at the 0.2 to 2.0 gram/tonne level. Gold is also
volcano-sedimentary enhanced along the margin of intensive Fe-Si alteration that
cover
? ? forms the ironstone footwall to the Prominent Hill mineralised
o re
ic a c breccias. High gold levels have been intersected in
te-sil F1
he m a ti crackle-veined dolomitic sediments in a halo above and around
e ly
S te
the copper mineralised breccias. This style of gold-only
S A: Plan View
mineralisation awaits detailed investigation and drill follow-up.
Gold grains are present in a broad range of grain sizes, from
grains of 800 microns readily observable by eye, down to five
microns or less detected by electron microscopy. Preliminary
B: Section View petrological and SEM observations of the gold grains indicate a
S N persistent association of gold, copper and iron oxide, implying a
0 100 200 300 400 500 common genetic association. In particular, gold grains are
Cover sequence rimmed with a copper rind, and contain minute copper inclusions
-100
(Figure 7). As a consequence, gold floats readily in standard
Volcano- sedimentary
sequence copper floatation tests with extremely high recovery-
Barren to-concentrate rates.
Hematite
Silica Magnetic
Domain
-300
core Ch
al
KEY CHARACTERISTICS OF OLYMPIC DAM
co Au Vein
ci Network Key elements of the Olympic Dam deposit presented by Cross et
te
al (1993), Esdale et al (2003), Gow et al (1994), Haynes et al
(1995), Oreskes and Einaudi (1990, 1992), Reeve (1990a,b),
High grade Reeve et al (1990) and Reynolds (2000, 2001) are summarised
Gold Ch
al c chalcocite
-500
Skin op below.
yr
ite Olympic Dam contains ore reserves in excess of 700 Mt
Main averaging 1.7 per cent Cu, 0.6 g/t Au, 3.6 g/t Ag and 0.5 kg/t
F1 Mineralised F2 U3O8 intimately associated with massive hydrothermal haematite
Breccia
Complex alteration and brecciation. The 2600 Mt global resource also
contains approximately 26 per cent Fe and 0.5 per cent Rare
-700 ?
? Earth Elements (REE), principally cerium and lanthanum, that
? ? ? are not recovered. The deposit occurs below 300 m of cover,
within the Olympic Dam Breccia Complex, that is in turn hosted
by the Roxby Downs Granite (Figure 8). Associated with the
FIG 4 - Schematic geological model for Prominent Hill.
breccia complex are largely contemporaneous, mafic and felsic
dykes, volcaniclastics, ash fall tuffs and associated high crustal
level, near-vent sediments. U-Pb zircon ages on volcaniclastic
At the broadest level, copper mineralisation displays a lateral tuffs and dykes associated with mineralised breccias are
and vertical zonation from chalcocite in the upper and more indistinguishable from the host granite intrusive and the Gawler
central portions of the breccia sheets to bornite, chalcopyrite and Range Volcanic extrusive ages (c. 1590 Ma).
pyrite with depth (Figure 5). In detail, significant variations Geophysically, the deposit is characterised by a large discrete
occur between discrete breccia bodies that are juxtaposed and bouguer gravity and residual gravity anomaly of approximately
variability of grade is also associated with metre-scale 17 mgals (500 100 m grid spacing). A magnetic anomaly
hydrothermal flow banding. A large part of the breccia system, sourced from deeper below the deposit (c. 4 km) may represent a
though highly anomalous in Cu, U, Au and REE, does not carry mafic intrusive complex. The gravity response directly reflects
economic grades of mineralisation. Chalcopyrite (-uraninite, the hydrothermal haematite brecciation and alteration of the
-pyrite) mineralised breccias are volumetrically the most granitic host rocks, and the overprinted hydrothermal system is
significant in terms of mineralisation intersected to date but in turn directly related to the volcanogenic vent setting.
generally of lower grade than chalcocite or bornite dominated Significant northwest- and northeast- trending crustal penetrating
intervals. structures occur in the vicinity of the deposit, and lower order
Uranium occurs in discrete parts of the system, often at deeper equivalent structures clearly control deposit geometry and grade
levels associated with fluorite and chalcopyrite, as small grains (Figure 8). High-level Hiltaba granite (c. 1588 Ma) and
of coffinite and uraninite. In these zones, U concentrations may associated sub-volcanic activity are coeval (within dating limits)
reach several thousand ppm, although through much of the with hydrothermal alteration and mineralisation.
breccias, the concentration is less than 100 ppm. This is a A central haematite-silica-barite core barren of mineralisation is
significant difference compared with Olympic Dam where surrounded by an area of approximately 3 by 3.5 km of
Uranium is present in greater proportion, about 30 per cent of the well-developed haematite-matrix breccias. A halo of variably
in-ground value compared to ten per cent at PH, and reportedly altered and brecciated granite (clast-supported breccia) extends up
occurs as pitchblende in the chalcocite-bornite zone and at the to a further 3 km out where it grades in to less altered granite. In
bornite-chalcopyrite interface. Cerium and lanthanum are present detail, the deposit is composed of a large number of irregular,
as Ce-apatite and Ce- and La- monazite and oxides (Figure 6). discrete breccia bodies. Breccias display a spectrum from
The main mineralised components (Cu, Au, Ur) all show an clast-supported simple jig-saw breccia types to matrix-supported
intimate relationship with haematite, with which they are heterolithic varieties. Repetitive lithification and rebrecciation
believed to be genetically coeval. have created a complex intermix of variably altered host and

120 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 COMMON GEOLOGICAL CHARACTERISTICS OF PROMINENT HILL AND OLYMPIC DAM

A B

C D

E F

G H

FIG 5 - Thin sections of main alteration and breccia styles at Prominent Hill.
(A) Sericitised and silicified corroded volcanic clasts in auriferous earthy haematite breccia (DP002, 200.60 m).
(B) Volcanic rock pervasively altered by dense, steely haematite (DP002, 440.5 m).
(C) Chalcocite mineralisation within bluish-grey haematite-silica matrix supported breccia (DP003, 463.5 m).
(D) Chalcopyrite mineralisation and compositional layering within dark grey haematite breccia (DP003, 543.2 m).
(E) Andesite with abundant aligned feldspar phenocrysts and late qz-cb veining (DP004, 188.5 m).
(F) Agglomerate with a variety of volcanic clasts in an Fe-rich volcanic (?) matrix and fine, white plagioclase
phenocrysts (DP004, 273.9 m).
(G) Altered quartz diorite (DP007, 122.7 m).
(H) Tensional fracturing within argillite (DP006, 413.9 m).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 121

A BELPERIO and H FREEMAN
FIG 6 - Backscattered electron micrographs of breccia
components, Prominent Hill.
breccia fragments intermixed with haematite matrix and a variety
of copper minerals largely as clasts and mineralised matrix. An IP
chargeability anomaly is recorded over the WSW to WNW
trending Fe-Si flooded volcanics and steely haematite-matrix
breccias.
Alteration is dominantly sericite and haematite with lesser
silica, chlorite, carbonate and magnetite alteration, and intensity
is correlated with degree of brecciation. Early magnetite
alteration has been largely overprinted by haematite. Silica
alteration dominates the haematite-quartz core where barite is
also enhanced. The iron rich mineralising fluids are believed to
have a significant mantle component as well as a superimposed
shallow crustal meteoric component (Johnson and McCulloch,
1995; Campbell et al, 1998).
122 Adelaide, SA, 19 - 22 September 2004
TABLE 1
Summary of key features Prominent Hill and Olympic Dam IOCG
deposits.
Common features
Regional association with high level Hiltaba/GRV magmatic event
Regional association with intersection of northwest- and
northeast-trending crustal structures
Proximal association with volcanic venting and associated
sedimentation
Major gravity anomaly directly reflects major zone of iron alteration
and local palaeotopographic highs
Magnetic anomalism in the vicinity may reflect mafic intrusives and
regional magnetite alteration
Lack of magnetic anomalism of the deposit reflects dominance of
haematite
Haematite, sericite, silica and chlorite alteration haloes (chlorite minor
or absent in high Cu zones)
Barium, fluorine, cerium, lanthanum, uranium, copper and gold
anomalous haloes
Hydrothermal breccia (jigsaw clast breccia to poly phase heterolithic
breccias)
Lateral and vertical zonation of sulfides (pyrite-chalcopyrite-bornite
-chalcocite) reflects primary redox control
Central zone of Fe-Si-Ba alteration devoid of mineralisation
IP chargeability anomaly associated with the Fe-Si-Ba barren core
Enhanced gold mineralisation around margin of Fe-Si-Ba alteration
front
Enhanced gold in close proximity to chalcocite-bornite zones
Copper, gold and uranium occur mostly as free granules dispersed
within haematite matrix
Gold grains have coating of chalcopyrite and inclusions of haematite
and chalcopyrite
Polymict mineralised inclusions reflect synchronicity and reworking of
mineralised grains
Significant differences
Main host rocks are granite at OD and volcanics at PH
Breccia clast dominantly sericitised and haematised granite at OD and
volcanics at PH
Fe carbonates at OD, Ca and Mg carbonates dominant at PH
Dextral structural style interpreted at OD, PH interpreted as sinistral jog
zone
Uranium mostly occurs as pitchblende in upper parts of
chalcocite-bornite zone at OD
Uranium mostly occurs as sporadic grains of coffinite and uraninite in
chalcopyrite zone at PH
Uranium comprises about 30 per cent of in-ground mineralisation at
OD, about ten per cent at PH
Gold comprises about ten per cent of in-ground mineralisation at OD,
about 30 per cent at PH
Sulfide mineralisation displays a broad lateral and vertical
zonation from chalcocite in the upper and more central portions
of the breccia complex to bornite, chalcopyrite and pyrite with
depth and lateral extent. Ore zones form only a small proportion
of the breccia complex, though anomalous Cu, U, Au and REE
occur throughout. Copper minerals occur mostly as disseminated
grains within the breccia matrix. Gold, in addition to its
widespread association with copper in the breccia, also occurs in
higher-grade zones around the haematite-silica core and in the
upper parts of the chalcocite-bornite zone. Mineralisation is
intimately associated, and considered contemporaneous with,
introduction of iron.
PACRIM 2004
 COMMON GEOLOGICAL CHARACTERISTICS OF PROMINENT HILL AND OLYMPIC DAM

FIG 7 - Photomicrographs of gold grains from Prominent Hill drillhole DP010 showing inter-relationships between haematite (hm),
chalcopyrite (cp), gold (Au) and carbonate (cb).

Strong structural control is evident in the locus of high grade
mineralised zones. Both NE-SW and NW-SE trending structures
are present, with individual breccia bodies having a NW trend
within a WNW envelope.
COMPARISONS BETWEEN PROMINENT HILL
AND OLYMPIC DAM IMPLICATIONS FOR
EXPLORATION
Exploration for Olympic Dam-style deposits on the Gawler in
the 1980s and 1990s was based on limited published data,
incomplete geological and geophysical information and poorly
defined models (eg Roberts and Hudson, 1983; Rutter and
Esdale, 1985). With additional information now available for
Olympic Dam (Reynolds, 2000, 2001; Esdale et al, 2003), high
quality regional data sets over the Gawler Craton from the South
Australian Geological Survey, a better understanding of
mineralising fluids (Skirrow et al, 2002) and a significant body
of new information released for the Prominent Hill deposit,
major improvements are being made in practical aspects of
exploration for these types of deposits. Most exploration
companies have reviewed and updated their key exploration
parameters in the light of the Prominent Hill data.
Table 1 compares key exploration criteria for each of Olympic
Dam and Prominent Hill. In addition to dispelling the notion that
Olympic Dam is somehow a unique or singular occurrence, the
information arising from the Prominent Hill highlights the key
common criteria required for establishing more rigorous
exploration models. In particular, the comparative data highlight
a common association with near-vent volcanism and associated
hydrothermal alteration and brecciation. They also highlight very
similar alteration and mineralisation assemblages, proximal and
distal haloes, and common zonation patterns.
Significant differences relate to the nature of the host rocks
and the relative proportions of Cu: Au:U that may be expected in
the mineralising system. Significant refinement of model-based
exploration strategies is being implemented by Companies
exploring for IOCG deposits. The remarkable similarities in ore
mineralogy, zonation and paragenesis between Olympic Dam
and Prominent Hill, and the similar tectonic setting, indicate
these two deposits formed by very similar genetic processes.
Together they define the highly prized, haematite-dominated end
member of the iron oxide copper gold family of deposits.
Key pointers for explorers are the relationship with bimodal
extrusives, the association with regional crustal structures, the
common presence of intercalated sediments with volcanics,
indicating high structural levels, and the dominant iron alteration
event that the mineralisation is associated with.
Other defining features of this deposit type that can aid
exploration include the extreme Na depletion associated with the
haematite alteration event, the absence of magnetite in the
mineralised zone, and alteration vectors involving chlorite, silica,
sericite and haematite.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 123

A BELPERIO and H FREEMAN

Limit of granite-rich and

haematite-rich breccias

Resource outline
Ore zones

0 1000

METRES
201035-009

FIG 8 - Geological interpretation of the Olympic Dam breccia complex (Reynolds, 2001).

124 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 COMMON GEOLOGICAL CHARACTERISTICS OF PROMINENT HILL AND OLYMPIC DAM
ACKNOWLEDGEMENTS
Data on Prominent Hill was collected as part of extensive
exploration activities undertaken by Minotaur Resources Limited
from 2001 to 2003, in partnership variously with Billiton
Exploration Australia, BHP Billiton, Normandy Exploration and
Oxiana Ltd. Numerous people from these companies assisted
with data collection and interpretation. Petrological assistance
and discussion were provided by Ian Pontifex, Kathy Ehrig and
BHP Newcastle Laboratories. Geophysical input was provided
by John Hart and geological interpretations by consultants
Richard Flint and Leigh Rankin.
REFERENCES
Belperio, A P, 2001. The Minotaur BHP Billiton alliance and the search
for the next Olympic Dam orebody in South Australia, Australian
Journal of Mining, extended abstracts of the 2nd Annual Copper
Forecast Conference, Sydney, August 2001.
Campbell, I H, Compston, D M, Richards, J P, Johnson, J P and Kent, A J
R, 1998. Review of the application of isotopic studies to the genesis
of Cu-Au mineralisation at Olympic Dam and Au mineralisation at
Porgera, the Tennant Creek district and Yilgarn Craton, Australian
Journal of Earth Sciences, 45:201-218.
Carter, D, Belperio, T and Freeman, H, 2003. The discovery of the
Prominent Hill copper-gold deposit, South Australia, in Proceedings
NewGenGold 2003 Case Histories of Discovery, pp 102-116.
Cross, K C, Daly, S J and Flint, R B, 1993. Olympic Dam deposit, in The
Geology of South Australia, Vol 1, The Precambrian. Geological
Survey of South Australia, Bulletin 54 (Eds: J F Drexel, W V Preiss
and A J Parker).
Daly, S J, Fanning, C M and Fairclough, M C, 1998. Tectonic evolution
and exploration potential of the Gawler Craton, South Australia,
AGSO Journal of Australian Geology and Geophysics,
17(3):145-168.
Esdale, D, Pridmore, D F, Coggon, J, Muir, P, Williams, P and Fritz, F,
2003. The Olympic Dam copper-uranium-gold-silver-REE deposit,
South Australia, in Geophysical Signatures of South Australian
Mineral Deposits, Special Publication 12 (Ed: M C Dentith), pp
147-168 (Australian Society of Exploration Geophysicists).
Ferris, G M, Schwarz, M P and Heithersay, P, 2002. The geological
framework, distribution and controls of Fe-oxide Cu-Au
mineralisation in the Gawler Craton, South Australia. Part I:
Geological and tectonic framework, in Hydrothermal Iron Oxide
Copper-Gold and Related Deposits: A Global Perspective, Vol 2 (Ed:
T M Porter), pp 9-31 (PGC Publishing: Adelaide).
Flint, R B, 1993. Mesoproterozoic, in The Geology of South Australia,
Vol 1, The Precambrian. Geological Survey of South Australia,
Bulletin 54 (Eds: J F Drexel, W V Preiss and A J Parker).
Gow, P A, Wall, V J, Oliver, N H S and Valenta, R K, 1994. Proterozoic
iron oxide (Cu-U-Au-REE) deposits: Further evidence of
hydrothermal origins, Geology, 22:633-636.
Hart, J and Freeman, H, 2003. Geophysics of the Prominent Hill
prospect, South Australia, in Geophysical Signatures of South
Australian Mineral Deposits, Special Publication 12 (Ed: M C
Dentith), pp 93-100 (Australian Society of Exploration
Geophysicists).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Haynes, D W, 2002. Giant iron oxide copper-gold deposits: Are they in
distinctive geological settings? in Giant Ore Deposits:
Characteristics, Genesis and Exploration, CODES Special
Publication 4 (Eds: D R Cooke and J Pongratz) (University of
Tasmania).
Haynes, D W, Cross, K C, Bills, R T and Reed, M H, 1995. Olympic
Dam ore genesis: A fluid mixing model, Economic Geology,
90:63-74.
Johnson, J P and McCulloch, M T, 1995. Sources of mineralising fluids
for the Olympic Dam deposit (South Australia): Sm-Nd isotopic
constraints, Chemical Geology, 121:177-199.
Minotaur Resources, Annual Reports 2002, 2003. Adelaide.
Oreskes, N and Einaudi, M T, 1990. Origin of rare earth element-enriched
hematite breccias at the Olympic Dam Cu-U-Au-Ag deposit, Roxby
Downs, South Australia, Economic Geology, 85:1-28.
Oreskes, N and Einaudi, M T, 1992. Origin of hydrothermal fluids at
Olympic Dam: Preliminary results from fluid inclusions and stable
isotopes, Economic Geology, 87:64-90.
Reeve, J S, 1990a. The discovery and evaluation of the Olympic Dam
deposit, in Geological Aspects of the Discovery of Important
Minerals in Australia (Eds: K R Glasson, A H Blaikie and J T
Woodcock), pp 125-133 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Reeve, J S, 1990b. The discovery and evaluation of gold mineralisation
within the Olympic Dam deposit, in Geological Aspects of the
Discovery of Important Minerals in Australia (Eds: K R Glasson, A
H Blaikie and J T Woodcock), pp 57-58 (The Australasian Institute
of Mining and Metallurgy: Melbourne).
Reeve, J S, Cross, K C, Smith, R N and Oreskes, N, 1990. Olympic Dam
copper-uranium-gold-silver deposit, in Geology of the Mineral
Deposits of Australia and Papua New Guinea, Monograph 14 (Ed: F
E Hughes), pp 1009-1035 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Reynolds, L, 2000. Geology of the Olympic Dam Cu-U-Au-Ag-REE
Deposit, in 2000 Hydrothermal Iron Oxide Copper-Gold and
Related Deposits: A Global Perspective, (Ed: T M Porter) Vol 1, pp
93-104 (PGC Publishing: Adelaide).
Reynolds, L, 2001. Geology of the Olympic Dam Cu-U-Au-Ag-REE
Deposit, MESA Journal, 23:4-11.
Roberts, D E and Hudson, G R T, 1983. The Olympic Dam
copper-uranium-gold-silver deposit, Roxby Downs, South Australia,
Economic Geology, 78:799-822.
Rutter, H and Esdale, D J, 1985. The geophysics of the Olympic Dam
discovery, Exploration Geophysics, 16:273-276.
Skirrow, R G, Bastrakov, E, Davidson, G, Raymond, O L and Heithersay,
P, 2002. The geological framework, distribution and controls of
Fe-oxide Cu-Au mineralisation in the Gawler Craton, South
Australia. Part II: Alteration and Mineralisation, in Hydrothermal
Iron Oxide Copper-Gold and Related Deposits: A Global
Perspective, (Ed: T M Porter), Vol 2, pp 33-47 (PGC Publishing:
Adelaide).
125
The Sams Creek Peralkaline Granite Hosted Gold Deposit,
Northwest Nelson, New Zealand A New Variant on Alkaline
Intrusion-Related Gold Deposits
R L Brathwaite1 and K Faure2
ABSTRACT
At Sams Creek a peralkaline granite porphyry dyke hosts gold-bearing
sulfide-quartz-siderite veins. The granite dyke intrudes
Ordovician-Silurian metapelite and quartzite, and has thin lamprophyre
dykes along its contacts. The mineralised veins and the granite and
lamprophyre dykes have been deformed during several phases of folding,
the youngest of which is Early Cretaceous. The granite has been
hydrothermally altered, with early magnetite-sideritebiotite alteration
overprinted by sulfide and quartz-siderite veins and quartz-perthite-albite-
siderite-pyrite-rutilesericite alteration assemblages. The sulfide veins
are composed of arsenopyrite + pyrite gold galena sphalerite
chalcopyrite pyrrhotite graphite, with minor quartz and siderite. The
gold contains significant silver (up to 30.8 wt per cent) and ranges in
fineness from 692 to 844. Immiscible liquid-rich fluid inclusions, in
quartz protected from recrystallisation by enclosing arsenopyrite, contain
carbonic liquid that homogenised at temperatures of 320 - 355C.
The Sams Creek gold deposit is distinguished from the orogenic gold
deposit type by its stockwork veining, magnetite-sideritebiotite
alteration, high sulfide content, ore mineralogy and the absence of
mineralisation in the metapelite country rocks. Features of the Sams
Creek gold deposit do not fit neatly into existing classifications of
granite-related gold-bearing deposits. The peralkaline composition of the
Sams Creek granite and the absence of molybdenite and paucity of
chalcopyrite, are major points of difference with gold-rich porphyry
copper deposits. The presence of a low sulfidation-state ore mineral
assemblage, carbonate alteration and carbonic fluid inclusions at Sams
Creek are typical of many of the reduced granitic intrusion-related type of
gold deposit. However, notable differences are:
1. the reduced type granites are metaluminous and calc-alkaline; and
2. Sams Creek only has high As in contrast to elevated Bi, W, As, Mo,
Te, and Sb in the reduced type.
The Sams Creek deposit also shows affinities with gold deposits
associated with alkaline magmatism, and it appears to be a new variant of
this type as an As-Au deposit associated with a peralkaline granite and
lamprophyre.
INTRODUCTION
Gold mineralisation at Sams Creek, 25 km south of Takaka in
northwest Nelson, was discovered in 1974 by GW Patterson of
CRA Exploration. The mineralisation was subsequently
investigated by CRA Exploration (Hawke, 1982, 1984;
Rosengren, 1985; Clemenston, 1987), and is currently being
explored by Oceana Gold Ltd (formerly GRD Macraes Ltd)
(www.oceanagold.com). The gold is associated with
arsenopyrite-pyrite-quartz-siderite mineralisation in a peralkaline
granite porphyry dyke, which intrudes lower Palaeozoic
metasedimentary rocks. The first published record of the dyke
noted that it was a recrystallised riebeckite microgranite (Shelley,
1984). Windle and Craw (1991) considered that the granite dyke
had been buckled and dismembered by folding, but they
concluded that the mineralising fluids were of metamorphic
1. MAusIMM, Institute of Geological and Nuclear Sciences Ltd, Rafter
Laboratory, PO Box 31-312, Lower Hutt, New Zealand. E-mail:
b.brathwaite@gns.cri.nz
2. Institute of Geological and Nuclear Sciences Ltd, Rafter Laboratory,
PO Box 31-312, Lower Hutt, New Zealand.
E-mail: k.faure@gns.cri.nz
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
origin. Tulloch (1992) described the mineralogy and chemistry of
the granite as peralkaline and noted that the granite was altered
by late and/or postmagmatic fluids. Osborne (1997) suggested
that the mineralisation was produced by late stage magmatic
hydrothermal fluids. The mineralisation is confined to the granite
and the ore mineralogy, hydrothermal alteration and stable
isotope compositions of the vein minerals are consistent with a
magmatic-hydrothermal origin (Faure, Brathwaite and de Ronde,
2003).
Here we describe the Sams Creek deposit as a granite-hosted
gold-sulfide deposit which, although it has been deformed and
largely recrystallised shows many features that are similar to the
reduced granite related (eg Thompson and Newberry, 2000; Lang
and Baker, 2001) and alkaline intrusion related gold deposit
types (eg Jensen and Barton, 2000).
REGIONAL SETTING
The Sams Creek area lies within the Eastern Sedimentary Belt of
the Takaka terrane (Figure 1), a belt of Lower Palaeozoic
metasedimentary rocks (Grindley, 1980; Cooper, 1989;
Rattenbury, Cooper and Johnston, 1998). This terrane was
deformed and metamorphosed under lower greenschist facies
conditions prior to intrusion of the Late Devonian (367 Ma)
Rameka Complex (Cooper, 1989). The terrane was subsequently
intruded by granite and granodiorite of the Early Cretaceous
Separation Point Suite (Tulloch, 1983; Muir et al, 1995). Three
phases of deformation have been identified in the region: a first
phase of recumbent folding (F1), a second of inclined folds (F2)
about subhorizontal north-south trending axes, and a third (F3) of
localised northeast to southeast trending folds (Grindley, 1980;
Jongens, 1997). Bradshaw (2000) has shown the Early Devonian
Baton Formation is affected by the F1 and F2 folding, and
175E
TASMAN SEA
TASMAN BAY
Takaka
AF
SS
41S
SS
DF
TT
BT
KB TT Sams Creek
10 km Fig 2
FIG 1 - Regional location map of Sams Creek area, west Nelson.
Buller Terrane (BT), Takaka Terrane (TT), Karamea Granite
Batholith (KB), Separation Point Suite Granitoids (SS), Riwaka
Complex (RC), Anatoki Fault (AF), Devil River Fault (DRF). Inset
shows location of Figure 2.
127
R L BRATHWAITE and K FAURE
therefore at least the F1 folding is Mid Devonian. Radiometric
dating indicates that a northeast-trending regional schistosity in
the area northwest of Takaka, which is correlated with the third
phase of deformation, was associated with the emplacement of
the Separation Point Batholith in the Early Cretaceous (Challis et
al, 1995). The region has also been affected by late Tertiary and
Quaternary folding and faulting, as indicated by deformation of
mid-Tertiary sedimentary rocks, which unconformably overlie
the older rocks.
DEPOSIT GEOLOGY
Sams Creek is a deeply incised tributary of the Takaka River on
the eastern side of the Lockett Range. The peralkaline granite
porphyry, which hosts the gold-sulfide mineralisation, intrudes
Ordovician-Silurian metapelite, metasandstone and quartzite of
the Wangapeka Formation and Hailes Quartzite (Grindley, 1980;
Rattenbury, Cooper and Johnston, 1998). The tabular
morphology of the granite porphyry body, with a thickness of
30 to 40 m and extent of at least 7 km (Figure 2), together with
its discordance to bedding in the metasedimentary rocks indicate
that it was emplaced as a dyke.
The metasedimentary rocks have been folded and regionally
metamorphosed to a lower greenschist facies assemblage of
quartz-albite-muscovite-chlorite. Shelley (1984) described a
recumbent fold (F1) at Sams Creek which has been refolded by
northerly-trending F2 folds. Hickey (1986) mapped the F2 folds
as being inclined with subhorizontal axes and also identified F3
folding about steeply plunging axes. A prominent crenulation
cleavage is axial plane to the F2 folds (Shelley, 1984; Hickey,
1986; Windle and Craw, 1991), and this cleavage also extends
into hydrothermally altered lamprophyre, which occurs along the
margins of the granite dyke. In the Anvil Zone (Figure 2),
Windle and Craw (1991) mapped deflection of the cleavage
around the margins of granite bodies that have been
dismembered by deformation. This together the sinuous outcrop
pattern of the granite and the presence of extensive shearing
FIG 2 - Geological map of the Sams Creek area, Upper Takaka. Modified from Tulloch (1992) and based on mapping by CRA Exploration
(Clementson, 1987) and Hickey (1986).
128 Adelaide, SA, 19 - 22 September 2004
within the dyke and along many of its contacts indicate that it has
been deformed and folded.
The unaltered granite porphyry is composed of scattered
phenocrysts of perthite (0.5 - 5 mm in length), arfvedsonite,
aegerine and quartz (up to 4 mm) in a fine grained (~0.2 mm)
hypidiomorphic granular aggregate of perthite, quartz,
plagioclase, arfvedsonite and aegerine. Myrmekitic intergrowths
of quartz and plagioclase are common. Ilmenite, rutile and
fluorite are accessory minerals. The arfvedsonite is partly
replaced by rims of riebeckite. The sodic mineralogy and
geochemistry, with low Al2O3 and Sr, coupled with high values
of Zr, Nb, Ga and Y, are characteristic of a peralkaline (Na2O +
K2O>Al2O3) A-type granite (Tulloch, 1992). The absence of
primary magnetite and a low magnetic susceptibility (30 SI units
10-5) in the least altered granite classifies it as a reduced, low
oxidation-state granite (Thompson and Newberry, 2000).
Thin (0.2 - 3.0 m) dykes of ankerite-chlorite-sericite altered
lamprophyre occur along the contacts of the granite porphyry
dyke, and within the adjacent metapelite. The occurrence of
inclusions of lamprophyre in the granite adjacent to its contacts
with lamprophyre suggests that the granite was intruded into the
lamprophyre (Figure 3). The least altered samples of the
lamprophyre contain relict primary phenocrysts of plagioclase,
titaniferous augite, brown amphibole, biotite, magnetite, ilmenite
and apatite, which together with chemical analyses indicate that
the lamprophyre is a camptonite (Windle and Craw, 1991).
Contacts of the lamprophyre and metapelite are commonly
sheared, but in some cases a sharp contact has been preserved
and a crenulation cleavage extends from the metapelite into the
altered lamprophyre.
A single K-Ar date of amphibole from the granite gave an age
of 226 1 Ma (Tulloch, 1992), but this age is probably a
minimum due to resetting during the thermal and deformation
events that have affected the area. Despite high Zr contents of
1180 - 1380 ppm, the granite contains insignificant amounts of
zircon, and attempts to separate it for U-Pb dating have been
unsuccessful.
PACRIM 2004
 THE SAMS CREEK PERALKALINE GRANITE HOSTED GOLD DEPOSIT, NORTHWEST NELSON, NEW ZEALAND

grained biotite. In the western outcrops, this magnetite-rich

0.5 mm
FIG 3 - Inclusion of altered lamprophyre in siderite altered
microgranite. Photomicrograph in plane transmitted light.
Hydrothermal alteration and mineralisation
Three stages of alteration and associated mineralisation are
recognised within the granite. Stage I is represented by a
magnetite-sideritebiotite assemblage and consists mainly of
magnetite and siderite grains that are clustered in and around the
remnants of arfvedsonite and aegerine phenocrysts (Figure 4). In
a few samples magnetite and siderite are accompanied by fine
assemblage is absent and a quartz-sericite-siderite alteration
assemblage is present. A K-Ar date of 101 2 Ma on a
sericite-altered granite sample from the western outcrops may
represent either the age of the hydrothermal alteration, or a
closure temperature for muscovite/sericite following uplift after
the Early Cretaceous thermal event that has affected the region
(Tulloch, 1992).
Stage II alteration/mineralisation consists of quartz-pyrite
veins which are cut by the third stage of sulfide and
quartz-siderite veins (Figure 4). The third stage of mineralisation
is the main stage of ore mineralisation and the only gold-bearing
stage. Adjacent to the Stage III veins, the granite is silicified to a
pervasive quartz-perthite-albite-siderite-rutile-pyritesericite
assemblage (Figures 5a and 5b). Remnant arfvedsonite and
aegerine are replaced by siderite, and magnetite is replaced by
pyrite and rutile. A fourth stage of pyrite-rich veins (1 - 2 mm
thick) commonly occurs in zones of more intense deformation.
These late pyrite-rich veins are aligned parallel to a schistose
fabric in the host granite (Figure 5b). Country rock metapelites
are locally altered to a carbonate-sericite-pyrite assemblage
along dyke contacts.
Gold is associated with arsenopyrite and pyrite in an irregular
stockwork of deformed sulfide-quartz-siderite veins that are
confined to the granite porphyry dyke. In some drill hole
intersections the Stage III sulfide-quartz-siderite veins occur over
the full width of the dyke. Individual sulfide veins are relatively
thin (1 mm to several centimetres, Figures 4 and 5) and consist of
mainly arsenopyrite and pyrite, with minor galena, sphalerite,
gold, chalcopyrite, pyrrhotite and graphite. Quartz and siderite
are locally present in the sulfide veins and also occur as separate
veins (Figures 4, 5a and 5b). Graphite occurs very locally in
association with arsenopyrite (Figure 6a). Gold (2 - 100 m in
size) is confined to sulfide-rich veins, where it is locally
common. It is associated with galena, arsenopyrite, sphalerite
and chalcopyrite (Figures 6b and 7a), and is found particularly as
inclusions with galena in arsenopyrite and pyrite crystals.
Electron microprobe (EMPA) analyses show that the gold
contains significant silver (15.6 - 30.8 wt per cent Ag) with a
range in fineness of 692 - 844 (average of 781). This range in
fineness is consistent with microscope observations of significant
variation in colour of the gold from deep to pale yellow. Within
individual samples the composition of the gold appears to be
fairly uniform.

Stage II quartz-pyrite veins

Stage III
apy-py-(gn-sl-Au-
cp-po-graph) +
Stage I magnetite-siderite-biotite quartz-siderite

10.0 mm

FIG 4 - Alteration and mineralisation stages within the Sams Creek granite. Stage I magnetite-sideritebiotite
altered granite (mottled black) cut by 1-2 mm thick Stage II quartz-pyrite veins (white) and by a 3 - 5 mm thick
Stage III arsenopyrite-pyrite-(galena-sphalerite-gold-chalcopyrite-pyrrhotite-graphite) vein (silver grey)
and associated quartz-siderite (white) vein. Photo of drill core.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 129

R L BRATHWAITE and K FAURE

a b

py
apy

sid

10.0 mm apy 10.0 mm

FIG 5 - Sulfide veins in silicified microgranite. (A) Arsenopyrite-rich veins (apy) and quartz-siderite veins (sid). (B) Deformed arsenopyrite-rich
vein (apy) and two generations of pyrite-rich (py) veins. Note alignment of late pyrite veins along cleavage. Photos of drill core.

a b apy

sl

qtz qtz
Au py
sl cp
graphite

gn
apy
apy
0.05 mm 0.05 mm

FIG 6 - Mineralisation. (A) Graphite with arsenopyrite (apy) and quartz (qtz). (B) Gold (Au) associated with galena (gn), chalcopyrite (cp) and
sphalerite (sl) with arsenopyrite (apy) and quartz (qtz). Photomicrographs in reflected light.

A number of prospects have been identified along strike of the TABLE 1

Sams Creek dyke including Riordans, Western Outcrops, Doyles,
Results of chemical analyses of granite and lamprophyre in drill
South Eastern Traverse, Main Zone, Anvil Zone and Barrons core samples.
Flat, but only one of these, the Main Zone, has been
systematically drilled to date (Figure 2). The current resource Metals (ppm) Au Ag As Zn Pb Cu Mo Bi
estimate for Sams Creek is 10.7 Mt at 2.12 g/t Au containing
Average 1.8 1.1 6972 324 148 16 5.6 0.4
729 000 oz at a 0.7 g/t Au cut-off and limited to a depth of 140 m
(Oceana Gold, formerly GRD Macraes, Quarterly Report Number 260 249 249 249 249 249 249 249
31 December 2003). This is based on limited drilling completed Maximum 9.2 34.5 43 500 4350 2050 320 26 7
to date for the Main Zone.
different sampling scales of metre lengths for the geochemical
Metal geochemistry analyses and few millimetres for the polished thin sections.
Geochemical analyses of selected elements from intersections of
the granite and adjacent lamprophyre dykes in four drillholes Deformation textures
(SC43, 44, 45 and 48) were provided by Oceana Gold Ltd.
One-metre long drill core samples were analysed for gold by fire The granite and the Stage III sulfide and quartz-siderite veins
assay and As, Cu, Pb, Zn, Mo and Bi by atomic absorption have been deformed and recrystallised. Quartz crystals in the
spectroscopy (AAS) or by inductively coupled plasma mass Stage II veins commonly show a schistose fabric, with the quartz
spectroscopy (ICP-MS). The results (Table 1) confirm the crystals elongated parallel to the vein walls. Vein pyrite and
mineralogical observations in showing moderate As (average arsenopyrite, being competent sulfide minerals are extensively
0.7 wt per cent) representing arsenopyrite, as well as local fractured and brecciated. Fractures within arsenopyrite are
locally filled by galena and gold (Figure 7a). Galena is a ductile
concentrations of zinc and lead as sphalerite and galena, with low
sulfide mineral, which under deformation readily migrates to low
copper as traces of chalcopyrite, and insignificant Mo and Bi.
pressure sites. Vein quartz shows local development of
Inter-element correlations show a strong correlation between Au cataclastic and mortar textures with broken quartz fragments in a
and As (r = 0.75) and weak correlations between Au and Ag (r = matrix of fine grained recrystallised quartz. Quartz pressure
0.28) and Au and Pb (r = 0.32). The weak geochemical shadows are formed around larger arsenopyrite crystals (Figure
correlations between Au and Ag and between Au and Pb contrast 7b), indicating that the arsenopyrite crystals are pre- or
with the relatively high silver content of the gold and the common syn-tectonic. Quartz in phenocrysts and veins shows undulose
association of gold and galena. This may be a reflection of the extinction, deformation bands and deformation lamellae.

130 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE SAMS CREEK PERALKALINE GRANITE HOSTED GOLD DEPOSIT, NORTHWEST NELSON, NEW ZEALAND

Fluid inclusions such as CH4 and N2 in addition to CO2, particularly because

Vein quartz is extensively recrystallised and contains tiny,
single-phase fluid inclusions (<2 m in size) as a monophase
haze. Early fluid inclusions are rare and are only preserved in
vein quartz that is associated with massive arsenopyrite veins in
the Main Zone (Figure 8a). The rigidity of the arsenopyrite
appears to have protected the quartz and its contained inclusions
from recrystallisation. The early inclusions occur as small
(<10 m) two- and three-phase immiscible liquid-rich carbonic
inclusions (Figure 8a) and two-phase liquid-rich inclusions. At
room temperature, the immiscible liquid-rich carbonic inclusions
contain two liquid phases (carbonic-rich and H2O-rich), with or
without a vapour phase as a CO2 vapour bubble.
A previous study by Windle and Craw (1991) described
CO2-rich inclusions from the western part of the dyke (Western
Outcrops and Riordans), which coexisted with liquid-rich
inclusions and therefore represented trapping of immiscible
carbonic- and liquid-rich fluids at temperatures of 260 - 340C.
The carbonic phase and clathrate melting temperatures of these
CO2-rich inclusions indicated that the carbonic phase was mainly
CO2.
In this study, the carbonic inclusions homogenised by
disappearance of the H2O-rich phase at temperatures of 320 to
355C, which are probably less than the actual trapping
temperatures (cf Shepherd, Rankin and Alderton, 1985). During
freezing runs the carbonic phase melted (TmCO2) at temperatures
of -62.0 to -66.0C and a clathrate phase melted (TmC) at 11.6 to
12.5C. These data suggest that the fluid contains other gases
TmCO2 is well below the -56.6C triple point of pure CO2 (eg
Diamond, 2003). The presence of CH4 in the fluid is consistent
with the occurrence of graphite in the mineralisation.
The liquid-rich carbonic inclusions have Th values of 258 to
310C and contain clathrates, as shown by TmC values in the
range of 9.2 to 11.6C. Liquid CO2 was not observed in these
inclusions and a TmCO2 value of -57.6C indicates near pure CO2
in the carbonic phase. Tmi values of -1.7 to -6.2C convert to
apparent salinities of 1.7 to 6.2 wt per cent NaCl equiv.
Remnants of early fluid inclusions occur within deformed vein
quartz as decrepitated fluid inclusions and as healed empty
inclusions (Figure 8b). These remnant fluid inclusion
morphologies are similar to those described from a
metamorphosed ore deposit by Giles and Marshall (1994). The
healed inclusions were interpreted by these authors as leaked
inclusions which re-equilibrated to a lower bulk density under
conditions of high internal overpressure.
COMPARISION OF THE SAMS CREEK GOLD
DEPOSIT WITH OROGENIC AND INTRUSION
RELATED GOLD DEPOSITS
Commonalities with orogenic lode gold deposits, are the low
sulfidation state of the ore mineral assemblage (arsenopyrite and
pyrite), carbonate alteration and carbonic fluid inclusions at
Sams Creek (Table 2). However, significant differences are that
the Sams Creek deposit has a high sulfide content, contains
significant galena, sphalerite and chalcopyrite and no scheelite or

a b
apy

gn
apy Au

0.05 mm 0.1 mm

FIG 7 - Deformation and recrystallisation textures. (A) Fractured arsenopyrite (apy) with one fracture infilled by galena (gn) and gold (Au).
(B) Recrystallised quartz as pressure shadow around arsenopyrite (black). Photomicrographs in reflected light.

a b

liquid CO2 with small CO2 vapour bubble

apy

0.01 mm 0.01 mm

FIG 8 - Fluid inclusions in vein quartz. (A) Immiscible carbonic liquid-rich inclusions in quartz enclosed by arsenopyrite (apy). Note three phase
inclusion containing H2O-rich liquid and liquid CO2-rich bubble enclosing small CO2-rich vapour bubble. (B) Secondary healed empty
inclusions. Photomicrographs in plane transmitted light.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 131

R L BRATHWAITE and K FAURE

TABLE 2
Comparison of Sams Creek gold deposit with other types of gold deposits.
Sams Creek Porphyry Cu-Au-Mo Reduced granite Alkaline intrusion Orogenic
Au deposit deposits Au deposits related Au deposits Au deposits
Granite: type A I I A and I No direct spatial
composition Peralkaline Calc-alkaline Sub-alkaline Alkaline-peralk relationship
oxdidation-state Oxidised-reduced Oxidised Reduced Oxidised
Lamprophyres Yes No Local Variable Variable
Tectonic setting Anorogenic Subduction-related Continental inboard Continental inboard Orogenic slate belts
plutonic arcs plutonic arcs plutonic arcs
Vein style Stockwork Stockwork and breccia Stockwork and breccia Stockwork Planar fault-hosted
(shallow) Planar (deep)
Metals+Fe+Au As, Zn, Pb, Ag Cu, Mo, Pb, Zn, Ag Bi, Mo, W, Sb, (Sn) Cu, Mo, Te As, W, Sb
Sulfide content High 10 - 30 % High Low <5 % High Low <3 %
Ore minerals+gold apy, py, gn, sl, cp, po py, cp, bn, mo po, py, apy, mo, sch cp, bn, py, mo, mt, hm py, apy, sch, stbn
Sulfidation-state Low High Low Moderate Low
Alteration: early mt-sd-(bt) kspar-bt-mt qtz-ser ab-kspar ser-cb kspar-bt-mt-anh cb-(ser) ser-cb-qtz-chl
Alteration: late qtz-sd-py-rt-(ser)
Fluid inclusions Carbonic V-rich and high salinity CO2-rich and Mod-high salinity and Liquid-rich carbonic
high salinity CO2-rich
References This study Sillitoe (2000) Thompson and Jensen and Barton (2000) Groves et al (1998)
Newberry (2000)

Abbreviations: ab = albite, anh = anhydrite, apy = arsenopyrite, bt = biotite, bn = bornite, cb = carbonate, chl = chlorite, cp = chalcopyrite,
ep = epidote, gn = galena, kspar = potassium feldspar, mo = molybdenite, mt = magnetite, peralk = peralkaline, po = pyrrhotite,
py = pyrite, qtz = quartz, sch = scheelite, ser = sericite, sd = siderite, sl = sphalerite, stbn = stibnite.

stibnite, and has early magnetite-sideritebiotite alteration.
Additional differences from orogenic gold deposits are the
stockwork-type veining and absence of mineralisation in the
metapelite country rocks at Sams Creek. Windle and Craw
(1991) proposed that the Sams Creek deposit was of the slate
belt (orogenic gold type) and contended that the mineralisation
was confined to the granite because of its distinctive Fe3+-rich
composition. However, this view is at odds with their description
of deformation textures in the quartz and sulfide veins, which
indicates that the mineralisation pre-dated the deformation and
metamorphism. It is conceivable that the mineralisation could be
syntectonic, but the lack of any through-going structural control
on the veins and the gross relationship of the mineralisation and
the alteration to the granite host rock are inconsistent with a
syntectonic origin (cf Marshall and Gilligan, 1993). Also, its
A-type chemistry, suggests that the Sams Creek granite was
emplaced in an anorogenic tectonic setting.
As listed in Table 2, features of the Sams Creek gold deposit
show similarities and differences with the various types of
gold-rich porphyry and granite-related deposit types. The high
sulfide content, stockwork style sulfide-quartz veining and the
presence of an early magnetite-biotite hydrothermal alteration
are features of many gold-rich porphyry Cu deposits (eg Sillitoe,
2000). However, the tectonic setting, high-SiO2 and peralkaline
composition of the Sams Creek granite, the absence of
molybdenite and paucity of chalcopyrite, the low sulfidation state
of the ore mineral assemblage and the carbonic fluid inclusions
are major points of difference.
A low sulfidation-state ore mineral assemblage (with
arsenopyrite, pyrite and pyrrhotite), carbonate alteration and
carbonic fluid inclusions are typical of the reduced-granitic
intrusions type (eg Thompson and Newberry, 2000; Lang and
Baker, 2001). However, there are some notable differences
between the reduced-granitic intrusions type and the Sams Creek
gold deposit. The Sams Creek granite is A-type and peralkaline,
whereas granites of the reduced type are generally I-type and
metaluminous and sub-alkaline or rarely alkaline granite. The
Sams Creek gold deposit has a high sulfide content and only has
high As in contrast to a low sulfide content and elevated Bi, W,
As, Mo, Te Sb in the reduced type.
An association of gold-rich deposits with alkaline magmatism
has also been recognised (eg Jensen and Barton, 2000). The
Sams Creek gold deposit shows affinities with this type, although
there are some differences (Table 2). The alkaline
intrusion-related type covers a range of compositions from
alkaline to peralkaline and are typically oxidised. Early
magnetite-biotite alteration and late carbonate alteration are
typical, as at Sams Creek. Some examples of this deposit type
have an association with alkaline lamprophyres (eg Rock et al,
1989), such as the Porgera gold deposit in Papua New Guinea.
However, alkaline granites with high SiO2, comparable with the
Sams Creek granite, are commonly associated with MoAu
mineralisation and have moderate to high, rather than low,
sulfidation states in the ore mineral assemblage.
Gold deposits associated with granitoids are a diverse group
and the Sams Creek gold deposit, while showing some features
of the various groups, appears to be a new variant of the alkaline
intrusion-related type, as an As-Au deposit associated with a
peralkaline granite and lamprophyre.
CONCLUSIONS
Stockwork vein mineralisation at the Sams Creek gold deposit,
consisting of quartz, siderite, and arsenopyrite + pyrite Au
galena sphalerite chalcopyrite pyrrhotite graphite, is
restricted to a peralkaline granite porphyry dyke that intruded
metasedimentary rocks. The granite dyke and the sulfide and
quartz-siderite veins within it have been deformed and
recrystallised during two phases of folding, the youngest of
which is of Early Cretaceous age from the regional geological
history. Therefore the emplacement of the granite and associated
mineralisation is at least pre-Early Cretaceous.
The presence of stockwork veining, magnetite-sideritebiotite
alteration, a high sulfide content with significant galena,
sphalerite and chalcopyrite, and the absence of mineralisation in
the metapelite country rocks distinguish the Sams Creek deposit
from the orogenic gold deposit type. In relation to granite type,
sulfide content and alteration, the Sams Creek gold deposit
resembles alkaline intrusion-related gold deposits, whereas in its

132 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE SAMS CREEK PERALKALINE GRANITE HOSTED GOLD DEPOSIT, NORTHWEST NELSON, NEW ZEALAND
low sulfidation-state ore mineralogy with arsenopyrite, it is more
akin to reduced-granite gold deposits. The association of a
peralkaline granite with alkaline lamprophyre is a key feature
and suggests that the occurrence of these rock types provides a
target for exploration of analogues of the Sams Creek deposit in
other terranes. The fact that the Sams Creek was only discovered
in the 1980s is an enigma, because the gold-bearing
sulfide-quartz veins at Sams Creek crop out in a region that was
intensively prospected for gold in the late 1800s.
ACKNOWLEDGEMENTS
Oceana Gold Ltd provided access to drill core and information
from the Sams Creek gold prospect, and in particular we thank
Lachlan Reynolds, Sean Doyle, Peter Grieve and Austin Osborne
for their help. Cornel de Ronde provided the impetus to get our
research on Sams Creek started. Andy Tulloch and Ian Graham
reviewed the manuscript and provided helpful comments. The
study was funded by the New Zealand Foundation for Research,
Science and Technology (CO5X0207).
REFERENCES
Bradshaw, M A, 2000. Base of the Devonian Baton Formation and the
question of a pre-Baton tectonic event in the Takaka Terrane, New
Zealand, New Zealand Journal of Geology and Geophysics,
43:601-610.
Challis, G A, Grapes, R and Palmer, K, 1995. Chromian muscovite,
uvarovite, and zincian chromite: products of regional metasomatism
in Northwest Nelson, New Zealand, The Canadian Mineralogist,
33:1263-1284.
Clementson, I M, 1987. Takaka River PL 31-1195 and Cobb River PL
31-1196, Sams Creek project, report on exploration March 1985 to
May 1987, CRA Exploration Pty Ltd, Unpublished open file
company report (Ministry of Commerce M1014: Wellington).
Cooper, R A, 1989. Early Paleozoic terranes of New Zealand, Journal of
the Royal Society of New Zealand, 19:73-112.
Diamond, L W, 2003. Introduction to gas-bearing aqueous fluid
inclusions, in Fluid Inclusions: Analysis and Interpretation (Eds:
I Samson, A Anderson, D Marshall), pp 101-158 (Mineralogical
Association of Canada Short Course Series Volume 32).
Faure, K, Brathwaite, R L and de Ronde, C E, 2003. Gold mineralisation
in the polymetallic Sams Creek peralkaline microgranite, South
Island, New Zealand, Journal of Geochemical Exploration,
78-79:613-616.
Giles, A D and Marshall, B, 1994. Fluid inclusion studies on a multiply
deformed, metamorphosed volcanic associated sulphide deposit,
Joma mine, Norway, Economic Geology, 89:803-819.
Grindley, G W, 1980. Sheet S13 Cobb, Geological Map of New Zealand
1:63 360, (DSIR: Wellington).
Groves, D I, Goldfarb, R J, Gebre-Mariam, S G, Hagemann, S G and
Robert, F, 1998. Orogenic gold deposits: a proposed classification in
the context of their crustal distribution and their relationship to other
gold deposit types, Ore Geology Reviews, 13:7-27.
Hawke, A A, 1982. Takaka project, New Zealand, report on 1982 field
and drilling program, Cobb River PL 31-1196, CRA Exploration Pty
Ltd, Unpublished open file company report (Ministry of Commerce
M0970: Wellington).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Hawke, A A, 1984. Sams Creek dyke, Cobb PL 31-1196 Takaka project,
New Zealand, report on drilling programme from 20.10.83 to
17.5.84, CRA Exploration Pty Ltd, Unpublished open file company
report (Ministry of Commerce M0982: Wellington).
Hickey, K A, 1986. Geology of Paleozoic and Tertiary rocks between
Upper Takaka and the Waingaro River, north-west Nelson, MSc
thesis (unpublished), University of Auckland.
Jensen, E P and Barton, M D, 2000. Gold deposits related to alkaline
magmatism, Gold in 2000, Society of Economic Geology Reviews,
13:279-314.
Jongens, R, 1997. The Anatoki Fault and the structure of the adjacent
Buller and Takaka Terrane rocks, Northwest Nelson, New Zealand,
PhD thesis (unpublished), University of Canterbury, Christchurch.
Lang, J R and Baker, T, 2001. Intrusion-related gold systems: the present
level of understanding, Mineralium Deposita, 36:477-489.
Marshall, B and Gilligan, L B, 1993. Remobilisation, syntectonic
processes and massive sulphide deposits, Ore Geology Reviews,
8:39-64.
Muir, R J, Weaver, S D, Bradshaw, J D, Eby, G N and Evans, J A, 1995.
The Cretaceous Separation Point batholith, New Zealand: granitoid
magmas formed by melting of mafic lithosphere, Journal of the
Geological Society of London, 152:689-701.
Osborne, M A, 1997. Report on the re-logging of Sams Creek core,
Macraes Mining Co Ltd, Unpublished open file company report
(Ministry of Commerce M3960: Wellington).
Rattenbury, M S, Cooper, R A and Johnston, M R (compilers), 1998.
Geology of the Nelslon area, Institute of Geological and Nuclear
Sciences 1:250 000 geological map 9, (Institute of Geological and
Nuclear Sciences Limited: Lower Hutt, New Zealand).
Rock, N M S, Groves, D I, Perring, C S and Golding, S D, 1989. Gold,
lamprophyres and porphyries: what does their association mean?,
Economic Geology Monograph, 6:609-625.
Rosengren, P G, 1985. Sams Creek project PL 31-1196 Cobb River,
report to Mines Division on drilling programme May 1984 to
February 1985, CRA Exploration Pty Ltd, Unpublished open file
company report (Ministry of Commerce M0993: Wellington).
Sillitoe, R H, 2000. Gold-rich porphyry: descriptive and genetic models
and their role in exploration and discovery, Gold in 2000, Society of
Economic Geology Reviews, 13:315-345.
Shelley, D, 1984. Takaka River recumbent fold complex, Nelson, New
Zealand, New Zealand Journal of Geology and Geophysics,
27:139-149.
Shepherd, T J, Rankin, A H and Alderton, D H M, 1985. A Practical
Guide to Fluid Inclusion Studies, (Blackie: London).
Thompson, J F H and Newberry, R J, 2000. Gold deposits related to
reduced granitic intrusions, Gold in 2000, Society of Economic
Geology Reviews, 13:377-400.
Tulloch, A, 1983. Granitoid rocks of New Zealand a brief review,
Geological Society of America Memoir, 159:5-19.
Tulloch, A J, 1992. Petrology of the Sams Creek peralkaline granite dyke,
Takaka, New Zealand Journal of Geology and Geophysics,
35:193-200.
Windle, S J and Craw, D, 1991. Gold mineralisation in a syntectonic
granite dyke, Sams Creek, Northwest Nelson, New Zealand Journal
of Geology and Geophysics, 34:429-440.
133
Disseminated Gold Mineralisation in a Schist-Hosted
Mesothermal Deposit, Macraes Mine, Otago, New Zealand
D Craw1, D J MacKenzie2 and B Petrie3

ABSTRACT
The Macraes mine is producing 175 000 ounces Au/year from a
mesothermal deposit with total resource of at least five million ounces.
The deposit is hosted in greenschist facies Otago Schist, a widespread
Mesozoic metasedimentary sequence that forms the basement of the
South Island of New Zealand. Mineralisation was controlled by a late
metamorphic shear zone which had both brittle and ductile structures
accompanying mineralisation. The shear zone dips shallowly (about 15),
subparallel to the shallow-dipping regional foliation, and mineralisation
occurred during southwesterly-directed thrust motion. Mineralisation
occurred under greenschist facies conditions (about 350C) during uplift
of the Otago Schist in the late Jurassic. No syn- or late-metamorphic
igneous bodies occur within the schist belt. The hosting shear zone is up
to 120 m thick, and is traceable for more than 30 km along strike. The
shear zone is truncated by an underlying low angle normal fault that
juxtaposes mineralised lower greenschist facies rocks on to unmineralised
upper greenschist facies rocks.
The whole hosting shear zone has been hydrothermally altered to some
degree, but mineralogical evidence for this alteration is subtle. Altered
rocks contain disseminated siderite, kaolinite, and rutile, rather than
calcite, epidote and titanite from the peak metamorphic assemblage. FIG 1 - Locality map showing principal mineralised vein systems
Alteration is most pronounced in the more strongly deformed rocks, (heavy lines; partly after Williams, 1974) in the Otago Schist of the
which anastomose around less-deformed lenses on the 1 - 20 m scale. southern South Island, New Zealand. Principal faults that cut the
Quartz veins occur locally, but they are thin (typically <1 m) and schist belt are also shown. The location of the Macraes mine, and
discontinuous on the 1 - 10 m scale. Mining at Macraes was initially the area of Figure 2 are indicated in the box to the north of the city
focussed on a zone of quartz veins exposed at the surface in historic of Dunedin. Inset shows location of Figure 1 (box) in the South
workings. Expansion of the mine along strike in the hosting structure has
Island of New Zealand.
revealed large volumes of mineralised rocks without quartz veins. Gold is
disseminated in schist (up to 5 ppm), accompanied by sulfide minerals
(principally pyrite and arsenopyrite), with only minor silicification. Some value in constructing future exploration programs. This paper
of the disseminated mineralisation was accompanied by addition of reports on our on-going research project on the nature of
hydrothermal graphite (up to 3 wt per cent), and sericitic alteration of mineralised rock and its structural control in the Macraes
metamorphic albite. Graphitisation and sericitisation has controlled deposit, and the relationship between the host structure and the
location of individual shears, and vice versa. Hydrothermal addition of regional geology.
As, W, Au, and S has occurred along shear zones and in associated schist
via grain boundary fluid migration. There has also been minor addition of
Bi, Sb, Hg, and Mo. This style of disseminated hydrothermal alteration is REGIONAL SETTING
difficult to detect in surface outcrops where quartz veins are absent.
The Macraes mine is hosted in the Otago Schist belt of southern
Nevertheless, these mineralised rocks constitute an important
large-tonnage, low-grade resource at the mine. The concept of
New Zealand (Figures 1 and 2). For regional mapping
disseminated gold, without quartz veins, is an under-utilised exploration convenience, the schist is subdivided into textural zones (TZ;
model in metamorphic terranes such as the Otago Schist. Turnbull et al, 2001) based on the degree of reconstitution. The
schists are derived from Mesozoic metasediments that were
deformed, metamorphosed and uplifted in the late Mesozoic
INTRODUCTION (Mortimer, 1993). The rocks of the belt grade from unfoliated
The Macraes mine is the largest gold mine in New Zealand, and metasediments (TZ I; greywackes and argillites), through slates
is a mesothermal style deposit sited in a regional scale and phyllites (semischists, TZ IIA and IIB) to strongly
mineralised structure. The mine extracts gold from mineralised segregated and foliated schists (TZ III and IV). Schists in the
schist in a large open cut operation which produces >3 t of gold vicinity of the Macraes mine are in TZ III and IV (Figure 2). The
per year with typical bulk grade of 1.6 g Au/t. Reserves exceed pervasive foliation in TZ III and IV schist is defined by
alternating layers of quartz and albite rich (light) and micaceous
50 t of gold. The structure which controlled mineralisation is
(dark) segregation laminae (Figure 3). TZ III and IV schist have
traceable for at least 30 km, and modern mining has focused on
greenschist facies mineralogy of quartz, albite, muscovite,
only 6 km of this strike length. Because of the high potential for chlorite, epidote, titanite and calcite. Sporadic biotite and garnet
further development in this structure, and the possibility of other occur in TZ IV, and mica grainsize in TZ IV schist (typically
similar structures in the vicinity, knowledge of the deposit is of 0.25 mm) is distinctly coarser than in TZ III schist (typically
0.1 mm) (Figure 3).
1. Geology Department, University of Otago, PO Box 56, Dunedin, Greenschist facies Otago Schist was uplifted after
New Zealand. metamorphism in the Jurassic and early Cretaceous (Little et al,
1999). Lower greenschist facies rocks (TZ III) were uplifted
2. Geology Department, University of Otago, PO Box 56, Dunedin, between 160 and 130 Ma (Adams and Graham, 1997). Higher
New Zealand.
grade TZ IV rocks were uplifted later than TZ III rocks, between
3. Mineral Deposit Research Unit, University of British Columbia, 120 and 100 Ma (Adams and Graham, 1997). A prominent
Vancouver V6T 1Z4, Canada. rectilinear set of northeast and northwest striking normal faults

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 135

D CRAW, D J MacKENZIE and B PETRIE

FIG 2 - Geological map of East Otago showing the Hyde-Macraes Shear Zone along the Footwall Fault, and principal other mineralised
quartz vein swarms. Representative stretching lineations and mesoscopic fold axes (partly after Winsor, 1991) are shown on both sides of
the Hyde-Macraes Shear Zone boundary. Major mafic metavolcanic horizons are also shown. Textural zones IIA, IIB, III and IV are as
defined by Turnbull et al (2001).

developed through the schist in the Cretaceous (Bishop and
Laird, 1976; Turnbull et al, 1993; Figure 1). Juxtaposition of TZ
III and TZ IV rocks occurred along some of these regional scale
normal faults in the early to middle Cretaceous (Deckert et al,
2002; Forster and Lister, 2003). Likewise, TZ III schist was
juxtaposed against TZ I and II rocks by middle Cretaceous
normal faulting. Many of the Mesozoic normal faults have been
reactivated as reverse faults in the late Cenozoic (Mutch and
Wilson, 1952; Turnbull et al, 1993).
There is no evidence for igneous activity during deformation,
metamorphism and uplift of the Otago Schist. Felsic volcanism
occurred in the Cretaceous, associated with extensional faulting
(Bishop and Laird, 1976), but no plutons are preserved in the
area. Alkaline mafic volcanic rocks were extruded in east Otago
during the Miocene (Coombs et al, 1986). Some of these
volcanic rocks contain xenoliths of crustal rocks incorporated
during emplacement, and there is no evidence in these xenoliths
for pre-Tertiary plutons beneath the schists of east Otago.
REGIONAL STRUCTURES IN THE MACRAES
MINE AREA
The largest regional structure near the Macraes mine is the
Waihemo Fault (Figures 1, 2 and 3). This northwest striking
structure was active as a major normal fault in the middle
Cretaceous (Bishop and Laird, 1976) when it juxtaposed TZ III
schists against lower grade rocks, with northeast side downward
motion. This fault has since been reactivated and forms a
prominent 2 km high scarp with lower-grade rocks upthrown on
the northeast side.
TZ III and IV schists have been juxtaposed in the vicinity of
the Macraes mine along a fault known locally as the Footwall
Fault (Figure 3). This fault dips gently northeast, subparallel to
the schist foliation, and strikes northwest. The Footwall Fault is
reputed to intersect or merge with the Waihemo Fault at depth
(Figure 3; Angus et al, 1997). The Footwall Fault zone consists
of crushed and sheared rocks from both sides of the fault. The
Footwall Fault is traceable for about 30 km between basins with
Tertiary cover sediments at either end (Figure 2). Beyond the
Tertiary sedimentary basin at the NW end of the Footwall Fault,
a related structure continues along strike. The Cap Burn Fault
(Forsyth, 2001) juxtaposes TZ II and III schist against TZ IV
schist to the southwest in the northernmost part of the Rock and
Pillar Range (Figure 2). This fault is marked in places by a
moderately (54 - 75) north-dipping shear zone up to 6 m wide
with a normal sense of movement, and locally may be shallower
dipping (Dickie, 1998). The hangingwall is marked locally by
brown Fe-carbonate alteration and veins and some earlier
unmineralised high angle quartz veins.

136 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DISSEMINATED GOLD MINERALISATION IN A SCHIST-HOSTED MESOTHERMAL DEPOSIT, MACRAES MINE

FIG 3 - Geological cross-section through the Hyde-Macraes Shear Zone (modified after Angus et al, 1997). Photomicrographs of
hangingwall and footwall schists are in plane-polarised light, and horizontal field of view is 4.8 mm. (A) Massive psammitic schist with
quartz and albite (clear grains), muscovite (elongate grains) and larger subhedral garnet. Typical muscovite grains are 250 microns long.
(B) Segregated pelitic schist with quartz and albite metamorphic segregations separated by darker layers of chlorite and muscovite.
Typical muscovite grains are 120 microns long.

Sets of steeply-dipping quartz veins with gold fill northwest
striking normal faults in several areas near to Macraes (Figure 2).
The vein swarms at Nenthorn and Lots Wife areas (Figure 2)
formed at shallow (<5 km) structural levels (Craw and Norris,
1991) and are small (metre scale) and discontinuous. The
Barewood veins (Figure 2) have formed along a set of regional
scale normal faults that are traceable for at least 20 km
(MacKenzie and Craw, 1993). Gold occurs in and adjacent to
(within <1 m) the well-defined quartz veins in all these
mineralised areas.
HYDE-MACRAES SHEAR ZONE AND MACRAES
MINE
The Macraes mine is located in the Hyde-Macraes Shear Zone.
This is a zone of variably sheared rocks that occurs in the lower
greenschist facies TZ III rocks immediately above the Footwall
Fault (Figure 3). The shear zone ranges from 2 to 120 m thick.
Deformation and mineralisation in the shear zone predated the
juxtaposition of different metamorphic grades and textural zones
of schists along the Footwall Fault, and the shear zone is
truncated by the Footwall Fault (Figure 3). The Hyde-Macraes
Shear Zone dips gently northeastwards, subparallel to the schist
foliation and the later Footwall Fault. Within the shear zone,
sheared rocks (metre scale) anastomose around pods of
less-deformed schists that are geochemically indistinguishable
from the sheared rocks (MacKenzie et al, 1999). Shearing was
accompanied by folding locally, and progressive shearing
disrupted early-formed folds (Teagle et al, 1990). Shear fabrics
and fold disruption patterns imply that the shear zone developed
during southwest-directed thrusting (Teagle et al, 1990).
Hydrothermal alteration occurred pervasively along the shear
zone. The altered rocks of the shear zone are overlain by
lithologically, structurally, and texturally similar lower
greenschist facies rocks (Angus, 1993; Craw et al, 1999). The
altered schist is generally indistinguishable from structurally
overlying unaltered schist in hand specimen where the rocks are
unmineralised. Accessory titanite and epidote have been replaced
at the sub-millimetre scale by secondary rutile, quartz, calcite,
siderite, kaolinite and chlorite (Craw et al, 1999). Some
recrystallisation of metamorphic micas (muscovite and chlorite)
accompanied hydrothermal alteration and deformation. Quartz
veins generally lie subparallel to the shear zone and are
discontinuous along it. Many of these veins have been deformed
by post-emplacement shearing, and some are partially
recrystallised via crystal-plastic mechanisms. Swarms of
steeply-dipping laminated quartz veins cut across less-sheared
rocks in the shear zone, forming stockwork arrays (Angus et al,
1997; DeRonde et al, 2000).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 137

D CRAW, D J MacKENZIE and B PETRIE

Sporadic mining of gold and scheelite occurred between 1880s

and 1940s (Williamson, 1939; Williams, 1974). This historic
mining focused on discontinuous shallow-dipping quartz veins
(metre scale) that were followed underground. The current
Macraes mine opened in 1990 with a single pit, and mining has
subsequently migrated along strike through several other pits.
The deposit is being mined with an average grade of about
1.6 ppm Au, although some mineralised zones contain up to
5 ppm Au. Gold is closely associated with sulfide minerals
(mainly pyrite and arsenopyrite) that occur locally in the
Hyde-Macraes Shear Zone. The sulfide mineralisation occurred
in a wide range of structural settings, in quartz veins and
disseminated in hydrothermally altered rocks without quartz
veins (Craw et al, 1999). Scheelite accompanies gold and
sulfides in mineralised rocks, especially in quartz veins, but
scheelite is not saved at the current mine. Ore at the first pit to be
mined was dominated by quartz veins, with relatively little gold
disseminated in altered schist (Angus, 1993). Subsequent pits
had differing proportions of disseminated and quartz vein ore.
Some pits were dominated by disseminated gold in mineralised
schist and relatively little quartz vein ore.

MINERALISED SCHIST WITH DISSEMINATED

GOLD
Some mineralised schist is difficult to distinguish from
unmineralised schist in hand specimen. Minor silicification and
millimetre scale quartz veinlets cement the otherwise fissile rock,
but the metamorphic folds and textures are largely preserved
(Figure 4a, b). Coarse (mm scale) pyrite grains are evident in
some samples (Figure 4b), and finer grained sulfides are
scattered through most mineralised samples. These sulfides have
grown across and replaced pre-existing schist metamorphic
fabric and minerals (Figure 5a). Early stages of mineralisation
were accompanied by recrystallisation of muscovite in
micaceous rocks, forming a new micaceous fabric locally within
the shear zone. In addition, fine-grained secondary muscovite
replaces albite (Craw et al, 1999). Black graphitic material
occurs in microshears at micron to millimetre scales, both
parallel to and cross-cutting the metamorphic foliation (Figure
4c). Many of these black shears contain pyrite and arsenopyrite,
with micron scale gold blebs in the sulfides (Figure 5b).
Sulfidic black shears are commonly resheared to give shiny
black surfaces (mm to cm scale). These black shears are
cataclastic with grainsize down to micron scale. Several well-
defined zones (1 - 5 m thick) with abundant black sheared rocks
cut through the Hyde-Macraes Shear Zone. Quartz veins (metre
scale) are locally developed in and adjacent to these black sheared
rocks, and some extensional quartz veins (cm scale) cut steeply
across the schistosity above and below the black sheared zones.

GEOCHEMISTRY OF ALTERATION AND

MINERALISATION
Hydrothermally altered but unmineralised rocks of the
Hyde-Macraes Shear Zone are geochemically indistinguishable
from their schist precursors (Craw, 2002). Geochemical changes
are recognisable only in mineralised rocks, and even there the
changes are subtle (Craw et al, 1999; Craw, 2002). Hydrothermal
replacement of albite by muscovite in mineralised rocks results
in minor enrichment in potassium and depletion in sodium, but
these changes barely show up through primary lithological
variations (Craw et al, 1999). Arsenic enrichment is the principal
indicator of mineralisation, and ore grade rocks typically have
>1000 ppm As. There is a close relationship between carbon
enrichment and arsenic enrichment (Figure 6), and also gold
occurrence, reflecting the widespread occurrence of auriferous
arsenopyrite in black sheared rocks (above). Elevated sulfur
contents, principally as replacement pyrite, with subordinate
FIG 4 - Photographs of hand specimens of mineralised schist
from Macraes mine. Prominent folded light and dark layering is
the metamorphic segregation and foliation fabric. This fabric is
largely unchanged by deformation and alteration accompanying
mineralisation. (A) Mineralised micaceous rock with
synmineralisation recrystallisation of micas (in dark patches),
and minor sulfidation and silicification. (B) Micaceous zones
(dark) have been extensively replaced by scattered pyrite
(light squares and blobs). Minor silicification accompanied
emplacement of unmineralised quartz veinlets (white).
(C) Black cataclastic shears with micron scale graphite, pyrite
and arsenopyrite form dark seams parallel to the metamorphic
fabric on the left, and cross-cut that fabric on the right. Minor
silicification accompanied emplacement of quartz veinlets (white).

138 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DISSEMINATED GOLD MINERALISATION IN A SCHIST-HOSTED MESOTHERMAL DEPOSIT, MACRAES MINE
FIG 5 - Photomicrographs of mineralised schist from Macraes
mine. (A) Transmitted light view with crossed polars of variably
recrystallised metamorphic micas (left side), cut by dark graphite
and sulfide mineral bearing shears (top and top left). Coarse pyrite
and arsenopyrite are intergrown on the right, replacing micaceous
rock, The coarse sulfides are partially surrounded by a veneer
of quartz (white). (B) Incident light view of arsenopyrite (pale
background) and enclosed sphalerite grain (dark, centre right)
with micron scale blebs of gold (in white ellipses).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
10000
As, ppm
mineralised rocks
100
host rocks
1
0.1 1 10
Non-carbonate carbon, wt %
FIG 6 - Chemical data (partly from Craw et al, 1999; Craw, 2002)
showing the occurrences of elevated arsenic in black graphitic
shears in mineralised rocks at Macraes mine.
arsenopyrite is a significant geochemical indicator of
mineralisation in some ore grade material. This sulfidation of the
primary schist has also formed minor amounts of sphalerite,
galena, and chalcopyrite, but there has been no enrichment in Zn,
Pb or Cu (Craw, 2002). There has been minor addition of Bi, Sb,
Hg, and Mo to mineralised rocks (Craw, 2002).
DISCUSSION
Relationship to regional structure
Mineralisation in the Hyde-Macraes Shear Zone occurred under
lower greenschist facies conditions, with associated
recrystallisation of muscovite, chlorite and quartz. The
mineralisation accompanied uplift of the host schists through the
brittle-ductile transition at about 300 - 350C (McKeag et al,
1989; Craw et al, 1999; deRonde et al, 2000). Earliest stages of
mineralisation occurred during ductile deformation, with folding,
crystal-plastic quartz deformation, and development of a
recrystallised mica fabric. These features were partially
overprinted by more cataclastic deformation in the latter stages
of shearing, characterised by the black shears (above).
Localisation of these shears was probably facilitated by
recrystallisation of micaceous schists, sericitisation of albite, and
addition of graphitic shear seams in the early stages of alteration
and mineralisation. In a regional context, this stage of schist belt
evolution coincided with a transition from km scale ductile
folding to more focused shear zone deformation with brittle and
ductile characteristics (MacKenzie and Craw, 2001). The
mineralising fluid was of metamorphic origin (Craw, 2002), and
mineral deposition occurred during fluid mixing (DeRonde et al,
2000; Craw, 2002). The fluid flow was focused by the structural
transition to a narrow shear zone from more pervasive but
low-volume metamorphic fluid percolation.
The Hyde-Macraes Shear Zone is traceable as a mineralised
structure, in TZ III schist, for the 30 km length of the Footwall
Fault that forms the shear zones lower boundary (Figure 2 and 3).
TZ III schists are exposed above the eastern end of the Cap Burn
Fault, along strike from the Footwall Fault (Figure 2), but sheared
rocks similar to the Hyde-Macraes Shear Zone have not been
found along the Cap Burn Fault. Minor shearing and hydrothermal
alteration along the Cap Burn Fault (above; Figure 2) appears to
be related to that fault, not an earlier structure. If the
Hyde-Macraes Shear Zone persisted to the northwest of its
presently-known extent, it is not exposed at the current erosion
level.
139
D CRAW, D J MacKENZIE and B PETRIE
Disseminated gold as an exploration target
Historic mining in the Otago Schist basement focused on the
quartz veins, typically with grades of >15 g/tonne (Figure 1;
Williams, 1974). Subsequent regional exploration has had a
similar focus. There was no significant mining in adjacent host
rocks because mineralisation was thought to have been restricted
to the veins themselves, although secondary enrichment
dispersed some gold into soft fault rocks beside veins during
weathering (Craw and MacKenzie, 1992). Consequently, the
Otago Schist veins constitute small exploration targets, and any
new mines developed within these veins would be relatively
low-tonnage operations. Recognition of widespread mineralised
rocks without quartz veins in the current Macraes mine, however,
suggests that past mining and exploration strategies may have
had too limited a focus. The Macraes example demonstrates that
there is potential for large tonnage mining targets within the
Otago Schist, albeit at lower grade than the veins.
The historic Otago situation as described above is not unique,
and many similar mesothermal vein systems throughout the
world have traditionally been mined along quartz veins only.
Other workers have noted more recently, however, that
hydrothermal alteration in these systems extended well beyond
the walls of the quartz veins (eg Phillips, 1986; Bierlein et al,
2000). There is clearly a need to re-evaluate exploration
programs targeted at mesothermal systems, and aim for the larger
target of mineralised rocks without quartz veins. While the
geochemical signature of these targets is more subtle than in the
quartz veins, the elements involved are the same (eg arsenic in
Otago; above). The strong structural control of the mineralised
system at Macraes suggests that structural mapping of shear
zones formed during uplift of the host rocks through the
brittle-ductile transition may be a first step in delineating
potential targets.
CONCLUSIONS
The Macraes mine is a mesothermal deposit hosted in
greenschist facies Otago Schist. Mineralisation was controlled by
a late metamorphic shear zone (at least 30 km long), and the
shear zone dips shallowly (about 15), subparallel to the
shallow-dipping regional foliation. The shear zone is up to 120 m
thick and is truncated by an underlying low angle normal fault
that juxtaposes mineralised lower greenschist facies rocks on to
unmineralised upper greenschist facies rocks. Mineralisation
occurred during southwesterly-directed thrust motion, associated
with both brittle and ductile structures, as the schist belt was
uplifted from metamorphic depths in the late Jurassic. No syn- or
late-metamorphic igneous bodies occur within the schist belt,
and the mineralisation was related to late metamorphic fluid
flow. The whole hosting shear zone has been hydrothermally
altered to some degree, but mineralogical evidence for this
alteration is subtle. Alteration is more prominent in mineralised
schists where graphite, pyrite and arsenopyrite are prominent.
Quartz veins occur locally, but they are thin (typically <1 m) and
discontinuous on the 1 - 10 m scale. Gold is disseminated in
schist with pyrite and arsenopyrite that have replaced
metamorphic micas. The concept of disseminated gold, without
quartz veins, is an under-utilised exploration model when
searching for mesothermal gold systems in metamorphic terranes
such as the Otago Schist.
ACKNOWLEDGEMENTS
This research was funded by the NZ Foundation for Research,
Science and Technology, University of Otago, and GRD Macraes
Ltd. We appreciate the help and support of staff at OceanaGold
who facilitated parts of the study on the Macraes mine site,
especially Paul Angus, Mark Mitchell, Bill Yeo, and Lindsay
Maw. Discussions with Jane Forsyth and Nick Mortimer of GNS
140 Adelaide, SA, 19 - 22 September 2004
helped with aspects of regional geology. Richard Norris and Rick
Sibson (OU Geology Dept) provided numerous insightful
discussions. Technical assistance was ably provided by Steve
Read, Damian Walls, and Brent Pooley.
REFERENCES
Adams, C J and Graham, I J, 1997. Age of metamorphism of Otago
Schist in eastern Otago and determination of protoliths from initial
strontium isotope characteristics, NZ J Geol Geophys, 40:275-286.
Angus, P V, 1993. Structural controls on gold deposition in the
Hyde-Macraes Shear Zone at Round Hill, Otago, New Zealand, in
Proceedings AusIMM NZ Branch Annual Conference, pp 11 (The
Australasian Institute of Mining and Metallurgy: New Zealand)
Angus, P V, De Ronde, C E J and Scott, J G, 1997. Exploration along the
Hyde-Macraes Shear, in Proceedings AusIMM NZ Branch Annual
Conference (The Australasian Institute of Mining and Metallurgy:
New Zealand).
Bierlein, F P, Arne, D C, McKnight, S, Lu, J, Reeves, S, Besanko, J,
Marek, J and Cooke, D, 2000. Wallrock petrology and geochemistry
in alteration haloes associated with mesothermal gold mineralization,
central Victoria, Australia, Econ Geol, 95:283-312.
Bishop, D G and Laird, M G, 1976. Stratigraphy and depositional
environment of the Kyeburn Formation (Cretaceous), a wedge of
coarse terrestrial sediments in central Otago, J Royal Soc NZ,
6:55-71.
Coombs, D S, Cas, R A, Kawachi, Y, Landis, C A, McDonough, W F and
Reay, A, 1986. Cenozoic volcanism in north, east and central Otago,
in Late Cenozoic volcanism in New Zealand (Ed: I E M Smith)
Royal Soc NZ Bull, 23:278-312.
Craw, D, 2002. Geochemistry of late metamorphic hydrothermal
alteration and graphitisation of host rock, Macraes gold mine, Otago
Schist, New Zealand, Chem Geol, 191:257-275.
Craw, D and MacKenzie, D J, 1992. Near-surface secondary gold
mobility and grain size enhancement, Barewood Mine, east Otago,
New Zealand, NZ J Geol Geophys, 35:151-156.
Craw, D and Norris, R J, 1991. Metamorphogenic Au-W veins and
regional tectonics: mineralisation throughout the uplift history of the
Haast Schist, New Zealand, NZ J Geol Geophys, 34:373-383.
Craw, D, Windle, S J and Angus, P V, 1999. Gold mineralisation without
quartz veins in a ductile-brittle shear zone, Macraes Mine, Otago
Schist, New Zealand, Mineralium Deposita, 34:382-394.
Deckert, H, Ring, U and Mortimer, N, 2002. Tectonic significance of
Cretaceous bivergent extensional shear zones in the Torlesse
accretionary wedge, central Otago Schist, New Zealand, NZ J Geol
Geophys, 45:537-547.
DeRonde, C E J, Faure, K, Bray, C J and Whitford, D J, 2000. Round Hill
shear zone-hosted gold deposit, Macraes Flat, Otago, New Zealand:
evidence of a magmatic ore fluid, Econ Geol, 95:1025-1048.
Dickie, B, 1998. Northern rock and pillar geology, BSc Hons thesis
(unpublished), University of Otago, Dunedin, New Zealand.
Forster, M A and Lister, G S, 2003. Cretaceous metamorphic core
complexes in the Otago Schist, New Zealand, Australian J Earth Sci,
50:181-198.
Forsyth, P J, 2001. Geology of the Waitaki area, Institute of Geological
and Nuclear Sciences 1:250 000 geological map 19. 1 sheet + 64 p,
Lower Hutt, New Zealand Institute of Geological and Nuclear
Sciences Ltd.
Little, T A, Mortimer, N and McWilliams, M, 1999. An episodic
Cretaceous cooling model for the Otago-Marlborough Schist, New
Zealand, based on 40Ar/39Ar white mica ages, NZ J Geol Geophys,
42:305-325.
MacKenzie, D J and Craw, D, 1993. Structural control of gold-scheelite
mineralisation in a major normal fault system, Barewood, eastern
Otago, New Zealand, NZ J Geol Geophys, 36:437-445.
MacKenzie, D J and Craw, D, 2001. A 3-dimensional section through the
Otago Schist, New Zealand, in Proceedings AusIMM NZ Branch
Annual Conference, pp 121-130 (The Australasian Institute of
Mining and Metallurgy: New Zealand).
MacKenzie, D J, Turnbull, I M, Mortimer, N, Craw, D and Norris, R J,
1999. Mapping of structural discontinuities in the Otago Schist as
potential exploration targets: preliminary results, in Proceedings
AusIMM NZ Branch Annual Conference, pp 71-84 (The Australasian
Institute of Mining and Metallurgy: New Zealand).
PACRIM 2004
 DISSEMINATED GOLD MINERALISATION IN A SCHIST-HOSTED MESOTHERMAL DEPOSIT, MACRAES MINE

Mortimer, N, 1993. Jurassic tectonic history of the Otago schist, New
Zealand, Tectonics, 12:237-244.
Mutch, A R and Wilson, D D, 1952. Reversal of movement on the Titri
Fault, NZ J Sci Tech, B33:398-403.
Phillips, G N, 1986. Geology and alteration in the Golden Mile,
Kalgoorlie, Econ Geol, 81:779-808.
Teagle, D A H, Norris, R J and Craw, D, 1990. Structural controls on
gold-bearing quartz mineralisation in a duplex thrust system,
Hyde-Macraes Shear Zone, Otago Schist, New Zealand, Econ Geol,
85:1711-1719.
Turnbull, I M, Craw, D and Norris, R J, 1993. Pre-Miocene and
post-Miocene deformation in the Bannockburn basin, Central Otago,
New Zealand, NZ J Geol Geophys, 36:107-115.
Turnbull, I M, Mortimer, N and Craw, D, 2001. Textural zones in the
Haast Schist: a reappraisal, NZ J Geol Geophys, 44:171-183.
Williams, G J, 1974. Economic Geology of New Zealand, Monograph 4.
(The Australasian Institute of Mining and Metallurgy: Melbourne).
Williamson, J H, 1939. The geology of the Naseby Subdivision, central
Otago, NZ Geol Survey Bull, 39.
Winsor, C N, 1991. Low-angle shear zones in Central Otago, New
Zealand their regional extent and economic significance, NZ J Geol
Geophys, 34:501-516.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 141

The Jianchaling Mine, Shaanxi Province, Peoples Republic of
China A Carlin-Type Gold Deposit?
M Erceg1, F Kaiqiang2 and T Leach3
ABSTRACT
The Jianchaling gold deposit is located about 1200 km south west of
Beijing in central China. The deposit was discovered in 1988 by the
Northwestern Geological Exploration Bureau (711 Exploration Team).
Sino Gold Limited, an Australian based company, developed the mine in
1998 and is currently operating a 1000 t per day underground mine,
producing about 80 000 ounces of gold per year. The pre-mine resource
for Jianchaling was 1.6 million tonnes at 9.85 g/t gold.
The deposit occurs within the east-west trending Mesozoic
Qinling-Dabie orogenic belt formed during the Yanshanian Orogeny. In
the Jianchaling region, gold mineralisation is primarily hosted by regional
west-northwest to east-southeast striking faults, the most important of
which, the reverse sinistral F145 fault, hosts the larger gold deposits
including Jianchaling.
At the mine scale, the most important zone of mineralisation is
localised within a shallowly north dipping dilation on the F145 fault.
Proterozoic dolomite forms the hangingwall and serpentinite and talc
schist the footwall to the fault.
Within the footwall, serpentinite is the most important host rock to
mineralisation. Initial alteration involves replacement of serpentine to
dolomite pyrite, talc to ankerite pyrite, and magnetite and chromite to
pyrite, the latter locally through a transitional hematite phase. Proximal to
feeder structures, the carbonate minerals have undergone partial
dissolution and replacement to quartz. Gold is related to arsenean pyrite
and marcasite rimming and replacement of early-formed pyrite, and
associated with iron release during the processes of dolomitisation,
silicification and carbonate dissolution.
The dolomite in the hangingwall exhibits only local dissolution,
collapse brecciation and silicification, dolomite and calcite veining and
minor gold mineralisation associated with sulfidation.
Anomalous nickel is locally associated with the gold ore and occurs
mainly as gersdorffite and minor millerite. Petrology studies indicate that
the nickel, which is associated with the gold, has been derived from the
alteration of pentlandite-chalcopyrite-pyrite-pyrrhotite-magnetite
mineralisation that formed during the cooling of the ultramafic intrusions.
Geochemical work indicates a strong Au-As-Hg element association.
This, along with other characteristics such as the tectonic setting,
alteration and mineralisation assemblages and interpreted ore formation
process, indicate that gold mineralisation at Jianchaling can be classified
as Carlin-type.
INTRODUCTION
The Jianchaling gold deposit is operated by Sino Gold Limited
and has produced approximately 450 000 ounces from an
pre-mine resource of 1.6 million tonnes at 9.85 g/t gold since it
commenced operation in 1998. The orebody is accessed by a
1.5 kilometre decline and mined by cut and fill and shrink-stope
methods. Ore is processed on site through a conventional
crushing and ball mill circuit, a sulfide concentrate produced
through froth flotation and dor produced from carbon in leach
treatment of the tails. Concentrate is sold to Chinese smelters.
1. Jianchaling Exploration Manager, Sino Gold Limited, XQG Village,
Hejiayan Township, Lueyang County, Shaanxi Province, 724308,
China.
2. Jianchaling Chief Mine Geologist, Sino Gold Limited, XQG Village,
Hejiayan Township, Lueyang County, Shaanxi Province, 724308,
China.
3. Consultant Geologist, Terry Leach and Co, 2713 Rings Road,
Coromandel, New Zealand.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
The most unique feature of the process is the fine grind required
to liberate the sulfide for the flotation process, 90 per cent pass
45 microns. Most of the gold is associated with this very fine
pyrite and reports to the concentrate.
The deposit was the subject of a two-year project that
commenced in 1999 undertaken by the University of Western
Australia. The first report (Vielreicher, 1999) described the
regional and local geology, as well as the tectonic evolution of
the west Qinling Mountains. The second report (Vielreicher,
2000) dealt with the pressure and temperature conditions during
lower greenschist-facies metamorphism, as well as aspects of the
relative timing of intrusive rocks and gold mineralisation. The
third report (Vielreicher et al, 2001) comments on the
geochemical aspects of post-emplacement modifications of the
stratigraphic pile in the close environment of the Jianchaling
mine and the tectonic implications of the geochemical results.
The report also summarises the major points of the research on
the setting of the Jianchaling gold deposit and incorporates a
synopsis of criteria for local and regional scale mineral
exploration in the west Qinling area.
This paper focuses on the results of a review of the Jianchaling
gold deposit conducted by Terry Leach during 2003. The review
included examination of outcrop and selected drill core and
discussion with site geologists. The purpose of the study was to
assess the ore mineralogy and associated alteration, and to
develop a conceptual model for the deposits formation, which
could be used regionally to rank exploration targets.
REGIONAL GEOLOGY
The Carlin-type gold deposits in China occur in two groups, one
on the southwestern margin of the South China Craton in the
Guizhou area and the other in the northwest Sichuan area on the
north side of the South China Craton. In both areas, Carlin-type
gold deposits coexist with epithermal ore deposits of Hg, Sb, As,
U, and Tl that together constitute two much larger metallogenic
provinces. Both areas also contain Mississippi Valley-type Pb-Zn
deposits and bedded barite deposits that are older (Hu Rui-Zhong
et al, 2002).
The South China Craton consists of Proterozoic metamorphic
rocks overlain by a thick carbonate and shale sequence of
Cambrian through Triassic age. Interaction between the South
China Craton and the North China Craton commenced in the
Devonian and the deposition of thick sequences of Devonian to
Triassic marine sedimentary rocks. These rocks were uplifted
and folded during the Early Jurassic to latest Cretaceous (190-65
Ma) Yanshanian Orogeny (Hu Rui-Zhong et al, 2002).
The initial Yanshanian stress regime caused major regional
scale sinistral faulting in the Jianchaling area as the South China
Craton collided with the North China Craton and rotated in a
clockwise direction (eastwards tectonic escape). By late
Yanshanian the stress regime had reversed as subduction of the
Pacific plate beneath both Cratons became dominant and the
Jianchaling area became a wrench and pull-apart basin belt
(Ratschbacher et al, 2003).
Igneous rocks associated with the Yanshanian Orogeny are
mainly found in eastern China, distant to the Carlin-type gold
deposit areas, and consist largely of granites and felsic volcanic
rocks. Although intrusive rocks are absent at most of the gold
143
M ERCEG, F KAIQIANG and T LEACH
deposits, a few contain small mafic dykes, quartz porphyries,
granite porphyries or dacite porphyries that were emplaced
during the Yanshanian.
LOCAL GEOLOGY
The Jianchaling deposit is hosted within an uplifted block of
Archean and Proterozoic rocks, the Mian-Lueyang Block (Figure
1). The Mianlue Suture Zone separates the Mian-Lueyang Block
from the Qinling-Dabie orogenic belt to the north. The Bashan
Fault separates the Mian-Lueyang Block from the South China
Craton to the south. The uplifted block exhibits a core of
Archean gneiss, overlain by Sinian (late Proterozoic) ultramafic
rocks, dolomite, limestone and shale.
The F145 Fault, within the Mian-Lueyang Block, subparallels
the Mianlue Suture Zone and hosts the Jianchaling gold deposit.
The hangingwall sequence of the F145 Fault is dominated by a
steeply north dipping turbidite-like sequence of massive to
FIG 1 - Jianchaling gold deposit: location and regional geological setting.
144 Adelaide, SA, 19 - 22 September 2004
laminated limestone and dolomite, with local shale and chert
horizons. The footwall sequence consists of variably foliated
serpentinite and talc schist and massive metagabbro.
The F145 Fault zone is the most prominent fault in the mine
area and can be traced along strike for several kilometres. The
fault strikes mainly west-northwest to east-southeast and with a
steep dip to the north-northeast. Variations in both dip and strike
related to mainly oblique, sinistral-reverse movement focused
metalliferous hydrothermal fluids along three discrete structural
orientations (Figure 2). Gold mineralisation was preferentially
deposited along:
1. a shallow north-northeast dipping dilation (main orebody);
2. shallowly to moderately southeast dipping northeast
striking cross-faults; and
3. shallowly south-southwest dipping vein sets.
PACRIM 2004
 THE JIANCHALING MINE, SHAANXI PROVINCE, PEOPLES REPUBLIC OF CHINA

FIG 2 - Jianchaling gold deposit: schematic cross-section.

PETROLOGY A distinct dark fine-grained metagabbro dyke that cuts the

Lithology
The Jianchaling gold deposit is hosted within serpentinised
ultramafic rocks that are in fault contact with dolomite and talc
schist.
The serpentinite consists mainly of serpentine (antigorite from
XRD) with variable talc and magnesite, and minor chromite. The
serpentine commonly exhibits well-formed mesh textures (Figure
3a). The original composition of the serpentinised ultramafic is
interpreted to have been a dunite or harzburgite composed of
mostly olivine with minor amounts of pyroxene. The results of
whole rock geochemical studies (Veilreicher et al, 2000) are
consistent with this interpretation.
serpentinite on the access track at the Greenwall 900 m south of
the mine. It is highly sheared and foliated, and is composed of
pyroxene/hypersthene, hornblende,and minor plagioclase and
epidote, which are set in a fine-grained groundmass of biotite
and sphene.
A leucocratic sample collected from within the orebody
exhibits possible porphyritic textures and was postulated by the
mine geologists to be a feldspar porphyry. However petrographic
analyses indicates that the rock is a talc-carbonate schist, and the
porphyritic texture was caused by granoblastic, euhedral
magnesite (from XRD) that is set in a fine-grained groundmass
of fibrous talc, minor carbonate (dolomite from XRD) and trace
quartz.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 145

M ERCEG, F KAIQIANG and T LEACH

A B

C D
FIG 3 - Jianchaling gold deposit: selected photomicrographs of styles of alteration. (A) Mesh-textured or reticular serpentine slightly altered
to dolomite; cut by thin magnetite stringers. Disseminated chromite and locally overgrowing magnetite, are in places partially altered to
hematite. Drillhole ZK78003, 91.6 m: field of view is 2.6 mm. XPL. (B) Altered serpentinite with pervasive dolomite and pyrite after hematite
and magnetite. Cut by dolomite veins with magnetite and hematite veinlets altered to pyrite. Drillhole ZK78003, 91.6 m: field of view is
2.6 mm. XPL. (C) Stylolite-like dissolution fractures in matrix of collapse breccia with local quartz replacement of dolomite. Late kaolinite fills
vughs. Drillhole ZK77507, 20.1 m: field of view is 2.6 mm. XPL. (D) Dolomite-quartz-sulfide alteration of serpentinite clasts in a vein matrix
of coarse calcite, that has subsequently undergone brecciation with vein clasts set in a fine calcite matrix. Drillhole ZK78003, 60.2 m: field of
view is 2.6 mm. XPL.

Serpentinite alteration and mineralisation Dolomite zone

Three distinct alteration zones are evident in the serpentinite,
from distal (to feeder structures) inwards to more proximal
settings. These zones, which occur from a hand specimen scale
of a few centimetres up to many metres in width, can be
summarised as follows.
Serpentine zone
The serpentinite is a dark green colour in hand specimen and
consists mainly of serpentine with variable abundances of talc,
magnesite, chrome spinel and magnetite. Serpentine formed in
response to hydration of mainly olivine during metamorphism.
Olivine may have altered to talc and magnesite instead of
serpentinite in higher stress regimes, although there is some
evidence to suggest that, at least in some cases, the
talc-magnesite may have altered earlier formed serpentinite. This
is clearly evident where magnesite veinlets cut the serpentinite
with some magnesite selvages after serpentinite occurring
immediately adjacent to these veinlets.
Magnetite appears to have post-dated the serpentine-talc-
magnesite assemblage. The magnetite occurs both as thin
discontinuous stringers and more continuous veinlets that cut the
serpentinite and magnesite veinlets, and locally overgrows and
replaces chromite (Figure 3a).
The dolomite zone is cream to tan colour in hand-specimen,
although where serpentinite minerals are still present, the zone
has a pale green colour.
In this alteration zone dolomite replaces serpentine, talc and
magnesite, while pyrite rims and replaces both chrome spinel and
magnetite (Figure 3b and Figure 4a). In many cases the pyrite
has totally pseudomorphed these oxides. Where dolomite has
replaced serpentine, pyrite is absent or occurs in very small
amounts. However pyrite forms relatively abundant inclusions in
dolomite that has replaced talc, reflecting the presence of more
available iron in talc schist possibly as magnetite. In addition,
XRD analyses indicate that the dolomite that replaces the talc is
transitional to more iron-rich ankerite. Where the dolomite has
replaced magnesite, sulfide minerals are absent.
Quartz-sulfide zone
This zone is a dark grey colour in hand specimen due to the
presence of very fine-grained sooty sulfide minerals and occurs
immediately adjacent to the feeder structures. Dolomite
generally persists in this zone although in most cases it has
undergone slight to locally almost total replacement by very
fine-grained granular quartz.

146 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE JIANCHALING MINE, SHAANXI PROVINCE, PEOPLES REPUBLIC OF CHINA

A B

C D
FIG 4 - Jianchaling gold deposite: selected ore microscopy. (A) Chromite (grey), partially replaced by pyrite (yellow), which in turn is rimmed
and repalced by arsenian pyrite (pale yellow). Drillhole ZK78003, 91.6 m: field of view is 0.6 mm RL. (B) Marcasite (pitted rhombs) and
arsenian pyrite (pitted pale yellow) replaces and overgrows pyrite (massive pale yellow) associated with altered serpentinite. Drillhole
ZK77507, 37.0 m: field of view is 0.6 mm. RL. (C) Sericite-quartz-sulfide shear, with pyrite (massive yellow) overgrown by millerite (massive
grey), then by later gersdorffite (pitted pale grey). Drillhole ZK 8557, 191.0 m: field of view is 0.6 mm. RL. (D) Tarnished native arsenic (dusty
pitted) forms massive to radiating fibrous aggregate grains that overgrow realgar (massive pale yellow) in calcite vein. Vughs (dark) are filled
by granular calcite showing internal reflections of realgar. Rock sample JCL-1: field of view is 1.3 mm RL.

Sulfides consist of either very fine-grained pyrite, minute
poorly crystalline, slightly anisotropic arsenean pyrite, and/or
anhedral to lath-like, rhombic or acicular and strongly
anisotropic marcasite. Where these minerals co-exist, pyrite is
early and marcasite generally overgrows and therefore post-dates
the arsenean pyrite (Figure 4b).
These sulfides overgrow and replace pyrite that formed during
dolomitisation of serpentine and sulfidation of iron oxides. They
also form minute intergrowths with quartz after dolomite,
especially where the dolomite has replaced talc and they fill
interstitial spaces between dolomite grains. This is especially
where some silicification has taken place, indicating volume loss
during quartz replacement and possibly also during
dolomitisation.
Dolomite alteration and mineralisation
The dolomite is sparry to micritic and has undergone initial
decarbonatisation that has resulted in the formation of collapse
breccias with clasts of dolomite set in a matrix of fine sparry
dolomite. Although silicification of the dolomite adjacent to the
main structures is commonly recorded by site geologists, in thin
section only trace amounts of quartz were observed to have
replaced dolomite. Dissolution of the matrix carbonate has
resulted in the formation of discontinuous stylolite-like fractures
that are filled by insoluble material including rutile and minor
arsenean pyrite. Minor kaolinite fills open spaces in these
fractures.
Dolomite breccias are a common feature in the hangingwall of
the main orebody and along the F145 fault above the Jianchaling
deposit. Mineralisation within the breccias tends to be narrow
and erratic.
DEPOSITIONAL SEQUENCES
The sequence of open space deposition including veins, veinlets,
breccias and vugh fill at Jianchaling is illustrated in Figure 6.
This sequence is similar to that observed in the wallrock
replacement alteration, with early depositional events associated
with sepentinisation, followed by dolomitisation then later
quartz-dominated depositional events associated with gold-nickel
mineralisation. Final stages of deposition are dominated by
calcite (Figures 5a-d).
The serpentinite is cut by generally discontinuous veinlets of
magnetite, which have undergone subsequent alteration to
hematite, then pyrite and arsenic-rich pyrite or marcasite during
the gold event. Locally magnesite veinlets cross-cut these
magnetite veinlets.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 147

M ERCEG, F KAIQIANG and T LEACH

A B

C D
FIG 5 - Jianchaling gold mine: handspecimen ore samples. (A) Dark green serpentinite cut by carbonate + regular vein with dark sulfide halo
that extends to pale green dolomite alteration. (B) Dark fine-grained sulfide rich ore cut by later carbonate + realgar veining. (C) Brecciated
dark sulfide ore with matrix of polyphasal carbonate and realgar. (D) Core sample of dark green sepentinite cut by thin sulfide + carbonate
veinlet with dark sulfide halo that extends to asymetric light green selvedge of dolomite alteration.

Dolomite-pyrite veinlets cut magnetite-magnesite veinlets and
are interpreted to be the first depositional phase that is associated
with the gold event. Fine-grained, granular quartz, locally with
minor abundances of dolomite, occur in later veinlets, and in
some places exhibit weak crustiform banding.
Silicification is locally accompanied by nickel and trace gold
mineralisation. In areas of intense silicification, shearing is
accompanied by early quartz-pyrite deposition, and subsequently
cross-cut by later shears and chrome-sericite (fuchsite) formation
and nickel mineralisation. In these shears, pyrite is locally
overgrown by millerite, then arsenean pyrite/marcasite, which in
turn by abundant late-stage gersdorffite. Sericite-sulfide shears
are cut by late-stage quartz-dolomite veinlets (Figure 4c). Sparse
quartz veinlets, that cut early dolomite veinlets, are intergrown
with chlorite, marcasite and millerite, the latter with minute
inclusions of gersdorffite.
Chlorite has formed after talc in some areas adjacent to
quartz-chlorite-sulfide shear zones. Chlorite alteration of talc is
accompanied by formation of abundant marcasite as well as
millerite and gersdorffite. The nickel sulfides are also intergrown
with dolomite (after serpentine) and intergrown and locally
overgrown by arsenean pyrite. In the latter case a light grey
isotropic sulfide, possibly galena, rims and overgrows the
gersdorffite in places.
The last stage of deposition is a calcite-dominated event with
early stage calcite veining containing intergrowths or inclusions
of orpiment, realgar and minor pyrite (Figure 5b). The metallic
metal native arsenic, though generally oxidised to a black sooty
colour, forms anhedral hexagonal plates and fibrous to acicular
and radiating aggregates (after arsenopyrite?) that are intergrown
with and also overgrow realgar in fractures and open vughs
(Figure 4d). Later finely banded calcite vein events are devoid of
sulfides, and locally form vein breccias (Figure 3d).
Nickel mineralisation
Anomalous nickel is locally associated with the gold ore and
occurs mainly as gersdorffite and minor millerite. Petrology
studies indicate that the nickel, which is associated with the gold,
has been derived from the alteration of pentlandite-chalcopyrite-
pyrite-pyrrhotite-magnetite mineralisation that formed during the
cooling of the Archean ultramafic intrusions.

148 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE JIANCHALING MINE, SHAANXI PROVINCE, PEOPLES REPUBLIC OF CHINA
FIG 6 - Alteration and mineralisation zones.
Ore samples from a nickel mine one kilometre southwest of
the Jianchaling mine consist of discontinuous stringers and
disseminated grains of sulfides in a completely serpentinised
ultramafic host rock. The sulfide minerals consist of pentlandite
that are intergrown with and/or overgrow interlocking
pyrrhotite-pyrite-chalcopyrite-magnetite. It is postulated that this
massive mineralisation was formed as sulfide minerals exsolved
from a cooling ultramafic melt during Archean emplacement.
High-grade nickel (>1 per cent Ni) intercepts in drilling at
Jianchaling mine are very similar to the ore zones from the
nickel mine, however in these cases, pentlandite has been
replaced by gersdorffite millerite and magnetite by hematite.
The nickel in the gold ore at Jianchaling is therefore interpreted
to have been mobilised from this much earlier stage of
mineralisation.
Textural relationships illustrate that gold and nickel
mineralisation events at Jianchaling were contemporaneous. The
association with arsenic in both nickel (gersdorffite) and
auriferous sulfides (arsenean pyrite and marcasite) further
supports this relationship.
CONCEPTUAL MODEL
Dilation along the F145 has facilitated the migration of ore fluids
in the region of the Jianchaling mine. The deposition of
dolomite, quartz and calcite in addition to nickel sulfides in veins
associated with this fault system, indicates the fluids were
transporting silica, carbonate, nickel, as well as arsenic and gold.
It is speculated that, through fluid-rock reactions, the
gold-arsenic mineralised fluids that migrated predominantly
through calcareous sediments were enriched in Ca/Mg carbonate,
whereas those that migrated through ultramafic rocks were
preferentially enriched in nickel.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Migration of these fluids through the serpentinised ultramafic
rocks set up a geochemical gradient that is zoned from a core of
quartz plus gold, arsenic and nickel sulfides, to a peripheral
dolomite plus pyrite and nickel sulfides (with some alteration to
hematite) zone. This replacement zonation reflects a migration,
along this gradient of H2O, SiO2, CaO, CO2, HS-, As, and Au from
fluid pathways, and the migration of MgO from the serpentinite.
FeO and Cr are fixed as pyrite and fuchsite (Figure 6).
Initially it was thought that the magnesium in carbonate veins
was derived from the silicification of the sediments. However the
paucity of carbonate dissolution and/or silicification in the
dolomite suggests that the magnesium associated with dolomite
veining was more likely derived from the much more extensive
quartz alteration of dolomite in altered serpentinite.
It is most likely that the nickel was derived from prexisting
pentlandite in the serpentinite, and this is confirmed by the
presence of nickel sulfide inclusions in pyrite associated with the
dolomitisation of the serpentinite minerals. The sulfidation of
this early nickel results in the formation of millerite and
gersdorffite immediately adjacent to and within fractures.
Gold deposition is interpreted to be the result of sulfidation of
iron, mainly in the form of magnetite and chromite in the
serpentinite and iron in serpentine. This iron was initially
released and made reactive during dolomite and pyrite alteration
of serpentinite and the alteration of magnetite initially to
hematite then to pyrite (Figure 7). Arsenian pyrite and marcasite
then rimmed and replaced pyrite in zones of dolomite dissolution
and silicification. The gold is reported (Vielreicher et al, 2002) to
be concentrated within the structure of the arsenic-rich pyrite
grains and rims with concentrations of up to 2500 ppm gold. It is
not known whether gold is enriched within contemporaneous
NiAsS minerals.
149
M ERCEG, F KAIQIANG and T LEACH

FIG 7 - Hydrothermal model.

The same gold and arsenic bearing fluid in contact with the
calcareous sediment has resulted in localised and minor
carbonate dissolution and quartz replacement. This was
accompanied by only trace arsenian pyrite formation in
dissolution fractures and breccia matrix in response to sulfidation
of reactive iron. The paucity of arsenic rich sulfides and nickel
mineralisation and corresponding low grades in the dolomite is
interpreted to reflect the very low iron content in these
sediments.
It is most likely that ore fluids were derived from sedimentary
rocks during the late phases of Yanshanian orogenic activity. Ore
fluid movement was controlled by regional structures through
relatively impermeable host rocks. The absence of Yanshanian
igneous rocks in the district makes it unlikely that the ore fluids
had a strong magmatic component.
Fluid inclusion studies of Carlin-type deposits elsewhere in
China (Hu Rui-Zhong et al, 2001) indicate that during
mineralisation fluid temperatures were in the range 250 - 150C,
salinities ranged from six to three wt per cent equivalent NaCl,
and CO2 was the major gas present. The ore fluids were
sufficiently acidic (pH ~4 - 6) to dissolve calcite. At these
conditions, the lack of appreciable silver or basemetals in the
ores and the presence of gold, arsenean pyrite and arsenic
minerals indicate that the ore fluids had a relatively high H2S
concentrations and that gold was transported as a bisulfide
complex.

150 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE JIANCHALING MINE, SHAANXI PROVINCE, PEOPLES REPUBLIC OF CHINA
A CARLIN-TYPE DEPOSIT?
The important features that characterise the Jianchaling gold
deposit are:
strong structural control within a major regional fault;
hosted within sedimentary and ultramafic rocks;
no obvious association with igneous rocks;
complex back-arc, pull-apart basin setting;
gold is micron size (refractory ore);
dominant sulfides are pyrite, arsenian pyrite and marcasite;
dolomitisation and silicification are main alteration
processes;
late Mesozoic age (Yanshanian);
Au-As-Hg geochemical association; and
associated Ni mineralisation (millerite and gersdorffite).
These characteristics compare favourably with those well
documented for other Carlin-type deposits in China (Zhiping and
Peters, 1998). However the association of nickel with gold
mineralisation is unusual and reflects the local geology.
In addition the primary mechanism of gold precipitation at
Jianchaling is postulated to be the reaction of an HS- (+Au +As)
bearing fluid with Fe in serpentinite to form auriferous arsenian
pyrite. Serpentinite is not a common host rock in other Chinese
Carlin-type deposits. An alternative mechanism for gold
precipitation and/or supply of Fe may apply in other deposits.
ACKNOWLEDGEMENTS
The authors are grateful to Greg Jones (former Asset Manager)
and Chris Cairns (former Geology Manager) with Sino Gold
Limited whose early support formed the foundations of this
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
paper and to Rod Hanson (Jianchaling Mine General Manager)
for continued encouragement. Thanks to Phillip Uttley (Chief
Geologist) and Dr Bob Ilchik (Senior Geologist) for critically
evaluating the paper. Thanks also go to the Jianchaling geology
team Zhang Xinsheng, Zhao Minsheng and Liu Junsheng who
know the deposit better than anyone.
REFERENCES
Hu Rui-Zhong, Su Wen-Chao, Bi Xian-Wu, Tu Guang-Zhi and Hofstra,
A H, 2002. Geology and geochemistry of Carlin-type gold deposits
in China, Mineralium Deposita, 37:378-392.
Ratschbacher, L, Hacker, B R, Calvert, A, Webb, L E, Grimmer J C,
McWilliams, M O, Ireland, T, Shuwen Dong and Jianmin Hu, 2003.
Tectonics of the Qinling (Central China): Tectonostratigraphy,
geochronology, and deformation history, Tectonophysics, 366.
Vielreicher, R M, 1999. The Jian Cha Ling Ni-Au project, SW Shaanxi
Province, central PR China: An example of epizonal,
listvenite-hosted gold mineralization, unpublished company report to
Sino Mining Australia Pty Ltd.
Vielreicher, R M, 2000. The Jian Cha Ling Ni-Au Project, SW Shaanxi
Province, Peoples Republic of China: comments on metamorphism,
hydrothermal alteration, timing of intrusive rocks, gold ore
assemblage and nickel mineralisation at JCL, unpublished company
report to Sino Mining Australia Pty Ltd.
Vielreicher, R M, Vielreicher, N M, Hagemann, S G and Jones, G, in
press. Fault asperity as the key to understanding
structurally-controlled, ultramafic rock-hosted, disseminated gold
mineralisation at Jian Cha Ling, central China, Mineralium Deposita.
Zhiping Li and Peters, S G, 1998. Comparative geology and geochemistry
of sedimentary-rock-hosted (Carlin-Type) gold deposits in the Peoples
Republic of China and in Nevada, USA, US Department of the
Interior, US Geological Survey, Open-File Report 98-466.
151
Distribution of Alteration and Mineralisation in Northern Carlin
Trend Gold Deposits, Nevada
T M Leach1

ABSTRACT
A series of petrological studies were carried out on the characteristics of
alteration and mineralisation in mainly structural-controlled (Deep Star,
Meikle, Deep Post) and predominantly lithological-controlled
(Post-Betze-Screamer and West Leeville) gold deposits in the northern
district of the Carlin trend, Nevada. Alteration and mineralisation in these
deposits are hosted in a sequence of calcareous clastic units of the
Devonian Popovich Formation, into which have been emplaced a number
of Jurassic intrusions, and syn- to post-mineral Eocene dikes.
Four hydrothermal events have been recognised in these deposits:
1. an initial Devonian base metal system and early diagenesis;
2. a Mississippian stage of alteration in conjunction with burial
metamorphism;
3. contact and hydrothermal alteration and base metal-gold
mineralisation associated with the emplacement of the Jurassic
intrusions; and
4. a Carlin-style gold system related to Eocene magmatism.
Early stage replacement alteration during the Carlin event is zoned from
proximal to distal from inferred fluid source as dominated by: residual
calcite dolomite arsenean pyrite illitic clay + arsenean pyrite
quartz + arsenean-pyrite/marcasite kaolinite + arsenean-marcasite
clays + Fe/Mg-carbonates arsenean-marcasite de-calcification. This
early stage of replacement alteration was the main gold mineralising event.
These replacement zones are overprinted by depositional assemblages that
are zoned from proximal to distal environments as dominated by: quartz
kaolinite/smectite As-sulfides calcite/barite, and which grade from
late-mineral to post-mineral episodes.
FIG 1 - Gold deposits of the Northern Carlin Trend, Nevada
The replacement assemblages are inferred to have formed in response (modified from Teal and Jackson, 1997).
to the progressive decrease in pH, of a moderately cool (200 - 250C),
mineralising fluid. They are also inferred to have formed in response to
cooling and sulfidation processes proximal to upflow zones, and somewhat lower grade, predominantly lithological- controlled
subsequence neutralisation caused by wallrock reaction in more distal deposits (Betze, 900 t @ 7.5 g/t Au; Screamer, 160 t @ 5.0 g/t Au;
outflow environments. It is proposed that the depositional zones were
caused by cooling of an outflowing fluid, and possible late stage mixing
and West Leeville, 180 t @ 15 g/t Au).
with surficial waters in distal settings.
These zonations in alteration and mineralisation can be a useful tool in GEOLOGICAL SETTING
the exploration for Carlin-type gold deposition.
The deposits of the northern Carlin trend are aligned along a
series of N to NNW striking faults; including the Leeville Fault
INTRODUCTION to the SE, the Beast-Genesis and Post faults in the central parts
of the district, and the Bootstrap-Dee faults in the NW. The
The Carlin Trend is a 60 km-long north-northwest alignment of
deposits are bounded to the south-west by the NW-striking
mainly sedimentary rock-hosted gold deposits that are located in
Castle Reef Fault.
north-eastern Nevada (Teal and Jackson, 1997). The trend is
made up of three separate clusters of ore deposits of which those Alteration and mineralisation in these deposits are hosted
in the northern Carlin trend are by far the most abundant and mainly in calcareous clastic units of the Devonian Popovich
largest (Figure 1). The deposits of the northern Carlin trend are Formation, which grades upwards into calcareous mudstone and
distributed over a strike length of more than 20 km from siltstone of the Devonian Rodeo Creek Unit, and are underlain by
dolomitic siltstones of the Silurian/Devonian Roberts Mountain
Bootstrap in the northwest to the Pete deposit in the southeast.
formation and massive dolomite units of the Ordovician Hanson
The total resource (past production and reserves) for the northern
Creek Formation. Fine-grained, intercalated chert, siltstone,
Carlin deposits is in excess of 70 Moz, making this district mudstone and local limestone units of the Ordovician Vinini
among the largest concentrations of gold in the world (Bettles, Formation have been thrust over the Rodeo Creek units along the
2002). Roberts Mountain Fault. Overlying the Palaeozoic rocks are
This paper is based on a series of petrological studies carried rhyolite tuffs, silts and gravels of the Tertiary Carlin Formation.
out on a number of deposits in the northern Carlin district over the The emplacement of pre-mineral Jurassic lamprophyre, and
past four years. This work was done on both high grade, mainly monzonite to diorite intrusions of the Goldstrike and Little
structural-controlled deposits (Meikle, 220 t Au @ 24.7 oz/t; Deep Boulder Stocks have formed hornfels and skarns at the contact
Star, 50 t @ 34 g/t Au; Deep Post, 84 t @ 25 g/t Au), and with the Palaeozoic sediments. Eocene to Oligocene (36.3 - 40.3
Ma; Bettles, 2002) biotite feldspar and rhyolite dikes locally host
1. Consultant, Research Associate, Department of Earth Sciences, mineralisation and it is postulated (Ressel et al, 2000) that these
Waikato University, Hamilton, New Zealand. dikes are genetically related to the gold mineralisation.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 153

T M LEACH
Meikle Deposit
The high grade, strongly structurally-controlled, underground
Meikle orebody is located within the northern extension of the
Post Fault system. Alteration and mineralisation are mainly
hosted in a series of silicified collapse breccias that occur within
the Popovich and Roberts Mountain Formation (Emsbo, 1999).
The breccias are structurally-controlled and associated with NW
and NE orientated high angle faults (Teal and Jackson, 1997).
Gold mineralisation is also hosted in Jurassic lamprophyre and
monzonite dikes that have been emplaced along the footwall of
the Post Fault.
Post-Betze-Screamer
These lower grade, mainly lithologically-controlled deposits are
hosted in the Devonian carbonate and clastic sedimentary rocks
that occur along the northern margins of the Goldstrike intrusion.
Irregular dikes and sills of the Goldstrike intrusion penetrate the
Palaeozoic sediments. Around the margins of these intrusions,
the sediments have been locally metamorphosed to skarn,
calc-silicate hornfels and marble. Alteration and mineralisation is
developed within the sediments and contact aureole, and to a
much lesser extent within fractures in the intrusion.
The deposit is divided into three orebodies: Post and Deep
Post are located to the east, and are bounded by the Post Fault;
Betze, which forms the central portion; and Screamer to the
west. Ore grade mineralisation is however continuous, forming
mainly a flat lying zone that extends over more than 2000 m
from east to west, with the Betze orebody occurring in an
up-faulted block between Screamer and Post (Bettles, 2002).
Deep Star
The high grade, underground Deep Star deposit is located within
a structural corridor between the Post and Beast-Genesis Faults,
at the SE margin of the Goldstrike Stock. The deposit is mainly
hosted in brecciated calc-silicate rocks of the Popovich
Formation, with minor ore extending into the intrusion (Heitt et
al, 2003). The emplacement of syn- and post-ore (38.3 Ma)
rhyolite dikes along bounding faults constrain the age of the
mineralisation.
West Leeville
The West Leeville deposit is currently being developed by
Newmont as an underground mine and is located in the southern
part of the northern Carlin gold district, approximately 2.3 km
north of the Carlin Mine. The deposit forms two flat lying six to
40 metre thick orebodies that are hosted in the upper and lower
portions of wispy laminated, silty limestone units of the Lower
Popovich/Upper Roberts Mountain Formations. The deposit
occurs along the footwall of the Bounding Fault, and is best
developed where this fault is intersected by the
northwest-striking Rodeo Creek Fault (Jackson et al, 1997)
TEMPORAL DISTRIBUTION IN ALTERATION AND
MINERALISATION
Four distinct hydrothermal events have been recognised in the
northern Carlin deposits (Figure 2) and these can be summarised
as the following.
Devonian base metal system
Base, and locally precious metal, mineralisation occur as
horizontal zones that are hosted in upper Popovich calcareous
mudstones and, in a number of the northernmost deposits of this
gold district, including Meikle and locally at Screamer, as
underlying discordant veins. This event consists of early stages
of ferroan (zebra) dolomite replacement of carbonate, an
154 Adelaide, SA, 19 - 22 September 2004
intermediate stage of sulfidation mineralisation (pyrite =
sphalerite > galena = boulangerite > tennantite), barite alteration,
silicification and associated collapse brecciation. There is also a
late stage stockwork barite-carbonate-quartz-sulfide vein
development. Gold occurs both in the native form and
encapsulated in pin-stripped pyrite. Many features of this early
alteration and mineralisation event suggest a MVT-style of
system (also Evans, 2000), although Emsbo (1999) favours an
exhalative (SEDEX) process.
In the Meikle area, karsting during seal level changes, initial
brecciation, calcite cementation and subsequent authigenic
quartz, dolomite, barite and pyrite deposition have been
attributed (Emsbo et al 2003) to early diagenesis.
Burial metamorphism
A polyphasal event of quartz (commonly strained) + carbonate
(dolomite, ankerite) vein formation accompanied by hydrocarbon
migration; and dissolution fracturing and silicification cross-cuts
the base metal mineralisation. These features are especially
abundant in the more siliceous units of the Rodeo Creek
Formation and in the upper plate siltstone and mudstones of the
Vinini Formation. The formation of these veins has been
attributed (Emsbo, et al, 2003) to deep burial metamorphism
during the early Mississippian.
Jurassic magmatism
The emplacement of the Jurassic intrusions has locally resulted
in the formation of zoned contact skarns, calc-silicate hornfels
and marbles, especially in the Betze deposit on the northern
margin of the Goldstrike Stock, and at Deep Star. This
magmatism was followed by a hydrothermal event that
comprises early, and local (eg at Deep Star) propylitic alteration
(epidote-chlorite-carbonate), followed by a widespread event of
quartz sericite pyrite dolomite alteration of intrusions,
hornfels/skarns, and sediments, and local intense silicification of
marble (Betze, Post) and upper limestone units (eg West
Leeville). Fracture-controlled base metal (pyrite sphalerite
galena bournonite boulangerite tennantite/tetrahedrite
chalcopyrite bismuthinite) mineralisation is associated with
quartz dolomite barite deposition in veins and breccias. Very
rare native gold/electrum mineralisation with these vein occurs in
these veins/breccias at Meikle (Emsbo, 1999) and elsewhere.
These veins cut both intrusions and host sediments, and are
attributed to the waning stages of Jurassic magmatism.
Eocene Carlin-type gold system
The main stage of gold mineralisation in this district is
associated with an Eocene hydrothermal system, which, at the
Carlin mine itself, is of course the type example of Carlin-style
gold deposits. This system shows evidence of a gradation from
an early episode of host rock replacement, to a later series of
polyphasal depositional events.
The replacement event can be separated into initial calcite
dissolution and local dolomitisation, followed by a stage of
quartz alteration/silicification (mainly of carbonate, both primary
and secondary, and calcareous sediments) and clay
alteration/argillisation, seen predominantly in the intrusions and
silicate-rich skarn and hornfels lithologies. Sulfidation of iron,
which was made reactive during earlier Carlin and/or previous
events, resulted in pyrite and arsenean pyrite/marcasite
formation, with associated gold mineralisation, and was syn- and
post-silicification and argillisation.
Early stages of deposition are sulfide-rich, consisting of
As-rich pyrite and marcasite in quartz illitic clay/kaolinite
veinlets and breccias, and are considered to be late-stage (gold)
mineral events. Realgar and orpiment are locally (especially at
PACRIM 2004
 DISTRIBUTION OF ALTERATION AND MINERALISATION IN NORTHERN CARLIN TREND GOLD DEPOSITS, NEVADA

FIG 2 - Sequence of events in northern Carlin deposits.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 155

T M LEACH
Betze, Deep Star and Deep Post) associated with later
carbonate-kaolinite, whereas stibnite was deposited very late
with quartz, calcite and/or rare barite. These late events are not
considered to be significantly auriferous (Cline, 2001).
Post-mineral fractures and breccias are sealed by barite,
kaolinite, smectite, Fe-carbonates (ankerite, siderite) and very late
calcite, gypsum and/or bitumen. Emsbo and Hofstra (2003)
consider late stage barite and calcite in the northern-most deposits
to be associated with a Miocene hydrothermal event. However at
Betze and Deep Post, barite is very locally intergrown with
arsenic/antimony-sulfide minerals and is therefore, at least in
places, associated with the final stage of the Carlin gold system.
SPATIAL DISTRIBUTION IN CARLIN-STAGE
ALTERATION AND MINERALISATION
In all the deposits studied, the Carlin gold event has been
overprinted onto alteration and mineralisation assemblages that
were formed during earlier hydrothermal events. At Screamer
and West Leeville, in most parts, the early events were not
strongly developed and therefore the recognition of the features
related to the Carlin system are readily apparent. However,
elsewhere (eg Meikle, Betze), some, if not all, of the earlier
formed assemblages are widespread and intense, and it is
difficult to confidently assign some mineral assemblages to a
particular event.
Replacement assemblages
The overall spatial distribution of alteration and mineralisation
during the Carlin-type replacement events, from distal to
proximal to the inferred source of the mineralising fluids, is
illustrated in Figure 3 and can be summarised as follows.
FIG 3 - Distribution of replacement alteration in northern Carlin deposits.
156 Adelaide, SA, 19 - 22 September 2004
Residual calcite aone
This zone occurs in the central part and deep in some of the
deposits (eg Betze, Deep Post) and consists of residual calcite
and other carbonate minerals (dolomite, ferroan dolomite and
ankerite), minor silicification and arsenean pyrite mineralisation
in dissolution vughs and fractures. It is speculated that the
Fe-carbonate minerals may be derived form the release of iron
during pre-Carlin alteration events in a manner similar to that
proposed by Fortuna et al (2003) at the Twin Creeks deposit.
Although alteration is weak, gold grades can locally be very high
(>1 - 2 oz/t Au), probably due to proximity to the source of the
mineralising fluids.
Dolomite zone
Both residual and hydrothermal replacement dolomite,
Fe-dolomite and/or ankerite occur in this zone, with variable
amounts of arsenean pyrite and secondary quartz and illitic clay.
All calcite was dissolved.
Illitic clay zone
This zone consists of illitic clay (illite, interlayered illite-smectite
with a low smectite content), arsenean pyrite, pyrite, quartz that
increases in abundances outward), and varying abundances of
residual carbonate that increases inwards towards the inner
dolomite zone. Most of the ore at Deep Post is hosted in the
illitic clay and dolomite zones (Figure 3).
Quartz zone
Quartz dominates over clay minerals in this zone, and the gold
mineralisation is associated with the formation of both
arsenean-pyrite and arseneanmarcasite. Silicification is most
PACRIM 2004
 DISTRIBUTION OF ALTERATION AND MINERALISATION IN NORTHERN CARLIN TREND GOLD DEPOSITS, NEVADA
intense in more carbonate-rich calcareous sediments and
marbles, and weaker in the intrusions, calc-silicate hornfels and
argillaceous sediments. Kaolinite increases in abundance in more
distal settings, whereas illitic clay co-exists with quartz in more
proximal settings. This zone hosts most of the orebody at Post,
Betze and Meikle (Figure 3), although in the latter case the
intense silicification may be attributable to pre-Carlin events.
Kaolinite zone
Kaolinite is the main replacement mineral and is commonly
accompanied by mainly marcasite mineralisation. Quartz
decreases in abundance in more distal settings.
Kaolinite-marcasite alteration is especially strong where hosted
by silicate-rich rocks (intrusions, skarns and argillaceous
sediments) and hosts most of the ore at the Deep Star and West
Leeville deposits (Figure 3).
Clay Fe/Mg-carbonate zone
The kaolinite zone grades outwards to a zone of residual
dolomite, secondary siderite and possibly ankerite, smectite and
rare marcasite. Alteration is generally weak in this zone, which
hosts the bulk of the mineralisation at Screamer.
De-calcified zone
Calcite dissolution and residual dolomite are features of this
zone. The extent of this zone varies considerably from deposit to
deposit.
Depositional assemblages
The distribution of minerals deposited in open spaces (vughs,
fractures, breccias) is illustrated in Figure 4. These mineral
assemblages post-date the replacement events and have therefore
FIG 4 - Distribution of deposition alteration in northern Carlin deposits.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
been superimposed on the earlier assemblages. The generalised
zonation in deposition minerals from deep and proximal to
upflow zones and early in the sequences of deposition, outward
to shallow, and distal environments and late in the depositional
sequence is summarised as follows:
1. quartz ( chalcedony) and/or illitic clay arsenean pyrite;
2. kaolinite and/or smectite As-marcasite/As-pyrite;
3. As-sulfides (realgar, proximal orpiment, distal) calcite
and/or barite; and
4. calcite and/or barite stibnite siderite/ankerite.
During the waning stages of the system, there can be
considerable telescoping of the distal/later depositional
assemblages onto those that were deposited earlier and or more
proximal to inferred fluid sources.
There is also a zonation in the nature of depositional events,
from fractures and local breccias (crackle to rotational) proximal
to upflow zones, to dominantly vugh fill in more distal
environments.
CONCEPTUAL MODELS FOR THE DISTRIBUTION
IN ALTERATION AND MINERALISATION
The postulated deep (>2 - 4 km) environment interpreted for the
Carlin deposits (Kuthen and Rose, 1992; Lamb and Cline, 1997)
has in the past precluded an epithermal environment being
invoked as a model for Carlin-type gold-deposits. However
recent studies on Eocene intrusions (Ressel et al, 2000),
paleo-reconstructions (Hickley et al, 2002), alteration textures
(Nutt and Hofstra, 2003) and alteration zonations (this paper), all
indicate a much shallower (<1 km) environment for Carlin
systems than has been previously recognised. In addition the
suite of trace elements associated with Carlin-type deposits
(Au-Ag-As-Sb-Tl-Hg) is identical to that associated with
157
T M LEACH
classical epithermal gold deposits (Cooke and Simmons, 2000)
formed at shallow (<1 km) crustal levels. A premise that Carlin
deposits formed at shallow levels has a significant affect on
understanding the processes that could have been involved
during alteration and mineralisation.
Wallrock replacement
Based on isotope and fluid inclusions data, Woitsekhowskaya
and Peters (1998) postulate that a hot (250C), low pH (4.4)
fluid, rising from depth to form the Post-Betze deposit. However
a central alteration zone of residual carbonate (Betze, Deep Post)
and a peripheral, and late, kaolinite alteration zone (all deposits),
strongly indicates that the original mineralised fluid probably
had a near neutral pH. It is most probable that the calcareous
sediments within the underlying Roberts Mountain and Hanson
Creek formations buffered the pH of the upwelling fluid. In
addition, the sequence of events during the evolution of the
Carlin system indicates that the fluid was initially in equilibrium
with carbonate and illitic clay, and therefore near neutral, and
became progressively more acidic in more distal settings.
As these fluids migrated to cooler, shallow levels, the pH of
the fluid would however have decreased and thereby initiated
dissolution of the carbonates. Iron released from the ferroan
dolomite and ankerite would then have been available for
sulfidation to take place. Sulfidation of reactive iron involves the
decrease in fluid pH and therefore increases the potential for
further carbonate dissolution. It is widely accepted that
Carlin-type gold mineralisation (Hofstra and Cline, 2000) is
closely intimately associated sulfidation of available iron by the
reaction :
Au(HS)- + Fe2+ + H2S H+ + FeS2 + H2 + Au0
In the northern Carlin deposits, this reactive iron was made
available for sulfidation to take place, from the earlier alteration
of minerals such as Fe-rich carbonate species, previously formed
sulfides, mafics in intrusions, and/or Fe-rich silicate minerals in
skarns and hornfels.
Cooling of the upwelling fluid, at these shallow levels in the
system, possibly accentuated by mixing with surficial waters,
would have resulted in a decrease in silica solubility and quartz
replacement of the carbonate. This silicification would have
resulted in further release of iron for sulfidiation processes and
therefore continued to decrease the fluid pH as it flowed laterally
and rose to shallower levels.
Proximal to the site of initial sulfidation and decrease in fluid
pH, the mineralising fluid would have been in equilibrium with
illitic clay. However in settings peripheral to sites of maximum
silicification, the fluid pH would have decreased significantly to
a pH of approximately four, thereby favouring kaolinite
formation. The association of marcasite, rather than pyrite, with
kaolinite and quartz indicates a more oxidising environment in
these peripheral alteration zones.
Although kaolinite is common in the northern Carlin deposits,
it is rare to absent in deposits elsewhere (eg Jerritt Canyon, Twin
Creeks, Getchell; Hofstra and Cline, 2000). The lack of kaolinite
in these deposits could be interpreted to indicate that a more
neutral (pH >5) upwelling fluid was associated with these
deposits. This may be a reflection of deeper/hotter environments
and/or lesser sulfidation than that which is necessary for
acidification and kaolinite formation.
The progressive sulfidation would have eventually resulted in a
decrease in sulfur activity of the outflowing fluid. At some point,
fluid neutralisation would therefore have dominated over
sulfidation and associated decrease in fluid pH. The occurrence
of smectite rather than kaolinite indicates a more neutral, and
cool (<100 - 150C, Corbett and Leach, 1998), environment in
these distal settings. The formation of Fe-rich carbonates
158 Adelaide, SA, 19 - 22 September 2004
(siderite/ankerite) with kaolinite and smectite further supports a
distal neutralisation process and also indicates mobilisation of
CO2 and iron by the low pH fluid out to the margins of the
system. Calcite dissolution continued until the outflowing fluid
was totally buffered by the host calcareous sediments.
Depositional events
The distribution of deposition-style alteration and mineralisation
(Figure 4) is interpreted as reflecting the progressive cooling of
upwelling hot fluid in shallower and/or more distal regions of the
system. This fluid was isolated from substantial rock reaction,
since the host intrusions and carbonate-bearing rocks had
undergone prior replacement alteration.
Illitic clays + quartz + sulfides were deposited at deeper levels
and early in the depositional sequence. The occurrence of illitic
clays implies that the upwelling fluid associated with the
depositional event was initially less neutral than that which
resulted in earlier replacement, probably due to some sulfidation
at deeper levels. Evidence for this deep sulfidation was observed
beneath the Meikle and Deep Star deposits. As in the
replacement events, quartz deposition was probably associated
with cooling. The occurrence of arsenean pyrite with quartz and
illitic clay indicates that significant iron was mobilised during
the late stage depositional events, at least proximal to the upflow
zone.
A decrease in pH as the fluid cooled at shallower levels and
during later depositional events resulted in kaolinite and/or
smectite becoming dominant over illitic clay and quartz. The
decrease in pH probably further facilitated mobilisation of iron,
with the deposition of arsenean marcasite over arsenean pyrite in
these lower pH more oxidising conditions.
The local deposition of realgar and orpiment (Deep Star,
Betze, Deep Post) indicate high levels of arsenic mobilisation by
the late stage mineralising fluid within feeder structures.
However these As-sulfide minerals are rare to absent in deposits
that are interpreted to have formed in more distal outflow
settings (eg West Leeville, Screamer). The association of these
arsenic-sulfides locally with barite implies that the deposition of
realgar and orpiment took place under significantly oxidising
condition, thereby inhibiting the mobilisation of gold. The
zonation of realgar to orpiment in more distal settings suggests a
progressive decrease in arsenic content of the outflowing fluid.
A late, outer zone of calcite Fe-Mg-carbonate and barite is
postulated to have formed in response to neutralisation of a
CO2-rich fluid within a cool oxidising environment. The
occurrence of stibnite peripheral to the arsenic-sulfide
depositional zone is similar to that encountered in classical low
sulfidation precious metal systems (Corbett and Leach, 1998)
and supports the inference of cool conditions around the margins
of the system. The local presence of gypsum in late stage clay
bitumen breccias (Meikle; Deep Post; Betze) implies that sulfate
deposition continued well into the post-mineral events, albeit at
much lower temperatures (<100 - 150C).
In Figure 4, it is shown that these depositional mineral
assemblages occur at progressive distances from the inferred
upflow zones. However in some of the deep-seated,
structural-controlled deposits there has been considerable
telescoping of the depositional assemblages due to pressure
draw-down, as the hydrothermal system waned during its latter
stages of evolution.
Isotope analyses of barite in Carlin-type deposits (Hofstra and
Cline, 2000) and of kaolinite at Deep Star and Deep Post
(Thompson, 2003) support a late stage mixing of the deep
hydrothermal fluid with surficial waters. However, there is an
overall lack of evidence for active phase separation (boiling)
during the formation of Carlin-type gold deposits (Hofstra and
Cline, 2000).
PACRIM 2004
 DISTRIBUTION OF ALTERATION AND MINERALISATION IN NORTHERN CARLIN TREND GOLD DEPOSITS, NEVADA
ALTERATION ZONES AND GOLD
MINERALISATION
Figure 3 shows a plot of the generalised distribution of
replacement aleration zoning relative to the ore zones for
selected northern Carlin deposits. It illustrates that there is very
little correlation between alteration zones and gold
mineralisation. At Deep Star, the ore zone coincides with intense
kaolinite-marcasite-quartz alteration (Heitt et al, 2003; Clode et
al, 1997), whereas at West Leeville (Jackson et al, 1997) and
Carlin (Kuhen and Rose, 1992) the ore zone occurs where
carbonate dissolution is intense, on the upper and lower margins
respectively, of the zones of intense silicification and
argillisation. At Meikle and Betze the ore zones closely correlate
with zones of intense silicification, whereas ore at Screamer
correlates with zones of residual Fe/Mg-carbonate and only weak
silicification and argillisation.
It is therefore postulated that release of reactive iron from the
host rocks is the controlling factor in gold mineralisation in
Carlin-type gold deposits. This is in turn controlled by the
amount of iron originally present in the host rocks, and the
availability of channelways for the mineralising fluid to access
this iron. These fluid pathways can occur along receptive
calcareous sediments, dilatant structures and/or dike contacts.
CONCLUSIONS
The northern Carlin deposits exhibit sequences of hydrothermal
events and zonations in replacement and deposition alteration
and mineralisation that are unique to Carlin-type gold systems.
Gold mineralisation is formed in response to sulfidation of
available iron, predominantly during the argillisation and
silicification replacement events, and extends into the early
quartz clay deposition events. However ore zones are not
restricted to any of the alteration zonations and may form from
proximal carbonate quartz zones outwards to distal clay
Fe/Mg carbonate zones.
Recognition of the sequence and styles and zonations in
alteration and mineralisation, when integrated with geological,
geophysical and geochemical data, can aid in the exploration and
development of Carlin-type gold deposits.
ACKNOWLEDGEMENTS
The author would like to thank Barrick Gold Corporation and
Newmont Mining Company for kindly granting permission to
present this paper, and Keith Bettles and Dr Ray Merchant for
patiently reviewing the manuscript prior to submission.
REFERENCES
Bettles, K, 2002. Exploration and geology, 1962-2002, at the Goldstrike
Property, Carlin Trend, Nevada, Society of Economic Geologists,
Special Publications, No 9, pp 275-298.
Cline, J S, 2001. Timing of gold and arsenic sulphide mineral deposition
at the Getchell Carlin-type gold deposit, North-Central Nevada,
Economic Geology, 96:75-90.
Clode, C H, Grusing, D G, Heitt, D G and Johnston, I M, 1997. The
relationship of structure, alteration and stratigraphy to formation of
the Deep Star deposits, Eureka County, Nevada, in Carlin-type gold
deposits field conference, Economic Geology Guidebook Series, (Ed:
T B Thompson) Vol 28, pp 239-256.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Cooke, D R and Simmons, S F, 2000. Characteristics and genesis of
epithermal gold deposits, in Gold in 2000, Society of Economic
Geology, Reviews in Economic Geology (Ed: S G Hageman and P R
Brown) Vol 13, pp 163-220.
Corbett, G C and Leach, T M, 1998. Southwest Pacific gold-copper
systems: structure, alteration and mineralization, Society of
Economic Geologist Special Publication No 6, 236 p.
Emsbo, P, 1999. Origin of the Meikle high grade gold deposit from
super-position of Late Devonian sedex and mid-Tertiary Carlin-type
gold mineralisation, PhD thesis (unpublised), 394 p, Colorado
School of Mines, Golden, Colorado.
Emsbo, P and Hofstra, A H, 2003. Origin and significance of posture
dissolution collapse breccias cemented with calcite and barite at the
Meikle gold deposit, northern Carlin trend, Nevada, Economic
Geology, 98:1243-1252.
Emsbo, P, Hofstra, A H, Lauha, E A, Griffin, G L and Hutchinson, R W,
2003. Origin of high grade gold ore, source of ore fluids components
and genesis of the Meikle and neighbouring Carlin-type deposits,
Northern Carlin-trend, Nevada, Economic Geology, 98:1069-1105.
Evans, D C, 2000. Carbonate-hosted breccias in the Meikle mine, Nevada
and their relationships with gold mineralisation, MsC thesis
(unpublished), Colorado School of Mines, Golden, Colorado, 266 p.
Fortuna, J, Kesler, S E and Stenger, D P, 2003. Source of iron for
sulphidation and gold deposition, Twin Creeks Carlin Type-Deposit,
Nevada, Economic Geology, 98(6):1213-1224.
Heitt, D, Dunbar, W W, Thompson, T B and Jackson, R G, 2003.
Geology and geochemistry of the Deep Star Gold Deposit, Carlin
Trend, Nevada, Economic Geology, 98(6):1069-1106.
Hickley, K, Haynes, S and Tosdal, R, 2002. Eocene reconstruction
project, Carlin-Jerritt canyon Region, NE Nevada, MDRU UBC,
internal Newmont report.
Hofstra, A H and Cline, J S, 2000. Characteristics and models for
Carlin-type gold deposits, in Gold in 2000, Society of Economic
Geology, Reviews in Economic Geology, (Eds: S G Hageman and P E
Brown) Vol 13, pp163-220.
Jackson, M, Lane, M and Leach, B, 1997. Geology of the West Leeville
deposit, Society of Economic Geologists, Guidebook Series, (Ed: T B
Thompson) Vol 28, pp 277-287.
Kuthen, C A and Rose, A W, 1992. Geology and geochemistry of
wall-rock alteration at the Carlin Gold Deposit, Nevada, Economic
Geology, 87(7)1697-1721.
Lamb, J B and Cline, J, 1997. Depths of formation of the Meikle and
Betze/Post Deposits, in Carlin-type gold deposits field conference,
Economic Geology Guidebook Series (Ed: T B Thompson) Vol 28,
pp 101-107.
Nutt, C J and Hofstra, A H, 2003. Alligator Ridge District, East-Central
Nevada, Carlin-type gold mineralisation at shallow depths, Economic
Geology, 98(6):1225-1242.
Ressel, M W, Noble, D C, Henry, C D and Trudel, W S, 2000.
Dike-hosted ores of the Beast deposit and importance of Eocene
magmatism in gold mineralisation of the Carlin trend, Nevada,
Economic Geology, 95:1417-1444.
Teal, L and Jackson, M, 1997. Geological overview of the Carlin Trend
Gold Deposits and descriptions of recent deep discoveries, in
Carlin-type gold deposits field conference, Economic Geology
Guidebook Series, (Ed: T B Thompson) Vol 28, pp 3-37.
Thompson, T B, 2003. Stable isotope geochemistry of kaolinite: Deep
Star and Deep Post Mines, Internal Newmont memo.
Woitsekhowskaya, M B and Peters, S G, 1998. Geological modelling of
alteration and gold deposition in the Betze deposit, Eureka County,
Nevada, US Geological Survey Open-File Report 98-338, pp
211-222.
159
Recent Investigations of Submarine Hydrothermal Activity in
Indonesia
T F McConachy1, H Permana2, R A Binns1,6, I Zulkarnain2, J M Parr3, C J Yeats3,
N D Hananto2, B Priadi4, S Burhanuddin5 and E P Utomo2
ABSTRACT
Through an ambitious but relatively low-cost program of marine
fieldwork, laboratory investigations and technology-transfer and training
activities, three Indonesia-Australia-led marine expeditions were
undertaken aboard KR Baruna Jaya VIII between 2001 and 2003. These
IASSHA (Indonesia-Australia Survey for Submarine Hydrothermal
Activity) cruises have sought to discover and investigate a variety of
potential ore forming hydrothermal systems in a range of tectonic
settings. All cruises have been successful, both scientifically and from the
point of view of training, creation of capabilities and capacity building for
such research in Indonesia. Over 60 scientists and technicians from
Indonesian and Australian institutions and agencies have been involved.
Three new sites and a possible fourth site of hydrothermal activity
were discovered in the Sangihe arc in northern Sulawesi. Surveys in the
Tomini basin tested a hypothesis based on previous data which suggested
this might be an isolated anoxic, brine-filled basin over 2000 m deep
where igneous heat sources existed to generate hydrothermal
mineralisation. The bay, however, was found to be open to adjacent deep
ocean circulation, and there was no evidence of massive base metal
sulfides. Subdued evidence of hydrothermal activity was found in the
Sunda Strait associated with the famous Krakatau volcano and the
extension of the seismically active Semangko Fault. The BANDAMIN
program (Indonesia-Germany) covered a portion of the Banda arc in
eastern Indonesia, and also found evidence of recent hydrothermal
activity and epithermal sulfide mineralisation.
Future discoveries of seafloor hydrothermal activity, sulfide mineral
deposits and associated biological systems are most likely to be within
archipelagic (ie territorial) waters, where the Government of Indonesia
has sovereignty. Hence, decisions about exploitation, evaluation, mining
and processing of mineral deposits and bioprospecting will come under
its direct jurisdiction, and not under the International Seabed Authority.
INTRODUCTION
Since the first discovery at 21N on the East Pacific Rise
(Edmond et al, 1979), the global inventory of sites where
massive sulfide mineral deposits are actively forming at
hydrothermal vents related to submarine volcanism has expanded
rapidly. Numerous examples along basaltic spreading ridges now
extend to every major ocean basin and divergent plate margin
(Stoffers et al, 2002; Edmonds et al, 2003). Equivalent activity is
now also known in a number of island arcs and backarcs at
convergent plate margins, particularly in the western and
southwestern Pacific region (eg Urabe et al, 1990; Herzig and
Hannington, 1995). The latter are important as they potentially
1. FAusIMM, CSIRO Exploration and Mining, PO Box 136, North
Ryde NSW 1670.
2. Earth Dynamics and Geological Hazard Division, Research Centre
for Geotechnology, Indonesian Institute of Sciences, Jl Sangkuriang,
Bandung 40135, Indonesia.
3. CSIRO Exploration and Mining, PO Box 136, North Ryde NSW
1670.
4. Department of Geology, Institute of Technology, Bandung 40135,
Indonesia.
5. Research Centre for Maritime Territory and Non Living Resources,
Ministry of Marine Affairs and Fisheries, Jl M T Haryono Kav 52-53
Jakarta Selatan 12770, Indonesia.
6. Department of Geology, Australian National University, Canberra.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
include close analogues for the geological settings of major base
metal and precious metal orebodies mined in ancient sequences,
especially where the associated volcanic rocks are felsic in
character as for instance at PACMANUS in the Manua back arc
of Papua New Guinea (Binns and Scott, 1993) and Sunrise in the
Izu-Bonin arc south of Japan (Iizasa et al, 1999).
Magmatic-hydrothermal activity in South-West Pacific and SE
Asia island arcs during the Tertiary has created a premier
copper-gold province with numerous, commercially attractive
base and precious metal deposits. This activity was assumed to
continue to the present day, especially in submarine sectors of
those arcs where modern technologies enable direct observation
of ore-forming processes in action, thereby providing new
approaches to understanding the interplay between igneous,
tectonic and hydrothermal phenomena and how they dictate the
location, size and tenor of mineral deposits; ultimately to assist
mineral exploration on land.
Until recently, however, the volcanic island arcs of Indonesia
have escaped such attention, but this has now been addressed by
two international collaborative programs (Figure 1). The
Indonesia-Australia Survey for Submarine Hydrothermal Activity
(IASSHA), involved three cruises with KR Baruna Jaya VIII,
focused particularly on the Sangihe arc extending from northern
Sulawesi towards the Philippine Islands; in Tomini Bay, Sulawesi,
including the vicinity of isolated Colo volcano; and in the Sunda
Strait including the neighbourhood of Krakatau volcano, (Binns,
Permana and Shipboard Participants, 2001a, b and c, 2004;
McConachy et al, 2001; McConachy, Permana and Shipboard
Participants, 2003; Permana et al, 2002; Yeats, 2002). The
BANDAMIN program (Indonesia-Germany) covered a portion of
the Banda arc in eastern Indonesia, finding evidence of recent
hydrothermal activity and mineralisation (Halbach et al, 2003).
These programs not only have provided invaluable training for
the many Indonesian scientists about how to undertake marine
expeditions to discover and locate seafloor hydrothermal activity
and associated deposits, but also have encouraged them and
government officers to consider management issues that must be
addressed by Government if such deposits are exploited in the
future.
The aim of this contribution is to give an overview of the main
results of these recent investigations of submarine hydrothermal
activity in Indonesian waters.
Methods
Methods used included single beam and high resolution
multibeam (<1000 m) echosounding to confirm the volcanic
morphology of targets, CTD-transmissometer casts and tow-yos
to detect particulate plumes associated with hydrothermal
activity, dredging, and bottom-tow video imaging. Single
channel seismic data were collected in Tomini Bay during
IASSHA 2001. During IASSHA-2003, rapid on-board analyses
of methane content (Schmitt et al, 1991), and pH measurements
of hydrocast samples were conducted as a further test for
hydrothermal plumes, and this was supplemented by post-cruise
analysis for diagnostic trace elements, particularly low level Mn
(McLaren et al, 1993). Plume particulates were captured by
filtering known amounts of seawater from the sample bottles
through pre-weighed polycarbonate filters under vacuum for post
cruise examination by scanning electron microscopy (SEM).
161
T F McCONACHY et al

C A M B O DI A
BURMA
Gulf
Andaman
of
10 N
Sea
0 VIETNAM Philippine
Thailand
T HAI LAN D Sea
PHILIPPINES
South China Pacific
Strait of
Malacca Sea Ocean
BA
BRUNEI
MALAY S IA Celebes KB
MALAY S IA Sea N
BW
Borneo
PAPUA
NEW
0
0
Kalimantan Tomini Bay GUINEA

Sumatra
Sulawesi

West
Ambon Papua
Java Sea
Banda
JAKARTA
SS/KR Madura Sea
Java KO

Denpasar EAST TIMOR

0 Arafura Sea
10 S Timor
Timor Sea
Indian Ocean
N
500 km

Submarine hydrothermal activity AUSTRALIA

0 0 0
100 E 120 E 140 E
TMCf001-04

FIG 1 - Map of Indonesia showing main tectonic features (Kertapati et al, 1992) and sites of submarine hydrothermal activity. Sites
in northern Sulawesi (KB, Kawio Barat; BW, Banua Wuhu; and N, Naung) and Sunda Strait (SS/KR, Sunda Strait/Krakatau) were
investigated by the IASSHA program (Indonesia-Australia), and the Komba (KO) site by BANDAMIN program (Indonesia-Germany).
The Banda Aceh (BA) site at the northern end of Sumatra was not investigated.

SANGIHE ARC Known mineralisation in the Sangihe arc includes a

IASSHA 2001 and 2003 cruises studied the arc and behind-arc
sectors of the Sangihe volcanic island chain extending
northwards from Quaternary volcanoes on the northeastern tip of
Sulawesi, near Manado, to the Kawio Islands near the border
with the Philippines. This chain lies above a west-dipping
Wadati-Benioff seismic zone whose subduction trench is
obscured by accreted sediments and mlange, forming the
Talaud-Mayu ridge. West of the main active chain and extending
northwards from Manado there is a subparallel ridge surmounted
by a number of high (>2000 m) seamounts of uncertain age.
From north to south, the chain includes the active subaerial
volcanoes Awu (1320 m elevation), Banua Wuhu (-5 m, near
Mahengetang Island), Karengetang (Api Siau, 1784 m) and
Ruang (752 m), as well as numerous smaller islands of volcanic
derivation (Figure 2).
The Sangihe arc is very volcanically active. For example, Awu,
the largest volcano of the Sangihe arc, has had five explosive
eruptions since 1711. An eruption in 1966 produced devasting
pyroclastic flows and lahars that killed more than 8000 people.
Banua Wuhu, near the island of Mahengetang forms a shoal less
than 5 m below sea level. Several ephemeral islands have been
constructed up to 90 m high, which have eventually disappeared
below sea level during the 19th and 20th centuries. Banua Wuhu
last erupted in 1968 when ocean water near the crater became hot
and turbid. Karangetang at the northern end of the island of Siau is
one of Indonesias most active volcanoes with more than 40 major
eruptions recorded since 1675 and many additional undocumented
small eruptions. Ruang forms a 4 5 km island and explosive
eruptions recorded since 1808 have been accompanied by lava
dome formation and pyroclastic flows.
porphyry-style Cu-Au occurrence of possible Pliocene age on
Taware Ridge and related colluvial gold; and nearby
high-sulfidation epithermal Au-Ag at the Binebase prospect on
the main island of Sangihe (Soeria-Atmadja et al, 1999). These
occurrences are considered by some workers (Meldrum et al,
1994) to lie in the Western Sulawesi porphyry belt that continues
north into Mindanao, Philippines, and possibly south to the Batu
Hijau porphyry Cu-Au deposit on Sumbawa.
In the Sangihe arc, the IASSHA-2001 and IASSHA-2003
expeditions examined numerous seamounts close to and between
the island chain, and to its west in what was initially, although
erroneously, considered by us a likely back arc zone (Figure 2).
Targets were selected on the basis of available bathymetry and
satellite altimetry (Smith and Sandwell, 1997); and were
numbered alphabetically in sequence, with new unofficial names
assigned to those that proved of special interest (Table 1).
Of the 19 sites examined, three showed positive evidence of
hydrothermal activity and were the subject of extensive
post-cruise research. One of these, Banua Wuhu, is an
occasionally emergent site of known volcanic and shallow
hydrothermal activity (Tomascik et al, 1997; Global Volcanism
Project, 2004). The other two, Naung and Kawio Barat, are
newly discovered deeper sites. A fourth site (Target R), also
showed positive signs of activity but was not investigated in
detail due to time constraints at sea. All four lie within the
volcanic island arc. Details of the remaining targets, where
evidence for hydrothermal activity was absent or subdued, are
provided in Table 1. They include all sites to the west of the
island chain, in a line (the Manado Line) of demonstrably or
presumably older edifices of uncertain affinity. Around 100 km

162 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RECENT INVESTIGATIONS OF SUBMARINE HYDROTHERMAL ACTIVITY IN INDONESIA

north of the Manado Line, we provide the first description of an The Manado Line
old, apparent hot-spot alkaline volcano (Maselihe; Binns et al, in
prep) rising above the Celebes microplate close to a reputed but On the rise north from Manado there is a chain of seven
uncertain site of submarine eruption in 1922. 2000 m-high conical edifices with crests descending northwards
from 310 m to 2000 m, referred to informally as the Manado
60 N Line (Figure 2). No transmissometer anomalies indicative of
Mindanao present-day hydrothermal activity were found, although weak
signals at 800 - 1200 m in the southern sector suggest that the
regions to either side, especially the southwest, deserve future
investigation. The upper flanks of three of these volcanoes
Hydrothermal Activity: (Target# D, E and G) were dredged. At the two shallower
Not tested edifices (crests at 400 m (E), 419 m (D)) we recovered coral and
None identified
shell fragments, and on one a fragment of manganese-encrusted
limestone. We suggest that these are relatively old, formerly

00
Uncertain/Subdued

-3 0
emergent, andesitic or basaltic volcanoes now significantly

-2 0 0 0
-3 0 0 0
Possible -4 0 foundered. From G (1794 m crest) we dredged large slabs of
00

Definite semiconsolidated foraminiferal sandstone thickly encrusted with

manganese oxide deposit. The manganese crust contains elevated
Active subaerial contents of Si, Al, Mg, Ca, Na and K indicating contamination
volcano Kawio Kawio Is
Barat R (~50 per cent) with volcanic detritus and possibly clays. The
Mineral occurrence
crust has relatively high Fe/Mn ratios, and contents of Ni
S
(0.27 per cent), Cu (0.04 per cent) and Co (0.03 per cent) are
-1000
-2 0 0 0
modest only. The nature and composition of the crust indicates
N
typical hydrogenic deposition on calcareous sediment settled
0 60 km
onto the presumed volcanic seamount during the <5 million
years since it ceased eruption. Therefore, Target G also appears
0
4 N to be a relatively ancient edifice, either formed by submarine
volcanism (from its shape, basaltic or andesitic) or with a long
Naung M history of submergence. The likely small areal extent of
Awu manganese-crust on the seamount crests and the low levels of
Maselihe Sangihe Is valuable elements (particularly Co) render the occurrence of no
economic significance. By comparison, Fe-Mn crusts in the
0 00

Palau-Federated States of Micronesia average 0.39 per cent Ni,

-4

0.10 per cent Cu and 0.45 per cent Co (Hein et al, 2000). Two
additional conical seamounts (Targets K and J), respectively
Banua Wuhu L 2300 and 2500 m high with crests at 1025 and 402 m, were
examined in deeper water between the Manado Line and the
active Sangihe arc. Neither showed hydrocast evidence of
K hydrothermal activity, and neither was dredged.
Celebes
Siau Is
Sea H
J Molucca Kawio Barat
Sea
This prominent feature, located 40 km west of the Kawio Islands
I (Figure 2) was targeted from satellite altimetry (Smith and
Manado Line

Tahulandang Is Sandwell, 1997).

G Ruang
-1 0 0 0

F Kawio Barat is a steep-sided, conical shaped feature, rising to

Biaro Is a depth of 1853 m from a 5 6 km base at 4300 m depth (Figure
0 E
2 N 3). The crest of the volcano is elongate NW-SE and is around
D 450 m higher than indicated by satellite altimetry. The 1900 m
C B A
bathymetric contour defining the crestal area is 1000 m-long and
up to 550 m across. Three crater-like features occur on the N, W
-
-3 4 0 0 Tongkoko and SW sides of the crest of the volcano, between the 2500 and
-2 0 0 0 0 0
00 -1 Manado Bitung 2250 m contours. These features are about 100 - 250 m in
00
0 diameter and 150 - 250 m deep. Irregular bathymetric contours
Lokon indicate significant NE-trending cross faulting. CTD results
established its hydrothermally active nature and encouraged
Mahawu
focused effort at this site.
Soputan
Hydrothermal plume
000
0
-1
00

North Arm
-2

Hydrothermal activity at Kawio Barat was first recognised by the

Sulawesi presence of a light transmissometer anomaly in the water column
ten to 90 m above bottom in hydrocast SH29, and confirmed
shortly afterwards by measuring anomalously high concentrations
of dissolved methane at the same position. A further two
1240 E 1260 E hydrocasts (SH30 and SH31), four dredges (SD13-16) and two
TMCf007-04 video tows (SV6 and SV7) were completed to further investigate
Kawio Barat. Results of CTD hydrocast SH31 are shown in Figure
FIG 2 - Map of Sangihe arc showing targets investigated by the 4. The profiles reveal a light transmission anomaly with near
IASSHA program. coincident high concentrations of dissolved methane and total

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 163

T F McCONACHY et al

TABLE 1
Sequential list of targets investigated by the IASSHA program (Indonesia-Australia) in the Sangihe arc region, northern Sulawesi,
Indonesia.

Target # Lat Lat Long Long True Appt Height Operations Morphology and Geology Hydrothermal Activity
crest crest
deg N min deg E min m m m
A 1 51.0 125 9.9 200 500+ M Low dome, flank of a volcano Not tested.
close to shore?
B 1 54.2 124 54.0 332* 630 1300+ M Off track, maybe not a volcano. Not tested.
*Sailed over flank; chart shows
332 m shoal.
C 1 51.4 124 41.9 310 1700 M,H Conical. Uncertain. Light transmission
anomaly of 0.8 per cent at 500 m.
D 1 54.7 124 41.8 419 1900 M,H,D Conical, fragments of Uncertain. Very weak light
limestone and coral, a small transmission anomaly at 888 m.
volcanic fragment. No
alteration. Possible founded
volcano.
E 2 2.9 124 42.3 400 1900 M,H,D Conical, fragments of limestone Uncertain. Very weak light
and coral, minor volcanic transmission anomaly 500 - 535 m.
fragment. No alteration.
Possible founded volcano.
F 2 10.4 124 42.5 1785 2200 N,H Conical, possible small Uncertain. Weak light transmission
breached caldera on N side of anomalies at 730 - 830 m (to SW),
crest. 790 - 1140 m (to NW), 820 - 1250
m (to E).
G 2 16.2 124 49.8 1794 1700 N, H, D Conical, slump or flow on W Possible. Weak light transmission
and S. Mn-oxide coated anomaly 1040 - 1100 m.
bioclastic limestones, no
alteration.
H 2 40.7 124 38.4 1999 2150 N,H Conical, with parasitic features None identifed.
on lower south and upper north
flanks, and a ledge on the upper
eastern flank.
I 2 26.3 124 44.3 2510 1100+ (N) Hummocky trace suggests flank Not tested.
crossed.
J 2 37.5 125 5.5 1025 2300 (M),N, Conical, multibeam (760 m) Not tested.
unreliable.
K 2 50.9 125 3.4 402 2500 M,H Conical. None identifed.
Banua 3 8.8 125 26.4 0 3000 M,D,H,V Banua Wuhu, parasite to Yes. Hydrothermally altered rocks,
Wuhu Mahengetang. Strongly altered weak light transmission anomalies.
volcanics.
L 3 7.3 125 29.0 447 120 M Small cone on a ridge. Not tested.
M 3 47.3 125 26.7 206 1000+ M Sharp cone with reef plateau at Not tested.
300 m.
Naung 3 46.5 125 22.4 514 800+ M,D Caldera, breached to NW. Fresh Yes. Weak light transmission
(N) to moderately altered and anomaly, anomalous CH4 and Mn.
fractured volcanic rocks.
Maselihe 3 39.6 124 34.2 2255 2300 N, D, H Possible irregular-shaped Co incident weakly elevated
(P) caldera, open to SE, lithified concentrations of CH 4 and Mn at
foram sandstone. Alkali basalt. 1800 m from possible distal
source?
Kawio 4 40.4 125 5.2 1832 3150 N, D, H, V Conical, fresh basalt, sulfide Yes. Strong plume signal.
Barat (Q) mineralisation, high temperature Photographs of vent field.
hydrothermal vent field, Recoverd sulfides. Evidence for
parasitic craters. boiling from CTD mounted on
camera sled.
R 4 40.1 125 14.9 918 1000 N, H Broad hummocky feature, Possible. Co incident elevated CH4
elongate NS. and Mn 980 - 1000 m, ~100 m
above bottom.
S 4 26.4 125 18.1 NA NA H Broad area tested by CTD for Uncertain. Coincident CH 4 and Mn
hydrothermal activity. at 500 m depth. Possible distal
plume from source unknown?

M = multibeam swath; N = narrow beam crossings; H = hydrocast; D = dredge; V = video tow

164 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RECENT INVESTIGATIONS OF SUBMARINE HYDROTHERMAL ACTIVITY IN INDONESIA

vent field dissolvable Mn (TDMn). Hydrothermal plumes are complex

SH29 CTD hydrocast -2 7 0 0 mixtures of dissolved chemical species, particulates and microbes.
SD15 dredge direction -2600
The pH of a deep seawater sample may be used as a proxy for the
SV06 camera tow
-2 5 0 0
amount of carbon dioxide dissolved in the seawater (CO2 + H2O
00 H2CO3). Some workers (eg Baker et al, 2002) suggest that
-2 4
around 75 per cent of a pH anomaly can be ascribed directly to
CO2, and which is considered to be magmatic in origin, based on
4041l N
the high concentrations implied for end member fluids (de Ronde
et al, 2001).
SD15
On board pH measurements of seawater samples collected in
the water column above Kawio Barat show complex patterns

16
SD
SH31
SV06 (both negative and positive shifts) in contrast to the near
SV07
SD13 SH30 coincident methane and TDMn anomalies, suggesting the
SH29 presence of possibly irregular concentrations of CO2 in the
-1900
14 plume.
SD
00 Based on the distribution of anomalous methane and TDMn,
4040l -20
the hydrothermal plume is between 150 and 260 m thick,
-2100
although the light transmission profiles suggest that the
0
-220 particulate portion of the plume is more confined. Measured
-2300 particulate concentrations in the plume vary from 30 to 65 g/L,
-2400 and particles mainly comprise ~10 m grains of hydrothermally
00
1 km -25 0 0
-26 derived Al-sulfate, pyrite, barite and calcite and volcanically
0
125 04
l
125 05 E
0 l 0
125 06
l
derived olivine, magnetite, pyroxene and possibly feldspar,
TMCf004-04
ilmenite/rutile and chromite. These combinations suggest either
FIG 3 - Bathymetric map of Kawio Barat submarine volcano recent volcanic eruption or entrainment of loose, volcanic ash
showing hydrocast, dredge and video tow sites and location of from the seafloor into a buoyant hydrothermal plume. Video tows
hydrothermal vent field; compiled from single beam echosounding that passed through the plume photographed distinct multiple
surveys conducted by the IASSHA program (September 2003). layers of milky water.

Methane (nl/l seawater) Mn (ug/l seawater)

0 20 40 60 80 0 1 2 3
1000 1000
CTD Pressure (dbar)
CTD Pressure (dabr)

1400 1400

1800 1800

pH Light Transmission (% )
7.64 7.84 8.04 88 88.5 89
1000 1000
CTD Pressure (dbar)
CTD Pressure (dbar)

1400 1400

1800 1800

FIG 4 - Pressure (~depth) versus methane, pH, light transmission and total dissolvable manganese for CTD-hydrocast SH31
over the summit of Kawio Barat showing a near-bottom hydrothermal plume. Methane and pH measurements by IASSHA 2003
shipboard scientist M C Pirlo. Post cruise total dissolvable Mn (TDMn) by analyst L Dotter (CSIRO).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 165

T F McCONACHY et al

Video tows/hydrothermal vent field

On the western side of the summit area at, (448.8N, 1254.9E)
1890 m below sea level, we photographed a hydrothermal vent
field (Figures 3 and 5). Over a distance of about 40 m,
spectacular shots reveal polycheat colonies, wholly and partially
covering rough, 1 - 4 m-relief, faulted bottom topography,
smoother pillow lavas, and in places, chimney-like substrates.
The position of the vent fauna is coincident with a temperature
anomaly of up to 0.25C above background, a sharp decrease in
light transmission and accompanying increase and decrease in
salinity that were measured by the CTD-transmissometer
mounted in the camera cage (Figure 6). Shimmering water is also
evident on the video at this location, as is possible grey smoker
activity. The presence of both an increase and decrease of
salinity with respect to seawater, suggests a high fluid
temperature of above approximately 370C resulting in boiling
in the subsurface. No ooze was photographed but loose volcanic
ash was easily disturbed by the camera hitting the bottom. From
the westerly movement of this fine ash, we infer the presence of
a westerly bottom current at this depth. This observation suggests
that the hydrothermal plume located by CTD-hydrocasts may
emanate from a different source(s), in the untested and
unexplored eastern summit area.

Rocks recovered by dredging

All four dredges at Kawio Barat recovered very fresh, or recently
erupted basalts from the crest, NE, W and S flanks of the volcano
(Figure 3). They are moderately to highly vesicular, and many
have glassy rinds. In places, a red-brown iron oxyhydroxide
deposit and/or white lacy biogenic material thinly coats the
rocks. Unfortunately, due to time constraints, the vent field
described above was not sampled. Several pieces of fresh basalt FIG 5 - Kawio Barat hydrothermal site-montage of bottom
recovered from the crest contain sulfides as do conspicuous photographs taken with sled-mounted deep tow video
cm-sized xenoliths of pale grey to cream altered rock. The camera-CTD system flown around 3 - 4 m above bottom. Note
xenoliths generally display partial to total alteration to X-ray rough faulted topography, spire-like features in foreground, and
amorphous clays, pseudomorphing former plagioclase and dense masses of polycheat colonies (tube worms). Total field of
pyroxene. Pyrite is minor to abundant, locally coarse grained, view of montage is approximately 30 m. (Montage by IASSHA
and, in places, replaces magnetite. One sample of basalt from 2003 Shipboard Science Journalist, G Burchfield).
dredge SD15 has a 5-cm ellipsoidal patch of alteration; its
apparently gradational boundary to basalt varies from white
4.0 80
laminated to brownish material. It is different from the other
xenoliths since it contains no pseudomorphs after phenocrysts,
consisting entirely of X-ray amorphous clay. There are zones, 3.9 34.9
possibly veinlets, of fresh glass with quench pyroxene microlites 79
cutting the xenolith, though not obviously connected to the 3.8 34.8
surrounding porphyritic host. These observations demonstrate
that some hydrothermal activity occurred prior to more recent Transmission (%)
Salinity (PSU)

lava flows. 3.7

78
T (oC)

Fresh basalts from SD14 and SD15 recovered from the S flank
and crest are similar, dominated by plagioclase phenocrysts with 3.6 34.6
varying proportions of clinopyroxene phenocrysts and magnetite.
The plagioclase and clinopyroxene are typically euhedral, 77
3.5 34.5
magnetite less so. Olivine is a minor phenocryst constituent and
tends to be more rounded, but curiously occurs around, rather 3.4 34.4
than within clinopyroxene. Most samples have glassy rinds, Temp
76
sometimes on both sides with the same phenocryst assemblage. Salinity
Melt inclusions are very abundant in the plagioclase and 3.3 34.3 Trans %
clinopyroxene phenocrysts. The high vesicularity of these basalts
at 2000 m depth indicates a volatile rich magma. 3.2 75
One dredge over the crest area (SD15) also recovered 1:01:00 1:02:00 1:03:00 1:04:00 1:05:00
relatively fresh basalt mineralised with marcasite and pyrite Ship's time (hr:min:sec) TMCf009-04
along fractures and filling vesicles. Because insufficient massive
sulfide was available for a bulk analysis, a preliminary estimate FIG 6 - Camera-sled CTD plot (temperature, light transmission,
was made of its probable metal content by comparing whole rock salinity and time) crossing 3 - 4 m above Kawio Barat hydrothermal
analyses of fresh and mineralised basalt samples. By using sulfur field. The temperature spike of 0.25C over the vent field coincides
as the guide and the highest values for base and precious metals, with a marked decrease in light transmission and decrease in
the estimated bulk composition of a massive sulfide is salinity. Nearby, small increases in salinity flank a temperature spike
approximately one per cent Cu, three per cent Zn, 0.7 per cent of 0.1C. These data suggest boiling of vent field fluid with
As, 75 ppm Sb and 10 ppm Au. subsurface or near surface temperatures of above ~370C.

166 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RECENT INVESTIGATIONS OF SUBMARINE HYDROTHERMAL ACTIVITY IN INDONESIA
In summary, a limited survey of Kawio Barat has
demonstrated unequivocal evidence of recent magmatic and
current hydrothermal activity. Kawio Barat is the deepest
submarine volcano yet found in Indonesian waters. The high
vesicularity of basalt suggests a volatile-rich magma, which is
indirectly supported by methane gas- and probable CO2 gas -rich
hydrothermal activity. Evidence of boiling, the presence of
possible chimney-like structures and disseminated and
anomalous base metal and gold values suggest that we are
witnessing the birth of a hydrothermal system, and further work
is clearly warranted.
Banua Wuhu
Banua Wuhu is as an active volcano whose summit is just
exposed at low tide. Gas bubbling, subsurface sonic activity, and
venting of hydrothermal fluids with temperatures around 50C
are known to occur on the summit at around 10 m depth, and
ferruginous oxide deposits several millimetres thick are common
(Tomascik et al, 1997). A multibeam bathymetric chart was
prepared to 1000 m and deeper narrow-beam echosounding was
conducted to show that Banua Wuhu is a parasitic feature on the
northwestern flank of adjacent Mahengetang Island (Figure 7),
also a volcanic construction, the combined edifice exceeding
3000 m in height. Hydrocast tow-yos following the 750 m
isobath on the southwestern and northwestern flanks of Banua
Wuhu failed to detect any significant hydrothermal plume, nor
did a deeper cast to the west. In a cast placed over the 1500 m
isobath on the southern side of Banua Wuhu (not shown in
Figure 7), however, there was a distinct but subdued light
transmission anomaly at 200 - 250 m depth. Because of strong
and variable tidal currents it is unclear whether the latter derives
from Banua Wuhu, nevertheless it was at in this depth range and
deeper that we discovered sites of hydrothermal alteration and
formation of bright white deposits.
FIG 7 - Bathymetry of Mahengetang Banua Wuhu, showing
CTD-hydrocasts, dredge sites and camera tows. Contours based
on multibeam survey by IASSHA program (June 2001).
A dredge (SD07, Figure 7) hauled a short distance from 650 m
depth up the western slope of Banua Wuhu recovered pervasively
altered porphyritic andesite containing disseminated pyrite and a
carbonate-chlorite-clay mineral assemblage, while a video tow
(SV05, Figure 7) descending this same slope from about 200 m
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
revealed small fields of bleached rock between volcanic outcrops
and talus before crashing spectacularly at 450 m depth onto a
mound of highly fragmented pale rocks with patches of white
deposit. With the camera sled we collected from this crash many
pieces of altered porphyritic volcanic rock equivalent to those
recovered by the nearby dredge, but here lacking pyrite. Similar
bleached rocks, many with Fe-stained exteriors, plus white
deposits were recorded by video (SV04, Figure 7) and
subsequently dredged (SD08, Figure 7) at 200 m depth to the
north of Banua Wuhu on a ridge descending from the
Mahengetang edifice. A shore party visited Mahengetang and
sampled outcrops of andesite lava and boulders of andesite from
a pyroclastic flow, as well as fragments of porphyritic pumice
said by residents to dominate the interior of the island.
Geochemical results from IASSHA 2001 (Binns, Permana and
Shipboard Participants, 2004) show the altered rocks are strongly
enriched in S, As, Cd, Sb and Te. Sulfur enrichment in particular
reflects pyrite replacing magnetite. This elevated trace element
assemblage is a magmatic signature, and is associated with
anomalous Au values (up to 20 ppb). Calcium is depleted during
alteration, with the break down of igneous plagioclase and
hornblende. In andesite exhibiting low intensity alteration, Fe,
Na, Mg and Mn are weakly depleted, but Si is weakly enriched.
In summary, Banua Wuhu is clearly a satellite system of
Mahengetang Island which displays alteration in the form of
addition of magmatophile elements.
Naung
This feature lies off the north coast of the active Awu volcano
(Figure 2) and was targeted from a 53 m shoal sounding on a
published nautical chart (this proved to be a drafting error, and
has been corrected in the new edition). Our surveys in IASSHA
2001 and 2003 demonstrated the presence of a subdued summit
crater or caldera about 600 m across, breached to the northwest
and surrounded on the other sides by cuspate knolls, the southern
most of which rises to 520 m depth at 346.45N 12522.37E
(Figure 8). Naung is circular in overall outline and not connected
by any ridge to Awu or other submarine structures. A flat-topped
submarine feature to the east (Target M-not investigated) has a
206 m deep summit surrounded by a platform at 300 m depth
that is presumed to be a foundered fringing reef.
SH34 CTD hydrocast
SD17 dredge direction 500 m
0 l
3 47.0
-900
-800
0
-70 -7
SD09
00
-6 SD
0 0 17 SH34
3046.5l N -514
-6 0 0
-7 0 0
00
-900
-8
-5 1 4
crest, depth (m)
125022.0l 125022.5l E 125023.0l
TMCf005-04
FIG 8 - Bathymetry of Naung showing operation sites for both
IASSHA 2001 and 2003 expeditions. Contours are based on
multibeam surveys.
The presence of hydrothermal activity is indicated by
anomalous methane of up to 70 nl/L, and a marked negative shift
in pH indicating CO2 enrichment. Unlike Kawio Barat, however,
there is no definite light transmission anomaly at Naung and only
very weakly elevated TDMn (Figure 9). These data suggest that
hydrothermal activity is manifested mainly by either gaseous,
167
T F McCONACHY et al

Methane (nl/l seawater) Mn (ug/l seawater)

0 20 40 60 80 0 1 2 3
0 0

CTD Pressure (dbar)

CTD Pressure (dbar)

200 200

400 400

600 600

800 800

Light Transmission (% )
pH
7.64 7.84 8.04 86.5 87.5 88.5 89.5
0 0

CTD Pressure (dbar)

CTD Pressure (dbar)

200 200

400 400

600 600

800 800

FIG 9 - Pressure (~depth) versus methane, pH, light transmission and total dissolvable manganese in CTD-hydrocast, SH34,
over Naung, north of Awu Island. Methane and pH measurements by IASSHA 2003 shipboard scientist M C Pirlo. Post cruise
total dissolvable Mn (TDMn) by analyst L Dotter (CSIRO).

particulate-poor emanations from the seafloor, or possibly from

degassing lava flows, or both. Dredge samples recovered include
fresh pyroxene-bearing basaltic andesites, but with heavily
Fe-stained and filled fractures in places. One hand-sized piece of
dolomitic limestone (90 per cent CaO, ten per cent MgO) was
also recovered from dredge SD17, near the southern crest (Figure
8) which suggests that Naung was shallower at some stage in the
past. Such a history of emergent volcanoes which then founder is
common along the Sangihe arc. No camera tow was conducted to
confirm this suspected hydrothermal activity.

Maselihe
We sought the site of a reputed submarine eruption in 1922 (van
Bemmelen, 1949), variously named Gunungapi (Sangihe map) or
Sangir (Volcanoes of Indonesia map) but listed as uncertain by
the Global Volcanism Program. Guided by a prominent shoal
defined by satellite altimetry (Smith and Sandwell, 1997) close
to the reported site, the IASSHA 2003 expedition found a large
conical feature elongate northwest, 18 45 km in diameter at its
base, rising 2300 m above a flat sea floor 4500 - 4600 m deep
(Figure 10) (McConachy, Permana and Shipboard Participants,
2003). The crestal area, about 3 km across, has four peaks of
which the shallowest rises to 2255 m below sea level at 339.6N
12434.2E. A possible breached caldera occurs between the
three southern peaks. The seamount, informally named Maselihe
(a local Sangihe legend where the palace of the Kendahe
Kingdom submerged), lies 55 km southeast of the position listed
by Global Volcanism Program, and is 750 m higher than FIG 10 - Bathymetric map of Maselihe, a new alkali hot spot in the
indicated by satellite altimetry. Celebes Sea.

168 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RECENT INVESTIGATIONS OF SUBMARINE HYDROTHERMAL ACTIVITY IN INDONESIA

CTD-transmissometer hydrocast and methane analysis over the

Sumatra
highest of the southern three peaks of the crestal area showed no
evidence of hydrothermal activity, though slightly elevated Kalianda
methane (29.2 nl/L) measured at 1800 m depth, is coincident with Sebuku Is
elevated total dissolvable Mn (0.16 g/L), and is possibly a signal
of a distal plume from activity in the Sangihe arc to the east. Two 6000l
SUH-07 SUH-02 Sebesi Is
of three dredge attempts successfully sampled the feature, both Krakatau

Se
Caldera SUH-01
recovering Early to Middle Pliocene semiconsolidated

ma
SUH-04 Rakata Is
foraminiferal sandstone. Three angular lumps to 50 cm across of

ngko
SUH-03
pale grey trachyte breccia with thin surficial Mn crusts, two of

Faul
which were embedded on the upper surface of semiconsolidated 0 l Java
6 30 S
sand blocks, were also recovered. Fresh volcanic glass, plagioclase

t
Panaitan Is
and pyroxene, which shows no signs of having undergone SUD-10
0 40 km
diagenesis, are the products of recent eruptions most likley from
subaerial volcano(es) along the Sangihe arc. 0
104 30 E l 0
105 00 l 0
105 30 l
106000l
Petrographically, the trachyte breccia is composed mainly of TMCf011-04

tiny pilotaxitic anorthoclase laths with rare phenocrysts and

glomerocrysts of simply-twinned anorthoclase with interstitial
kaolinite-smectite clay and scattered titaniferous magnetite
granules (Binns et al, in prep). Geochemically, the rocks have a
highly evolved alkaline affinity; Zr and Nb are highly enriched,
whereas Ba, Sr and Rb are highly depleted. REE patterns relative
to chondrite show LREE enrichment, flat HREE, and a
pronounced negative Eu anomaly (Binns et al, in prep).
Our preliminary interpretation suggests that Maselihe is
broadly the same age as oceanic basement of the Celebes Sea
Basin, rather than the product of a recent eruption, and therefore
represents an old hot spot in the Celebes Plate.
SUNDA STRAIT
IASSHA 2002 investigated the famous Krakatau volcano and the
seismically active Semangko fault which runs southeast across
the Sunda Strait from Semangko Bay in southern Sumatra and
passes to the west of Panaitan Island (Figure 11). The seawater
column within the Krakatau caldera is 5C to 8C warmer,
slightly less saline and more turbid than water elsewhere at these
depths in the Sunda Strait, indicating that Krakatau is
hydrothermally active (Figure 12). This conclusion is also
supported by post cruise analyses of seawater samples which
recorded up to 6.3 g/L Mn, levels which are comparable with
Temperature (C) Light Transmission (%)
10 12 14 16 18 20 22 24 26 28 30 70 74
0
SUH01
SUH02
SUH03
50
SUH04
FIG 11 - Map of Sunda Strait and operations referred to in the text
during IASSHA 2002 (September 2002).
Mn concentrations in hydrothermal plumes at the PACMANUS
hydrothermal site in Papua New Guinea (McConachy, 2002).
High resolution multibeam echosounder mapping of the caldera
floor revealed two 50 - 60 m high conical features which are
probable volcanic edifices. Neither edifice was successfully
sampled. However, dredge and grab samples of the floor and
walls of the caldera returned olive green ooze and fresh pumice
fragments with no indication of hydrothermal alteration, though
the smell of H2S gas was evident upon recovery on deck
(Permana et al, 2002; Yeats, 2002).
A 70 m-thick light transmission anomaly of four per cent
interpreted as being a possible hydrothermal plume was recorded
in SUH-07 around 80 m above bottom (1235 m), adjacent to the
Semangko Fault near the mouth of Semangko Bay (Figure 11).
The plume was confined to the north and east by sharp rises in
bathymetry and closed off by two additional hydrocasts deployed
~1.8 km SW and SE which recorded no anomaly and a weak
0.4 per cent transmission anomaly, respectively. Disappointingly,
follow up dredges failed to recover any material from this area.
Hydrothermally altered rocks, however, were recovered 70 km
SE, along the southern extension of the Semangko Fault to the
Salinity (psu)
78 82 86 90 32.5 33 33.5 34 34.5 35

100
Depth
(m)

150

200

250 TMCf010-04

FIG 12 - Comparison of temperature, light transmission and salinity profiles from CTD-hydrocasts in the Krakatau caldera (SUH-01, SUH-02
and SUH-03) and background in the Sunda Strait (SUH-04) during IASSHA 2002 (September 2002).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 169

T F McCONACHY et al
west of Panaitan Island (Figure 11). These include rounded to
subrounded oxidised boulders of poorly sorted greywacke and
conglomerate, weakly altered to moderately chlorite-altered
feldsparquartz-phyric volcanic and one piece of pale, silicified
feldspar-phyric lava with 0.5 to one per cent fine subhedral
pyrite, partially or entirely oxidised to Fe-oxide. Given their
oxidised nature, it is more likely that these are pieces of older
basement, rather than representing products of current
hydrothermal activity on the fault scarp. However, as no further
operations (and in particular, no CTD-hydrocasts) were carried
out in the area, this inference cannot be made with confidence.
Further work is required along the Semangko Fault to clarify
these observations.
TOMINI BAY
Tomini Bay, a virtually unexplored Neogene sedimentary basin,
is surrounded by the northern, western and southeastern arms of
Sulawesi (Figure 13). Our objective here was to test whether
modern sediment-hosted hydrothermal activity occurred on the
sea floor. Equivalent settings are metallogenically important in
ancient sequences, but no modern analogue is as yet known.
0
2 N
Celebes
Manado
Sea Bitung
North Arm Sulawesi
Gorontalo
Tomini Bay
Una Una Is
Gorontalo Basin
0
0 (Colo Volcano)
SC11
(-2022m) Togian Is
Palu
100 km
0
0
122 E 124 E
0
2S
, TMCf008-04
Ship track, seismic line; IASSHA 2001
SC11
(-2022m) Gravity core; depth of seafloor below sea level
Fault
FIG 13 - Map of Tomini Bay, northern Sulawesi, showing ship
tracks of IASSHA 2001, and features discussed in the text.
Two facts encouraged this expectation. First, the presence of
Colo volcano on Una Una Island at the eastern side of Tomini
Bay suggested potential presence of magmatic heat sources to
drive hydrothermal circulation systems and, secondly, previous
bathymetric compilations interpreted Tomini Bay as an enclosed
basin some 2000 m deep, separated from Gorontalo Bay to the
east by a sill more than 300 m high. With inflow of organic
debris from rivers draining the encircling heavily vegetated, high
rainfall landmass of Sulawesi, the possibility seemed strong for
anoxic and/or anomalously saline conditions to prevail on the
basin floor. Should appropriate faulting and deformation exist,
the essential elements would be present for formation of
sediment-hosted base metal sulfide deposits. Like the subaerial
volcanoes in the Sangihe arc, the Colo volcano is also active. It is
508 m high with a 2 km wide caldera containing a crater lake.
Eruptions in 1898 and 1900 produced a lava dome, building up
the main part of the island. Phreatic eruptions occurred in 1928
and 1948, and an earthquake was recorded in 1961 (Katili, 1975).
A catastrophic eruption in 1983 produced ash material and a
nue ardante that flowed to the SE and SW. Volcanic ash covered
Tomini Basin, western Sulawesi and extended as far as
Kalimantan (Katili and Sudradjat, 1990).
170 Adelaide, SA, 19 - 22 September 2004
We demonstrated, however, that the present-day sill between
Tomini and Gorontalo Bays is less than 170 m high, and it
involves only very gentle slopes. Like Gorontalo Bay, Tomini
Basin is filled by around 1 km of presumed late Tertiary
sediments comprising two distinct sequences separated by slight
unconformity. The lower sequence, which is continuous between
the two basins, is mildly folded and faulted, especially where it
thins at basin margins and where it is arched to form the
pediment of Una Una Island. The upper sequence of turbidites is
undisturbed and exists as two separated depocentres in the
respective basins.
Sedimentary cores from the centre of Tomini Basin (eg SC11,
Figure 13), did not penetrate hydrothermal deposits, and no
geochemical evidence was found in any sample for hydrothermal
activity. The upper, Colo-derived, 20 cm of sediment in SC11
from the centre of Tomini Bay shows a standard progression
from brownish, oxidised surface material to greenish-grey,
diagenetically reduced deeper sediment, showing that Tomini
Bay is not today an anoxic basin. A lower mud unit of SC11
appears predominantly of mafic to ultramafic provenance
(possibly correlating with the sediment and sedimentary rocks
recovered from the ridge extending NW from Una Una). The
lack of organic matter in this deeper sediment suggests that
oxidised conditions persisted earlier as well, but only a very
limited portion of the >500 m (800 ms two way travel time)
sedimentary fill has been examined. The seawater column at the
centre of the basin shows conventional oceanic profiles of
salinity, temperature, and density. No hint of a light transmission
anomaly indicative of hydrothermal activity was present at any
depth. These observations negate the likely presence in central
Tomini Bay of large-scale, near-surface modern analogues of
hydrothermal massive sulfide environments involving
hydrothermal venting of basinal or magma-derived fluids into
Molluca reduced sediments.
Sea
Enigmatic Plume Una Una
Whereas most hydrocasts taken in a circuit of Una Una island
126 E
0 (Figure 13) showed no evidence of hydrothermal activity, we
found and mapped a prominent cloudy plume centred above a
ridge NW of Una Una, at 200 - 250 m depth. On one cast, this
plume coincided with an alkaline pH anomaly. Two shallower
plumes are also evident in the light transmission data, centred at
around 150 m and 180 m water depth, and suggest that the source
of the plume is in this area. Particulates filtered from samples
collected in the plume were bright white in colour, unlike the
brown ferruginous particulates normally found above
high-temperature hydrothermal sites in mid-ocean and back-arc
volcanic regions. Under SEM, the particulates comprise
dominantly irregular-shaped calcite; and irregular to
ellipsoidal-shaped barite, celestite, anhydrite/gypsum, chromite
with Ni (and one olivine/serpentine), hornblende, Na-K feldspar,
chlorite, magnetite, possible quartz, and a single pyrite grain.
There is no sign of biotite, which is present in bottom sediments,
to suggest any land-derived material from Colo volcano. Two
video tows along the fault zones either side of the ridge failed to
record any classic hydrothermal vents or deposits representing
possible sources for the plume. Patches of white material,
however, were photographed among the jointed mudstone
outcrops and bouldery debris. We conclude that the plume is
unlikely to be associated with submarine turbidity currents, nor
with supersaturated-carbonate-rich tropical waters as none of the
particulates have the characteristic needles of aragonite attributed
to whitings. A more likely interpretation is that we may have
witnessed signs of hydrothermal activity driven by heat from
serpentinisation reactions, similar to those described in 1.5 Ma
rocks off axis near the Mid Atlantic Ridge (Kelley et al, 2001).
Ultramafics are exposed on land nearby, and underlie Una Una
Island as evidenced by the presence of ultramafic xenoliths in
volcanic products from the eruption of the Colo volcano on Una
Una (Binns, Permana and Shipboard Participants, 2004)
PACRIM 2004
 RECENT INVESTIGATIONS OF SUBMARINE HYDROTHERMAL ACTIVITY IN INDONESIA
In summary, although we did not achieve our objective of
finding sediment-hosted hydrothermal activity in the Tomini-Una
Una region, by no means was the basin fully tested. In particular,
it is possible that potential remains in the deeper parts of the
basinal sediment pile.
KOMBA
Two BANADAMIN expeditions in 2001 and 2003
(Indonesia-Germany) investigated the seafloor ~36 km north of
the island of Lomblen, near the subaerial Komba volcano in the
eastern Flores Sea (Figure 1) (Halbach et al, 2003; Halbach,
2003). Bathymetric mapping revealed a dissected seamount with
two summits separated by pull-apart plane at 800S 12348E.
CTD surveys measured temperature anomalies in the water
column above the NE summit indicative of hydrothermal activity.
Porphyritic potassium-rich, fresh to highly altered basaltic
trachyandesite and trachydacite were recovered from this feature.
The highly altered rocks contain sulfides, mainly pyrite with
lesser amounts of pyrrhotite and rare chalcopyrite and oxidised
gossanous equivalents. Halbach et al (2003) report that altered
rocks impregnated with several percent sulfide contain up to
0.15 g/t Au and 4 g/t Ag, though later fire assays yielded higher
concentrations which are being confirmed (Halbach, 2003). The
BANDAMIN team interpret the Komba area to represent
submarine epithermal-style mineralisation, with the presence of
albite, K-feldspar, and calcite (and the lack of kaolinite, alunite
and pyrophyllite) indicative of a more distal low-sulfidation
system. A second type of alteration assemblage, however,
comprising illite, amorphous silica and quartz, adularia and
sulfide is indicative of higher alteration temperatures that have
broken down plagioclase.
DISCUSSION
From the emerging knowledge of Indonesias submarine geology
and metallogenesis gained by these programs, future discoveries
of seafloor sulfide mineral deposits are most likely to be wholly
within archipelagic (ie territorial) waters, where the Government
of Indonesia has complete sovereignty. Decisions about
exploitation, evaluation, mining and processing of mineral
deposits and bioprospecting will come under the direct
jurisdiction of the Government of Indonesia, and not under the
International Seabed Authority. As such, it is the responsibility
of the Government of Indonesia to formulate policy which
encourages and sustains scientific research and attracts
commercial interest into seafloor hydrothermal systems for the
benefit of all Indonesians.
CONCLUSIONS
Up until 2001, there was little known about submarine
hydrothermal activity in Indonesian waters. Only two shallow
sites were known, and there had been no systematic surveys
undertaken in deeper water. We have presented here the
preliminary results of the work carried out over the past three
years and there are a number of papers being prepared that more
fully describe the results of the work. These recent investigations
into submarine hydrothermal activity by joint
Indonesia-Australia and Indonesia-Germany cruises have
underscored the potential of Indonesian waters with the
discovery of five new sites. All three active sites in the Sangihe
arc warrant further investigation, especially camera tows,
precision sampling and further bathymetric mapping. Many more
sites will inevitably be found by further surveys. The Krakatau
caldera is hydrothermally active, as well as potentially two other
sites associated with the Semankgo fault in the Sunda Strait
between Sumatra and Java.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
None of the occurrences, however, are in felsic-hosted settings,
an initial scientific hypothesis of the IASSHA program.
Although no massive sulfide deposits of the kind found in the
Papua New Guinea, Fiji and elsewhere have been located or
recovered to date in Indonesian waters, there are emerging and
tantalising indications that these exist, especially in the vicinity
of sites in the Sangihe arc, and potentially the Banda arc.
The concept of sediment-hosted massive sulfides in the Tomini
Bay was tested and found not to apply. Significant contributions,
however, have been made to the understanding of the marine
geology and tectonic evolution of the area.
Seafloor investigations of the type undertaken are exploratory in
nature, and researchers must be prepared to expect the unexpected.
The discovery of the Maselihe seamount in the Celebes Sea and its
interpretation as an evolved trachyte hotspot; and an enigmatic
plume near Una Una in Tomini Bay are just two examples.
ACKNOWLEDGEMENTS
The IASSHA program was funded by the Indonesian Institute of
Sciences, Ministry of Marine Affairs and Fisheries, CSIRO
Exploration and Mining, AusAID under the Australian
Government Sector Linkages Program, and, in the early stages
by CSIROs industry sponsors Phelps Dodge, Rio Tinto,
Newmont and Pasminco. Post cruise work conducted by the
Indonesian Institute of Sciences has benefited from a RUTI
(Riset Unggulan Terpadu Internasional, International
Competitive Research Program) grant. We are grateful to Captain
Danil and crew of KR Baruna Jaya VIII for their patient and
excellent work at sea, and to all shipboard scientists, too many to
individually mention here, who contributed to the success of the
IASSHA program. Indroyono Soesilo and Hery Harjono are
thanked for continued and unfailing support. Rob Hough kindly
provided SEM data.
REFERENCES
Baker, E T, Massoth, G J, de Ronde, C E J, Lupton, J E and McInnes, B I
A, 2002. Observations and sampling of an ongoing subsurface
eruption of Kavachi volcano, Solomon Islands, May 2000, Geology,
30(11):975-978.
Binns, R A and Scott, S D, 1993. Actively forming polymetallic sulfide
deposits associated with felsic volcanic rocks in the eastern Manus
back-arc basin, Papua New Guinea, Economic Geology,
88:2226-2236.
Binns, R A, Permana, H, McConachy, T F, Zulkarnain, I, Priadi, B and
Hananto, N, (in prep). Maselihe Seamount A hotspot volcano in the
Celebes Sea Basin, West Pacific Ocean.
Binns, R A, Permana, H and Shipboard Participants, 2002a. IASSHA
2001 cruise report volume 1: Introduction (Eds: J M Parr, N D
Hananto and I Zulkarnain), CSIRO Exploration and Mining Report
(unpublished), 982F, 49 p.
Binns, R A, Permana, H and Shipboard Participants, 2002b. IASSHA
2001 cruise report volume 2: leg A Tomini Gorontalo Basins (Eds: J
M Parr, N D Hananto and I Zulkarnain), CSIRO Exploration and
Mining Report (unpublished), 983F, 69 p.
Binns, R A, Permana, H and Shipboard Participants, 2002c. IASSHA
2001 cruise report volume 3: leg B Sangihe Arc (Eds: J M Parr, N D
Hananto and I Zulkarnain), CSIRO Exploration and Mining Report
(unpublished), 984F, 56 p.
Binns, R A, Permana, H and Shipboard Participants, 2004. IASSHA 2001
cruise report volume 4: post cruise research (Ed: J M Parr), CSIRO
Exploration and Mining Report (unpublished), 128 p.
de Ronde, C E J, Baker, E T, Massoth, G J, Lupton, J E, Wright, I C,
Feely, R A and Greene, R R, 2001. Intra-oceanic subduction-related
hydrothermal venting, Kermadec volcanic arc, New Zealand, Earth
and Planetary Science Letters, 193:359-369.
Edmonds, H N, Michael, P J, Baker, E T, Connelly, D P, Snow, J E,
Langmuir, C H, Dick, H J B, Muehe, R, German, C R and Graham,
D W, 2003. Discovery of abundant hydrothermal venting on the
ultraslow-spreading Gakkel Ridge in the Arctic Ocean, Nature,
421(6920):252-256.
171
T F McCONACHY et al
Edmond, J M, Measures, C, Grant, B, Sclater, F R, Collier, R, Hudson, A,
Gordon, L I and Corless, J B, 1979. On the formation of metal-rich
deposits at ridge crests, Earth and Planetary Science Letters,
46:19-30.
Global Volcanism Project, 2004.
http:www.volcano.si.edu/gvp/world/volcano.cfm, March.
Halbach, P, 2003. Alteration and mineralisation patterns in marine
volcanoes of the eastern Flores Sea, Indonesia. Talk (no abstract)
presented 13-16 December 2003, International Expo and Seminar on
Marine, Fisheries and Seafood Industries and Technology, Jakarta,
Indonesia. (Department of Marine Affairs and Fisheries: Republic of
Indonesia).
Halbach, P, Sarmili, L, Kag, M, Melchert, B, Post, J, Rahders, E and
Haryadi Y, 2003. The break-up of a submarine volcano in the
Flores-Wetar Basin (Indonesia): implications for hydrothermal
mineral deposition, InterRidge News, 12(1):18-22.
Hein, J R, Koschinsky, A, Bau, M, Manheim, F T, Kang, J-K and
Roberts, L, 2000. Cobalt-rich ferromanganese crusts in the Pacific,
Handbook of Marine Mineral Deposits (Ed: D S Cronan) pp 239-279
(CRC Press: Boca Raton, Florida).
Herzig, P M and Hannington, M D, 1995. Polymetallic massive sulfides
at the modern seafloor: a review, Ore Geology Reviews,
10(2):95-115.
Iizasa, K, Fiske, R S, Ishizuka, O, Yuasa, M, Hashimoto, J, Ishibashi, J,
Naka, J, Horii, Y, Fujiwara, Y, Imai, A and Koyama, S, 1999. A
Kuroko-type polymetallic sulphide deposit in a submarine silic
caldera, Science, 283(5404):975-977.
Katili, J A, 1975. Volcanism and plate tectonics in the Indonesian island
arcs, Tectonophysics, 26:165-188.
Katili, J A and Sudradjat, A, 1990. The devastating 1983 eruption of Colo
Volcano, Una-Una Island, central Sulawesi, Indonesia, in
Proceedings Sixth Regional Conference on Geology, Mineral and
Hydrocarbon Sources of Southeast Asia, Journal of Southeast Asia
Earth Sciences (GEOSEA VI), (Eds: A J Barber and R S K Phoa),
4(1):82 (Pergamon).
Kelley, D S, Karson, J A, Blackman, D K, Fruh-Green, G L, Butterfield,
D A, Lilley, M D, Olson, E J, Schrenk, M O, Roek, K K, Lebonk, G
T, Rivizzigno, P and AT3-60 Shipboard Party, 2001. An off-axis
hydrothermal vent field near the Mid-Atlantic Ridge at 308N,
Nature, 412:145-149.
Kertapati, E K, Soehaimi, A, Djuhanda, A and Effendi, I, 1992.
Seismotectonic map of Indonesia, Geological Research and
Development Centre, Jl. Diponegoro 57, Bandung 40122, Indonesia
(revised with additional data by E K Kertapati and I Effendi, 1998).
McConachy, T F, 2002. Preliminary Cruise Report DaeYang02, RV
Onnuri, Lihir-Bismarck Sea, Papua New Guinea August
25-September 7, 2002: CSIRO Exploration and Mining Report
(unpublished) 1009C, 55 p.
McConachy, T F, Permana, H and Shipboard Participants, 2003. Cruise
report, IASSHA 2003 aboard KR Baruna Jaya VIII September 2003.
CSIRO Exploration and Mining Report (unpublished) 1153R, 135 p.
172 Adelaide, SA, 19 - 22 September 2004
McConachy, T F, Binns, R A, Permana, H and Shipboard Participants,
2001. First survey for submarine hydrothermal vents in NE Sulawesi,
Indonesia, Eos Trans American Geophysical Union, Fall Meeting
Supplement, 82(47), Abstract T52A-0923.
McLaren, J W, Lam, J W H and Berman, S S, 1993. On-line method for
the analysis of sea-water for trace elements by inductively coupled
mass spectrometry, Journal of Analytical Atomic Spectrometry,
8:279.
Meldrum, S J, Aquino, R S, Gonzales, R I, Burke, R J, Suyadi, A, Irianto,
B and Clarke, D S, 1994. The Batu Hijau porphyry copper-gold
deposit, Sumbawa Island, Indonesia, in Mineral Deposits of
Indonesia: Discoveries of the past 25 years, Special Issue Journal of
Geochemical Exploration (Eds: T M van Leeuwen, J W Hedenquist,
L P James and J A S Dow), 50:203-220.
Permana, H, Zulkarnain, I, Hananto, N G and Shipboard Participants,
2002. Scientific final report of RUTI 2002, stage 1. Submarine
metallogenic resources and its management: case of Sunda Strait and
northern Sulawesi waters. Earth Dynamics and Geological Hazard
Division, Research Centre for Geotechnology, Indonesian Institute of
Sciences, Jl. Sangkuriang, Bandung, 40135, Indonesia, Internal
Report (unpublished), 22 p.
Schmitt, M, Faber, E, Botz, R and Stoffers, P, 1991. Extraction of
methane from seawater using ultrasonic vacuum degassing,
Analytical Chemistry, 63:529-532.
Soeria-Atmadja, R, Priadi, B, van Leeuwen, T M and Kavalieris, I, 1999.
Tectonic setting of porphyry Cu-Au, Mo, and related mineralisation
associated with contrasted Neogene magamatism in the Western
Sulawesi Arc, The Island Arc, 8(1):47-55 (Blackwell).
Smith, W H F and Sandwell, D T, 1997. Global seafloor topography from
satellite altimetry and ship depth soundings, Science, 277:1957-1962.
Stoffers, P, Worthington, T, Hekinian, R, Petersen, S, Hannington, M,
Tuerkay, M and F/S Sonne 157 Shipboard Scientific Party (DEU),
2002. Silicic volcanism and hydrothermal activity documented at
Pacific-Antarctic Ridge, Eos, Transactions, American Geophysical
Union, 83(28):301, 304.
Tomascik, T, Mah, A J, Nontji, A and Moosa, M K, 1997. The Ecology of
the Indonesian Seas, Part I, Chapters 1-12, pp 44-47, 529-537,
542-565. (No 5) (Periplus Editions, Hong Kong: UNSW Press Ltd).
Urabe, T, Auzende, J M, Deplus, C, Eissen, J P, Grimaud, D, Huchon,
Philippe, Ishibashi, J, Joshima, M, Lagabrielle, Yves, Mevel, C,
Naka, J, Ruellan, E, Tanaka, T and Tanahashi, M, 1990.
Hydrothermal deposits at the central spreading axis of the northern
Fiji Basin (SW Pacific), Preliminary report of the Nautile dives of
STARMER Cruise, Mining Geology, 40(2)(220):117-124.
van Bemmelen, R W, 1949. The Geology of Indonesia, 732 p (Government
Printing Office, The Hague).
Yeats, C J, 2002. Submarine hydrothermal and volcanic activity in the
Sunda Strait, Indonesia: summary cruise report KR Baruna Jaya VIII
IASSHA-2002, CSIRO Exploration and Mining Report
(unpublished) 1003C, 59 p.
PACRIM 2004
A Geological Review of the Tampakan Copper-Gold Deposit,
Southern Mindanao, Philippines
C Middleton1, A Buenavista2, B Rohrlach3, J Gonzalez4, L Subang5 and G Moreno6
ABSTRACT
Located on the island of Mindanao in the Southern Philippines, the
Tampakan copper-gold deposit represents a major high-sulfidation
epithermal deposit superimposed on an underlying porphyry copper
system. With a recent JORC compliant indicated and inferred resource
estimate of 1104 million tonnes at 0.65 per cent copper and 0.26 g/t gold
at a 0.3 per cent copper cut-off grade, Tampakan represents the largest
undeveloped copper-gold deposit in Southeast Asia. Since its acquisition
by Indophil Resources affiliate Sagittarius Mines Inc in August 2002, the
deposit is the subject of an ongoing preliminary feasibility study into
project viability that is initially focused on a higher-grade copper zone in
the north-central part of the deposit.
The deposit is hosted by a sequence of subaerial andesitic flows
(Tampakan Andesite Sequence) related to a highly eroded andesitic
stratavolcanic complex of Pliocene age. Detailed mapping and
geochronological dating has identified four volcanic cycles or centres that
have been sequentially developed and eroded. This stratovolcanic
complex is interpreted to lie on the northern extension of the Sangihe
volcanic arc that extends from Northern Sulawesi in Indonesia. The
Tampakan district is located within a complex tectonic setting dominated
by WNW trending wrench faults that form part of the trans-Mindanao
Cotabato Fault Zone. The Tampakan deposit is located near the
intersection of one of several WNW trending splay structures with a
prominent NNE trending dilational structure, one of many that transect
the host volcanic complex. High-sulfidation style copper-gold
mineralisation at Tampakan is broadly associated with and central to a
flat-lying tabular body of pervasive silica and silica-clay alteration
developed within a district- scale lithocap of advanced-argillic and
argillic hydrothermal alteration that extends over an area of 90 km2 within
the volcanic complex.
Deposit geology has been developed from a systematic program of
diamond drilling that currently comprises 102 completed holes (38 400 m)
with most holes ending in lower grade (0.2 - 0.3 per cent Cu)
porphyry-copper style mineralisation at downhole depths of 350 - 500 m.
Superimposed high-sulfidation mineralisation and associated alteration
passes transitionally to porphyry copper style mineralisation and
alteration at depth. The interface between these styles of mineralisation
can be quite irregular as a result of structural control on hydrothermal
fluids. At an approximate 0.3 per cent copper contour, the deposit
comprises a flat-lying to gently dipping, tabular body of mineralisation
extending over an area of 2 km by 1.6 km and ranging in thickness from
200 - 400 m. Although there is a close spatial correlation between these
mineralisation types, the earlier porphyry event has been dated at 4.25 Ma
and the later high-sulfidation event at 3.2 - 3.4 Ma.
1. FAusIMM, Former Director of Exploration and General Manager
Technical Services, C/O- Indophil Resources NL, Level 3, 99 Queen
Street, Melbourne Vic 3000. E-mail: middletonc@netspace.net.au
2. Project Manager, Tampakan Mineral Resources Corporation, JPM
Building, Bula-Lagao Road, General Santos City 9500 Mindanao,
Philippines.
3. Geological Consultant, 75 Wybalena Grove, Canberra ACT 2614.
Swiss Inn 1, Guevara Subdivision, Legaspi, Philippines.
4. Chief Geologist, Tampakan Mineral Resources Corporation, JPM
Building, Bula-Lagao Road, General Santos City 9500 Mindanao,
Philippines.
5. Senior Project Geologist, Tampakan Mineral Resources Corporation,
JPM Building, Bula-Lagao Road, General Santos City 9500
Mindanao, Philippines.
6. Senior Project Geologist, Tampakan Mineral Resources Corporation,
JPM Building, Bula-Lagao Road, General Santos City 9500
Mindanao, Philippines.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
High-sulfidation epithermal mineralisation and alteration is strongly
influenced by the interplay of stratigraphic and structural controls, in
particular NNE-trending dilatant faults. This style of mineralisation is
typically associated with silica and silica-clay (pyrophyllite-dickite-alunite)
hydrothermal alteration assemblages that have obliterated or overprinted
original textures and mineralogy. Zones of silica alteration often display
multi-phase brecciation and vuggy porosity resulting from intense acid
leaching. Sulfide mineralisation typically comprises the copper-rich species
digenite, bornite, and enargite, and to a lesser extent chalcocite and
covellite, and usually occurs as fractures, veins and vug-fills or as
disseminations at depth. Pyrite is usually present throughout and enargite
distribution appears to be preferentially associated with late-stage
structures and areas of strong silicification.
Porphyry-copper style mineralisation at depth comprises disseminated
chalcopyrite, bornite and pyrite that is associated with variably developed
quartz-stockwork veining. This mineralisation style is usually associated
with sericite-clay-chlorite anhydrite alteration of host andesite and
high-level hornblende diorite stocks and may represent the intermediate
shell of a porphyry system. The metal grade in structural zones within the
porphyry system is locally upgraded by high-sulfidation overprinting
events, resulting in a change of sulfide mineralogy and progressive acid
alteration of the host sequence.
INTRODUCTION
The Tampakan copper-gold deposit is a major high-sulfidation
epithermal deposit superimposed on a pre-existing low- to
moderate-grade porphyry copper system. It is located on the
island of Mindanao, approximately 45 km NNW of General
Santos City, a major growth centre in the southern Philippines
(Figure 1). With a recently published JORC compliant indicated
and inferred resource estimate of 1104 million tonnes at 0.65 per
cent copper and 0.26 g/t gold at a 0.3 per cent copper cut-off
grade (Hellman and Schofield, 2004) the Tampakan deposit
represents the largest undeveloped copper-gold deposit in
Southeast Asia.
The deposit was discovered by WMC Resources Exploration
Division in its first exploration drillhole which was collared
within the Tampakan project area in December 1992. This
discovery resulted from a systematic exploration program
comprising reconnaissance geological mapping and geochemical
drainage surveys followed by grid-based detailed geological
mapping, multi-element soil sampling, and ground magnetic,
electromagnetic and induced polarisation geophysical surveys
over a selected target area (Balkau et al, 1998). By the
completion of an advanced exploration drilling program in
September 1997, 85 diamond drillholes had been completed for a
total of 31 474 m. This, and subsequent infill drilling by an
Indophil Resources affiliate Sagittarius Mines Inc (Sagittarius),
has provided the basis for interpreting geological controls to
copper-gold mineralisation, associated alteration assemblages,
variations in sulfide mineralogy and metal zonation patterns.
Title to the Tampakan deposit and adjacent environment is
currently held under a Financial and Technical Assistance
Agreement (the Columbio FTAA) with the Philippines
Government, which allows a company with up to 100 per cent
foreign control to explore for, and develop major mineral
projects. The Columbio FTAA was granted to WMC Philippines
(WMCP) in March 1995. After statutory reductions it currently
covers an area of 304.9 sq km. A contiguous FTAA application
(Hillcrest) covering an area of 454.8 sq km forms part of the
overall Tampakan project area (Figure 1). In August 2002, the
173
C MIDDLETON et al

N. Cotabato
Sultan Kudar LUZON
at

DIGOS
Davao
Matanao Gulf

SAMAR
Columbio
Lake
Buluan
SDwag Pula River
LEYTE

10N
Bong Dalol Bukay-El
Sultan Kudarat MINDANAO
S. Cotabato
Salnaong Kiblawan
Koronadal
Bolol Bato 120E Map Area
Silway N.
Tampakan Snip Silway Tampakan
Mt. Cu-Au Deposit
Magolo 0 10
Other prospects
Kilometre
2293m.
Columbio FTAA
Mt. Matutum Hillcrest FTAA
Application

Davao D
Tampakan Drill Grid el Sur
South Co
(Figure 4) tabato
Provincial Boundary
Major Road
City or Town

GEN.
SANTOS
Location Diagram:
Tampakan
Sarangani Copper-Gold
Bay
Deposit

FIG 1 - Location diagram Tampakan copper-gold project.

FTAA title was transferred to Sagittarius following the purchase
of the project from WMC Resources through a right of first
refusal.
The Tampakan project area is located at the southern end of
the Central Mindanao Cordillera and is bound to the west by the
Cotabato Basin and to the east by the Sarangani Basin, both
relatively flat-lying areas. Most of the project area is
characterised by elevated topography. The Tampakan deposit is
located at approximate geographic coordinates latitude
062830N and longitude 1250300E, and lies along a
NNE-trending ridge at 1000 - 1300 m above sea-level. The Mt
Matutum Volcano (2293 m) which last erupted over 2000 years
ago lies 12 km south of the Tamapakan deposit.
Since acquiring an interest in the project in August 2002,
Indophil and Sagittarius have undertaken a comprehensive
review of historical project data and commenced a preliminary
feasibility study (PFS) focused on a higher-grade copper zone
within the Tampakan deposit. The PFS comprises a significant
program of infill and delineation diamond drilling for a revised
resource estimation and associated metallurgical and mining
studies and is due for completion by early- to mid-2005.
Concurrently, the company has initiated certain environmental
studies and is working closely with local communities, local
government units and government agencies to facilitate the PFS
and future project development. Community assistance and
development programs are either in place, or are being planned
in collaboration with local community and indigenous groups.
This paper is a general descriptive paper of the Tampakan
copper-gold deposit that relies heavily on the work of previous
WMC and current Indophil/Sagittarius geoscientists. A detailed
PhD research project focused on the tectonic evolution,
petrochemistry, geochronology and palaeohydrology of the
deposit and surrounding district has been undertaken and
documented by Rohrlach, 2002.
REGIONAL GEOLOGICAL AND TECTONIC
SETTING
Rohrlach (2002) provides a comprehensive review and
documentation of the tectonic evolution of Southeast Asia and in
particular the regional geological and tectonic setting of southern
Mindanao. His review incorporates the work of Hall (1996) and
other researchers. Southern Mindanao is characterised by a
highly complex syn- to post-collisional tectonic setting related to
both terrane accretion and arc-arc collision. The Tampakan
district lies at the southern end of the Central Mindanao
Cordillera and at the northern end of the Miocene- to
Pliocene-age Sangihe volcanic arc that extends southwards as an
active island arc through Balut Is and the Sangihe Islands in the
Molucca Sea, and into the northern arm of Sulawesi in Indonesia
(Rohrlach, 2002; Balkau et al, 1998). A major collisional event
between the Philippine Sea Plate and the Eurasian Plate in
southern Mindanao, that developed in the late Miocene and
extended through the Pliocene, has produced a prominent

174 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A GEOLOGICAL REVIEW OF THE TAMPAKAN COPPER-GOLD DEPOSIT

INTERPRETED GEOLOGY OF MINDANAO

(Modified after Domasig et al 1998)

RECENT ALLUVIAL COVER

PH
ILI

PHIL
ACTIVE VOLCANIC CENTRES

PP
INE
TERTIARY INTRUSIVE

IPPI
CENTRES

FA

NE T
UL
TERTIARY VOLCANICS

T
+ VOLCANO-SEDIMENTS

REN
PRE-TERTIARY BASEMENT

CH
SUBDUCTION ZONE
MAJOR FAULT
Cagayan De Oro City
H
C

DAVAO
CENT
CO
EN

RAL M
TA

-AGUS
BA
TO

INDA
R

AN TR
FA

NAO
T

UL

OUGH
CORD
T
U

L Z

ILLER
U

O
S

NE

A
C O TA

CO
TA
Davao

BA
TO
City
DA
Mt. Apo

BA
Zamboanga City GU
06.40N

SIN
MA
RA
NG Digos
BA

Mt.
Matutum
TO

ns ni
a
ni ga
ul
Pe ran
0 80
T

Gen. Santos City

Sa
R

E Tampakan District
KILOMETRE N
C
125.07E
H

FIG 2 - Interpreted geology of Mindanao. (modified after Domasig et al 1998).

WNW-trending sinistral strike-slip fault zone approximately
100 km wide referred to as the Cotabato Fault Zone (Figure 2).
This WNW-trending collision suture separates the southern and
northern Mindanao structural domains and represents a boundary
of crustal accretion.
The Tampakan district is located within the Cotabato Fault
Zone and lies on a thrust-fault-related ramp anticline that folds
an early Miocene volcanic basement. These folding and thrusting
events resulted from compression during both movement along
the Cotabato Fault Zone and an arc-arc collisional event between
the northerly trending Halmahera and Sangihe volcanic arcs
during the Pliocene Period. Both of these collisional events are
regarded as important in the evolution of metallogenetically
fertile magmas in the Tampakan district (Rohrlach, 2002).
Crustal compression and associated deformation in southern
Mindanao extended over a seven million year interval from 7 Ma
to the present, peaking in intensity between 4 Ma and 2 Ma
(Rohrlach, 2002). This peak in crustal compression occurred
almost synchronously with porphyry copper and high-sulfidation
epithermal Cu-Au mineralising events within the Tampakan
mineral district (Rohrlach, 2002).
The Tampakan district is flanked by two extensional basins,
the Cotabato Basin to the west and the Sarangani Basin to the
east, that predominantly contain sedimentary rocks derived from
the rapid erosion of adjacent uplifting volcanic sequences.
Formed initially as a fore-arc basin between the early Miocene
(16 - 18 Ma) volcanic arc sequence of the Daguma Range and
the Sangihe trench, the Cotabato Basin has more recently
developed as a back-arc basin to the Daguma Range resulting
from north-easterly subduction at the Cotabato Trench
(Rohrlach, 2002).
The Sarangani Basin is regarded as an intra-arc basin between
the southern Central Mindanao Cordillera (Tampakan district)
and the Sarangani Peninsula. The recently active Mt Apo and
Mt Matutum volcanoes lie along the south Central Mindanao
Cordillera.
TAMPAKAN DISTRICT GEOLOGY
Simplified geology of the Tampakan district as shown in Figure 3
is based on work by Buenavista et al (1995) as modified by
Rohrlach (2002) to incorporate structural detail. The Tampakan
district is bound to the east by the Central Cordillera Thrust, a
generally NNW-trending arcuate structure that separates the
district from the Sarangani Basin further eastwards. The
Cotabato Basin delimits the district to the west. The core of the
district is occupied by an early to mid Miocene-age basement
sequence of basalt and basaltic andesite lavas (Sulop Formation)
and associated volcaniclastic sediments that are folded along a
NW-trending ramp anticline. An intercalated limestone unit
within the basaltic andesite sequence has yielded a seawater
Sr87/Sr86 correlation age of 17.44 (+0.21/-0.8) Ma (Rohrlach,
2002). The exposed part of this sequence is delimited to the north
by a major WNW-trending left lateral structure referred to as the
Alip River Fault Zone and to the south it is unconformably
overlain by the Tampakan Andesite Sequence (TAS), the
principal stratigraphic unit of economic significance within the
district.
The Tampakan Andesite Sequence is interpreted to be a deeply
dissected and eroded Pliocene-age stratovolcanic complex
(Tampakan Stratovolcanic Complex) of andesitic composition
with an inferred original diameter varying from approximately

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 175

C MIDDLETON et al

V V V
Structural V V

Lineament Interpretation
SARANGANI
from

CE
Digital Elevation Model v BASIN

NT
RA
v

L
AL
IP

CO
RD
RIVE
6.40.00N R

ILL
FAU

ER
QU
LT

AN
Lake

EZ
ZON
E

ON
Buluan

TH
RU
ST
RA
NG
COTABATO
E
BASIN

+ +
+ +
+ +
6.30.00N +
RO
XA
S

Tampakan ** Mt. Magolo

Township
Lampitak * ** Akbang
Plugs Dome
Maar

BUA
YAN
Lambayong
Plugs FA
ULT

ZO
NE

0 RA
NG
0 2 4 6 8 10 E
Mt. Matutum
Quaternary
Kilometres Stratovolcano
124.50.00E 125.00.00E 125.10.00E

GEOLOGY OF THE TAMPAKAN DISTRICT

Southern Central Mindanao Cordillera

* * * Quaternary tuffaceous lacustrine sediments +++ Miocene diorite Faults:

Quaternary-andesite stratocone/pyroclastic flows Basalt & basaltic andesite (Early Miocene) Thrust faults

V V V Basalts flows and andesite Volcaniclastic sediments (Early Miocene) Strike-slip

Limestone Tampakan copper-gold deposit Normal

Massive hydrothermal silica
Volcanic sediments Medium
and silica clay bodies
Andesite-Dacite flow domes, plugs (Pleistocene)
Antiform, Synform axis
Pliocene volcaniclastic sediments
Advanced Argillic and Argillic alteration Annular / elliptical topographic anomaly
of Tampakan Andesite Sequence - (Pliocene)
Tampakan Andesite Sequence - (Pliocene)

FIG 3 - Geology of the Tampakan district Southern Central Mindanao Cordillera (modified after Beunavista et al, 1995 and Rohrlach, 2002).

18 - 24 km. Rohrlach (2002) recognised four andesitic volcanic
cycles (Cycles 1 to 4) within the complex using
geochronological studies (K-Ar, Ar-Ar and U-Pb) and suggested
a series of sequentially rebuilt and eroded volcanic centres over a
4.5 million year interval extending from ~7 Ma to ~2.5 Ma. The
distribution of the fourth cycle is currently manifest in digital
terrain models as a dissected, middle Pliocene-age, central vent
stratocone-type landform 8.4 km wide that has been erosionally
truncated to an average elevation of 1200 - 1300 metres above
current sea level (Rohrlach, 2002). The Tampakan copper-gold
deposit lies just below the western edge of the truncated upper
surface. The andesite sequence generally comprises a series of
grey to dark green porphyitic pyroxene-hornblende-andesite flow
units. Rorhlach (2002) suggests that the lack of vesicular
textures, magmatic breccias or interbedded pyroclastic horizons
may indicate that the main stratocone sequence erupted by quiet
effusive lava outpouring rather than explosive volcanism. Table
1, sourced form Rohrlach (2002), summarises geochronological
ages from the Tampakan Andesite Sequence and related
mineralising events.

176 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A GEOLOGICAL REVIEW OF THE TAMPAKAN COPPER-GOLD DEPOSIT

TABLE 1
Summary of geochronological dates for the Tampakan Andesite Sequence.

Cycle/Event Composition Age (Ma) Comments

High-sulfidation epithermal Described in text 3.23 - 3.34 Hosted by cycles 1, 3 and 4a. Broadly tabular stratabound mineralisation
mineralisation along the unconformity between cycle 3 and 4a andesites
Porphyry copper Described in text 4.24 - 4.26 Hosted by cycle 1 and 3
mineralisation event
Post- mineralisation dykes Andesite-diorite 3.04 0.04 Near coincident with the high sulfidation event but displaying
3.14 0.06 post-mineralisation characteristics
Post-porphyry intrusive Diorite 3.38 0.04 Near coincident with the high sulfidation event.
3.67 0.04
Pre-porphyry intrusive Diorite 5.37 0.32 Intrusive stock
Cycle 4b Andesite to dacite flow 1.47 Logdeck Andesite and younger intrusive dacite plugs, eg Nun Guon
dome complexes Dacite
Cycle 4a Andesites and associated ~3.0 - 3.9 Unconformably overlies other cycles
diorite stocks and dykes
Cycle 3 Andesite 4.39 - 4.76 Generally preserved in proximity to the Tampakan copper-gold deposit
Cycle 2 Basaltic andesite to 6-7 Mostly exposed around the western flank of the Tampakan Stratovolcanic
andesite Complex
Cycle 1 Andesite 8.34 Identified at the base of the deepest drillhole
Basement Sequence Basalt and basaltic 17 - 18 Folded basement sequence unconformably overlain by the Tampakan
andesites and associated Lower to Andesite Sequence
mid-Miocene

Information in the above tabulation is derived from Rohrlach (2002).

The Tampakan Andesite Sequence is intruded by a series of DEPOSIT GEOLOGY

hornblende diorite porphyry stocks and dykes of inferred coeval
age. In the near deposit environment, the preserved thickness of
andesitic flow units is 650 m between the current topographic
surface and diorite stocks intersected in drillholes at depth. The
Tampakan Stratovolcanic Complex is transected in the south by
the Buayan Fault Zone, a WNW strike-slip structure that
parallels the Alip River Fault Zone further north. Both of these
faults are strike-slip splays of the Cotabato Fault Zone. The
digital terrain model of the volcanic complex also indicates a
number of smaller circular and arcuate structures that may
represent paleo-volcanic features. Other WNW structures are
evident in the data as well as prominent orthogonal
NNE-trending structures that are important in the distribution of
copper-gold mineralisation, in particular the high-sulfidation
mineralisation event. The Tampakan copper-gold deposit lies at
the intersection of such WNW- and NNE-trending structures.
A significant part (>90 sq km) of the Tampakan Andesite
Sequence has been affected by widespread advanced-argillic and
argillic hydrothermal alteration that is transitional outwards to a
regional propylitic alteration facies. The Tampakan deposit is
spatially associated with one of several major bodies of enhanced
silicification located within the district-scale advanced-argillic
and argillic alteration envelope and lies approximately central to
it (Rohrlach, 2002).
Pliocene-age volcaniclastic sediments that flank the basement
sequence and the Tampakan Stratovolcanic Complex are regarded
as the erosional products of an actively uplifting adjacent
environment. Post-mineralisation andesite and dacite plugs and
extrusive flow dome complexes respectively intrude and
unconformably overlie the Tampakan Andesite Sequence. These
Pleistocene-age units also intrude sediments within the Cotabato
Basin to the west of Mt Matutum (Lampitak and Lambayong
Plugs). A thin (1 - 6 m) blanket of unconsolidated, andesitic,
pyroclastic crystal lapilli tuff sourced from Mt Matutum, a
Quaternary-age stratovolcano, occurs throughout much of the
district with the most recent eruption dated at 2000 years BP.
Preamble
The geology of the Tampakan deposit is constructed from a
drilling grid that covers an area of 2 km 1.6 km (3.2 sq km) in
the N-S and E-W directions respectively (Figure 4), and limited
surface exposures including siliceous and geochemically
anomalous advanced-argillic lithocap outcrops exposed along the
western margin of the deposit. Mineralogical logging of drill
core undertaken by WMC and Rohrlach (2002) was augmented
by PIMA spectroscopy and facilitated widespread
characterisation of alteration. These observations have been
further augmented by detailed logging of drillcore from
Indophils current infill and delineation drilling program.
At the time of writing a total of 102 diamond drillholes for
38 414 m of coring had been completed within the grid with
many holes terminating in low-grade (0.3 per cent Cu) porphyry
copper style mineralisation at downhole depths of 350 - 500 m.
The drilling density over the grid (deposit) varies from a broad
320 m 320 m collar spacing in the south and east, to a closer
160 m 160 m collar spacing elsewhere. Since acquiring an
interest in the project, Indophil has focused on a systematic
diamond drill out on approximate 80 m 80 m centres over the
northern part of an interpreted higher-grade copper zone (100 m
per cent copper and 1 per cent copper). The majority of holes
within the deposit have been drilled on E-W traverses and are
generally inclined at -60 to the east or west to facilitate the
intersection of inferred NNE-trending subvertical structures.
Four WMCP drillholes were inclined to the north to ascertain
any E-W structural control to mineralisation.
Figure 5, modified after Rohrlach et al (1999) and Rohrlach
(2002) is a schematic representation of stratigraphy, alteration
zonation and mineralogy within the Tampakan Deposit, whilst
Table 2 provides a summary of key features of host rocks,
breccias, alteration and associated mineralisation. Some aspects of
these are augmented or highlighted in the descriptive text below.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 177

C MIDDLETON et al

505,040mE
Exploration Base Camp TAMPAKAN
14 13 COPPER-GOLD PROJECT
16 15 77 8
716190N 71 PLAN PROJECTION
72
716110N 111 39
51
60
OF MINERALISATION
61 90
716030N 107 106

715950N
Da
84
92 85 21
AND DRILLHOLE
nla 5 105 37 23
22
715870N
gF
au 94 108 104
100 27 54
20
50
LOCATION
lt
102 103 46
715790N 109 64 40 43
82 81 91 62 42
110 Plan projection of high-grade zone
715710N 97 98 99 101
89 >100 metre % Copper and >1% Copper
88 7 96 32
715630N 18 17
4 19 6 55
Plan projection of mineralised zone
U at 0.5% Cu cut-off
D 44
49

ult
75 74 41 20 Drillhole with direction

Fa
47 24 26
56 715,310mN
31

wit
3 25 52 Drillhole with intersection > 100 metre

La
70 52 67 78 79 % Copper and >1% Cu
72 69 53 68
Fault
29
r

2
Corrido

1 ASPAC Camp 48
45 28 36
38 30 Ro
ad

Bangay
o Fault
U
Fault

D 0 500
34 33
87 35
to

86 Metre
Pula Ba

83

FIG 4 - Plan projection of mineralisation and drillhole location.

GENERALISED STRATIGRAPHY, ALTERATION ZONATION AND MINERALOGY

TAMPAKAN Cu-Au DEPOSIT
SCHEMATIC DISTRIBUTION of
ALTERATION and SULFIDE
MINERALOGY
1300 (Recent) MT. MATUTUM TUFF PHYLLOSILICATES
OXIDES SULFIDES
(Pleistocene) LOGDECK DACITE
Smectite

EARLY DIATREME
ARGILLIC

Phenocryst textures
Chlorite

1200
Illite

preserved
Phenocrystic vuggy silica
Chalcocite
Pyrophyllite

~ 4.25 - 3.93 Ma Pervasively silicified

Diaspore
Silicic

Alunite
Dickite

1100
Silica

hydrothermal breccias
Metres (ASL)

Enargite (Luzonite)
ADVANCED

unconformity
ARGILLIC

Clotted vuggy silica

Digenite

Clotted silica-clay
Silica
Clay

Conduit

Molybdenite

Granoblastic textures common

1000 TAMPAKAN Veined enargite-molybdenite
Silica

ult
Clay

ANDESITE zones (Feeder structures)

Pula Bato Fa

(Early to Middle
Pliocene)
Phenocrystic textures
variably obscured
Specular Hematite

900
Top of Gypsum Veining
INTERMEDIATE ARGILLIC

Phenocryst textures
Covellite

well preserved
Chalcophyrite
(SCC)

800
Gypsum after Anhydrite

(Middle to Late DIORITE/

ANDESITE
Sericite

Pliocene) Darker porphyry

DYKES
quartz +/- cp, bn, py
stockwork veins
Magnetite
Anhydrite
Illite

TAMPAKAN
Biotite

(Late
700 Gypsum veins
ANDESITE
Chlorite

Miocene)
Bornite

Relict Potassic Zones

Pyrite

(Late EARLY DIORITE

Miocene) STOCKS

FIG 5 - Generalised stratigraphy, alteration zonation and sulfide mineralogy (modified after Rohrlach, 2002 and Rohrlach et al, 1999).

178 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A GEOLOGICAL REVIEW OF THE TAMPAKAN COPPER-GOLD DEPOSIT
TABLE 2
Summary of key features of host rocks, breccias, alteration and mineralisation.
Category Distribution and description
Host rocks Tampakan Andesite Sequence (TAS) intruded by diorites of
varying composition and textures. Most mineralisation is hosted
by highly altered lithologies where original textures have been
largely obliterated.
Andesites In the immediate deposit environment the host andesite
sequence (TAS) attains a thickness 650 m and dips at
~10 - 20 to the WSW. The sequence comprises massive,
undifferentiated grey to green porphyritic flow units usually of
hornblende-pyroxene andesite composition.
Dioritic intrusives Hornblende diorite intrusives occurring as stocks and dyke-like
bodies occur usually at deeper levels. Original textures are
usually masked by intense sericite-chlorite alteration. Where
preserved, original textures vary from mg equigranular to
crowded porphyry textures.
Breccias A wide variety of breccias host mineralisation (see below).
Alteration zonation Three major alteration types recognised in a vertically zoned
pattern: argillic, advanced-argillic and intermediate argillic
(SCC). The argillic and advanced-argillic alteration are related
to a high-sulfidation mineralisation event.
Propylitic Generally restricted to higher stratigraphic levels at the
periphery of the deposit. Usually comprises chlorite-epidote
clay alteration.
Argillic Uppermost alteration within the Tampakan environment
predominantly hosted by cycle 4a of the TAS. Largely
comprises an assemblage of illite-smectite-chlorite that has
largely replaced original mineralogy but commonly preserved
original textures of host andesites.
Advanced argillic Underlies the argillic alteration zone and is widespread
extending into the surrounding district. Ranges in thickness
from 50 - 200 m but can attain 400 m in structural corridors.
Divided into three generally vertically distributed subtypes:
silicic, silica-clay and clay-silica.
Silicic Silica = 90 per cent. Three predominant textural variants:
massive silica, vuggy silica and siliceous breccias (see below).
Massive silica usually present in the uppermost levels of the
advanced argillic alteration occurring as either recrystallised
residual silica or as deposited silica from hydrothermal fluids.
In vuggy silica, voids are created by intense acid leaching.
Leaching of feldspar phenocrysts produces phenocrystic vuggy
silica. Coarse vuggy silica comprises 0.5 - 2 cm irregular
ovoid shaped vugs in a silicified matrix.
Silica-clay Silica 50 - 90 per cent. This advanced alteration subtype is
generally developed underneath and in sharp contact with the
silicic horizon, or enclosing lenses of massive silica alteration.
Three textural variants: massive silica clay an irregular
assemblage of aluminosilicates and sulfates (dickite,
pyrophyllite, diaspore, alunite precursor illite and sericite) and
silica with texturally destructive alteration, porphyritic
silica-clay and clotted silica-clay characterised by coarse
(0.2 - 2 cm) clay clots in a groundmass of massive silica or clay.
Clay-silica Clay 50 - 90 per cent. The silica-clay alteration subtype
(above)s transitional downwards to a clay rich assemblage
dominated by pyrophyllite, dickite and lesser diaspore alunite.
Significant precursor illite and sericite are present as relicts of
an overprinted intermediate argillic alteration (Rohrlach, 1999).
Clay-silica alteration is texturally destructive with pervasive fine
grained clays occurring interstitial to granoblastic silica and
coarse silica aggregates (Rohrlach, 2002).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Mineralisation
Altered andesites host the bulk of copper-gold mineralisation
within the deposit, in particular high-sulfidation
mineralisation (see below).
These intrusives are commonly altered to intermediate argillic
alteration assemblages with associated mineralisation as
described below.
Diverse mineralisation (see below)
The majority of copper-gold mineralisation within the
deposit, in particular higher grades, is associated with the
advanced-argillic alteration zone. Porphyry related
mineralisation (sensu-stricto) is generally 0.1 - 0.6 per cent
Cu.
At times geochemically anomalous in copper and gold.
Only minor copper-gold mineralisation is associated with this
alteration zone, usually in association with silicification or
faults.
Host to the majority of copper metal within the Tampakan
deposit, especially higher copper grades
Usually occurs as vug-filling sulfides and brittle
hydro-fracture, stringer and breccia veins. Classified into two
types based primarily on the enargite content. Type 1 is
considered to be paragentically early and consists
predominantly of py-en with minor luz, dig, cc, bn and cpy. It
is intimately associated with massive to vuggy silicification
and pyrophyllite and diaspore clays. Euhedral py and en are
the common vug filling sulfides, usually associated with
drussy quartz and high temperature clay fill minerals. Type 2
represents a low arsenic (= 150 ppm As) mineralisation
typically consisting of dig-bn and lesser en and mo. These
minerals are usually associated with kaolinite, dickite, rutile
and alunite with accessory smectite, illite and sericite
(Reynolds, 1997). Due to their localisation in the silicic
alteration cap, both of these types of mineralisation tend to be
grossly stratabound. Below the silica cap Type 1
mineralisation can taper into subvertical silica+quartz+
sulfide feeder breccias.
Mineralisation associated with the silica-clay and clay-silica
subtypes of advanced argillic alteration beneath the silicic cap
are subdivided into two types (Type 3 and 4) based largely on
the relative abundance of certain copper sulfide species. Type
3 mineralisation is characterised by dig-bn rich zones and
lesser cpy. Type 4 mineralisation is dominated by cpy-bn and
weak to trace amounts of dig. Continued below.
Type 4 mineralisation becomes more prevalent at depth as cpy
increases and bn decreases. The strongest mineralised
sections of these types are marked by a granoblastic texture
with strong sericite flooding. In both types the copper sulfides
usually occur as fine disseminated aggregates in the
groundmass and as fracture and cavity fills.
179
C MIDDLETON et al

TABLE 2 (continued)
Summary of key features of host rocks, breccias, alteration and mineralisation.

Category Distribution and description Mineralisation

Intermediate argillic Gradational with the overlying HS clay silica subzone this zone Type 5 mineralisation. This alteration zone is regarded as the
(SCC) comprises sericite-clay-chlorite (SCC) alteration with abundant outer shell of a porphyry system with copper grades typically
silica and pyrite. Illite is the predominant clay mineral. Textural in the 1.2 - 0.6 per cent range. Low to moderate density
preservation is common and this alteration is spatially related to pinkish quartz veins with py-cpy commonly observed.
extensive porphyry-stage quartz stockwork veining (Rohrlach, Significant specular hematite as primary qtz-cpy-hem veins
2002). A gypsum-anhydrite subzone comprising and cpy-hempy disseminations. Disseminated py-cpybn
gypsum-anhydrite veins has been defined at depth as an gradually diminishes with depth down to the
overprinting alteration. gypsum-anhydrite zone.
Potassic A relict potassic alteration assemblage comprising Volumetrically insignificant.
biotite-magnetite- anhydrite has been intersected near the
margin of host diorite intrusives in a few deep drillholes.
Breccias Several breccia types are recognised in the deposit environment,
including diatreme breccias, hydrothermal breccias, pebble
breccias (dykes), shear/fault breccias and sulfide/vein breccias.
Diatreme breccias WMC identified three small-scale diatreme breccia bodies in the Volumetrically insignificant.
central and southern portions of the deposit. These bodies are
characterised by strongly upward-flaring morphology that
extend to the current topographic surface and are progressively
overprinted at depth by the flat-lying advanced argillic blanket.
Hydrothermal breccias These volumetrically significant breccias are described as High-sulfidation mineralisation occurs in both clasts and/or
variably silicified hydrothermal breccias which are ubiquitously matrix and/or acid-leached vugs.
developed within the upper portions of the deposit and
especially within the silicic subfacies. They vary from 0.2 to
10 m in width, have sharp contacts with altered wall-rocks and
comprise sub-rounded to angular clasts of hydrothermal silica.
They vary from clay-matrix breccias (Figure 6c) through to
totally silicified 'ghosted' silica breccias. Cyclic brecciation is
evidenced by up to three successive clast-in-clast textures. Their
distribution within the silicified, upper portions of the lithocap,
their multiple silicification, brecciation and acid vuggy leaching
episodes all imply that the breccias are coeval with, and are an
integral part of, the highly dynamic high-sulfidation
mineralisation and alteration system.
Pebble dykes These breccia bodies are constricted within narrow subvertical Volumetrically insignificant.
permeable structures transecting the high sulfidation zones.
They are usually polymict, have very sharp contacts the host
rocks, and commonly consist of sub-rounded to sub-angular
clast of grey silica and porphyritic silica-clay altered clasts set in
a fine grained silica-clay matrix. High-sulfidation alteration and
mineralisation occur variably both within the clasts and the
matrix, suggesting that the breccias have a range of pre to
post-mineralisation timing relationships.
Shear/fault Recent infill drilling has encountered a number of multiple stage These breccia bodies have been noted to contain significant
shear/fault breccias. These are variably altered to silica-clay or copper and molybdenum mineralisation.
clay dominated assemblages and typically constitute
sub-rounded pebble sized altered fragments from the affected
sheared wall rock. The matrix is essentially clay gouge and
highly milled material containing very fine sulfides. Their
interpreted geometric distribution in the deposit follows a series
of subvertical structures coincident with the major up flow
conduits. These breccias are interpreted to both pre-date and be
synchronous with the overprinting high-sulfidation event.
Sulfide/vein Recent infill drilling has encountered several mineralised sulfide Sulfide breccias consist of a sulfide matrix and micro-fracture
and vein breccias occurring within the high-sulfidation zone. fill as py, veined cc-en and minor bn. The matrix of the
Sulfide breccias comprise generally narrow (<2 m) zones. silica-quartz may contain drussy cavities and veinlets
Occasional intercepts of clast -supported silica-quartz vein containing en and cpy inclusions. These breccias typically
breccias have been recognised. report appreciable gold grades but are usually limited in
distribution.

Note: Much of the information in this tabulation is derived from Rohrlach et al, 1999 augmented by recent observations from infill diamond drilling.
Legend: py-pyrite, en-enargite, luz-luzonite, dig-digenite, bn-bornite, cpy-chalcopyrite, cc-chalcocite, mo-molybdenite, hem-hematite, cov-covellite.

Host lithologies thickness in excess of 650 m between the current erosional

Well preserved original lithologies in the immediate deposit
environment tend to be restricted to peripheral drillholes, in
particular to the immediate north and east of the deposit, as a
consequence of pervasive alteration of the host sequence. The
host Tampakan Andesite Sequence attains a total preserved
surface and hornblende diorite stocks intersected at depth. The
sequence dips at ~10 - 20 to the WSW, generally subparallel to
the tabular zone of copper-gold mineralisation (see section on
mineralisation). Andesitic volcanics of the TAS are the
predominant host to copper-gold mineralisation within the
deposit, in particular high-sulfidation style mineralisation.

180 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A GEOLOGICAL REVIEW OF THE TAMPAKAN COPPER-GOLD DEPOSIT

Intrusives of hornblende diorite composition occurring as

stocks and dyke-like apophyses have been intersected at deeper
levels of drilling and are related to a more extensively developed
diorite body at depth. Original textures of these intrusives are
usually masked by intense chlorite-sericite alteration. Narrow
intercepts of pre-mineralisation andesite porphyry dykes
generally altered to clay-sericite-chlorite assemblages, are rarely
observed.
Dating of these intrusives (Rohrlach, 2002) has produced a
range of ages from pre-porphyry mineralisation to post-porphyry
and broadly contemporaneous with the high-sulfidation
mineralisation event. However, no intrusions have been dated
that are synchronous with the porphyry copper mineralisation
event (Table 1).

Alteration
A B
Three major alteration types (zones) are recognised in a
vertically zoned pattern within the host environment of the
Tampakan copper-gold deposit; argillic, advanced-argillic and
intermediate argillic (SCC). The intermediate argillic alteration
(SCC) is considered by the authors to be related to an earlier
porphyry-style mineralisation event, whilst the argillic and
advanced-argillic alteration are related to a later overprinting
high-sulfidation mineralisation event. Patchy domains of phyllic
alteration (sericite-quartz-pyrite) can occur at or near the
interface between the advanced-argillic and intermediate argillic
alteration zones. Minor intervals of relict porphyry-related
potassic alteration have been intersected near the margins of host
hornblende diorites in some deeper drillholes. Propylitic
alteration is generally restricted to higher stratigraphic levels at
the periphery of the deposit and is widespread in the surrounding
district. Most high-grade copper-gold mineralisation within the C D
Tampakan copper-gold deposit is restricted to the
advanced-argillic zone of alteration. The alteration types and
their textural subfacies as shown in Figure 5 and described in
Table 2 and below, are largely reproduced from Rohrlach et al FIG 6 - Selected core photographs. (A) TMPD101; 80 m. Massive
(1999) with some modification from the documentation of recent silica with alunite (silicic zone). Strong digenite/bornite with
infill drilling. Figure 6 shows core photographs of some of the enargite + pyrite mineralisation in microfractures. (B) TMPD114;
alteration types. 212 m. Clotted silica-clay alteration. Strong digenite/bornite +
pyrite mineralisation in clay clots (white) and microfractures.
(C) TMPD101; 1020 m. Polymict breccia. Note silicified clast in
Argillic alteration clast. Breccia matrix is clay altered. One vuggy silica clast contains
Argillic alteration is the uppermost alteration zone within the py+dg/bn mineralisation. (D) TMPD109; 234 m. Porphyritic
Tampakan deposit environment and is predominantly hosted by silica-clay alteration with ghosted quartz stockworks from porphyry
cycle 4a of the Tampakan Andesite Sequence (TAS) described by stage mineralisation. Minor py+dgen associated with clay (white).
Rohrlach (2002). This alteration comprises an assemblage of
illite-smectite-chlorite that has replaced most of the original
result of intense leaching under acidic conditions generated
mineralogy of host andesites but quite commonly preserved
during the high-sulfidation stage of mineralisation with the
original igneous textures. Although volumetrically minor in
silicic subtype representing zones of more intense leaching and
occurrence, some silica-altered (advanced-argillic) zones are also
deposition of hydrothermal silica.
locally present, usually proximal to faults. Only minor
mineralisation is associated with this alteration zone. Earlier stage porphyry-related quartz stockworks occur as
ghosted or relict textures throughout, but generally in the lower
parts of the advanced-argillic zone. In the silicic and silica-clay
Advanced-argillic alteration
subtypes of advanced-argillic alteration ghost quartz veinlets
A zone of advanced argillic alteration underlies the argillic are present in low density (<2 vol per cent/m), and have been
alteration zone and is the host to the majority of copper metal occasionally noted even in the upper levels of the acid leached
within the deposit. This alteration facies is widespread and silicic zone. In the clay-silica alteration subtype, these
extends into the surrounding district. It usually ranges in porphyry-related quartz veinlets are present up to 20 vol per
thickness from 50 m to 200 m and generally thins towards the cent/m. The quartz stockworks are wispy and generally
margins of the deposit. It can attain a maximum thickness in the unmineralised with original associated sulfides interpreted to
order of 400 m within structurally controlled corridors, eg the have been leached and remobilised. In the high-sulfidation
Pula Bato Fault corridor (Figure 4). Advanced-argillic alteration overprinted zone the porphyry-related quartz veinlets have
is divided into three generally vertically distributed subtypes: generally assumed a bleached colour of the latest stage
silicic, silica-clay and clay-silica depending on the relative ratios silicification, in contrast to the pink-coloured porphyry related
of silica and clay. These subtypes are represented pictorially in quartz stockworks in the intermediate argillic (SCC) alteration
Figure 5 and described in Table 2. This alteration type is the zone described below.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 181

C MIDDLETON et al
Intermediate argillic alteration (SCC)
The base of the clay-rich high-sulfidation alteration subtype is
gradational towards porphyry-related alteration composed
mainly of sericite-clay-chlorite (SCC) with abundant pyrite and
silica. Illite is the most common clay species. This alteration
zone is spatially related to extensive porphyry-stage quartz
stockwork veining (Rohrlach, 2002). Preservation of the original
porphyritic rock texture and pinkish quartz veins with
pyrite-chalcopyrite are commonly observed. Rohrlach, 2002
refers to this intermediate argillic zone as the sericite-chlorite
alteration zone and considers that it is a bona fide facies and not
related to late overprinting of one alteration style on another. The
intermediate argillic zone contains significant specular hematite
as primary quartz-chalcopyrite-hematite veins and as
chalcopyrite-hematitepyrite disseminations with hematite
possibly forming by hypogene oxidation of precursor magnetite
(Rohrlach, 2002).
A gypsum-anhydrite subzone manifested by the presence of
gypsum-anhydrite veins (five to 20 per cent by volume) has been
defined at depth as an overprinting alteration to the intermediate
argillic (SCC) alteration zone. In most cases the upper contact of
this subzone is sharp although it generally exhibits an undulating
morphology, often subparallel to the distribution of
advanced-argillic alteration. The upper contact of the gypsum
subzone beneath the thick advanced-argillic alteration within the
Pula Bato fault zone shows a steep downwards gradient and in
some cases has not been intersected in drilling. Previous workers
have interpreted the gypsum as product of hydration of anhydrite
veins (Reynolds, 1997). Rohrlach (2002) has interpreted this
undulating gypsum horizon as a downward migrating
sulfate-saturation front, where sulfates dissolved from the
advanced-argillic alteration zones by descending meteoric waters
are re-precipitated as low temperature gypsum veins at depth.
Potassic alteration
A relict biotite-bearing potassic alteration assemblage related to
porphyry style mineralisation has been intersected near the
margin of host hornblende diorite intrusives in a few deep WMC
drill holes. The potassic assemblage consists of
biotite-magnetite-anhydrite that is overprinted by widespread
sericite-chlorite (intermediate argillic) alteration (Rohrlach,
2002). Within the potassic-altered zone, magnetite is locally
present at up to eight volume per cent. Rohrlach has interpreted
that the level where these patches of biotite-magnetite-anhydrite
assemblage are found may be the transition zone between the
intermediate argillic and potassic facies. Whilst this alteration
type might be a vector towards a higher-grade core to a porphyry
system, drilling has yet to intersect high-grade porphyry related
copper mineralisation.
Breccias
Several breccia types are recognised within the deposit
environment, including pebble breccias, hydrothermal breccias,
shear/fault breccias, sulfide/vein breccias and diatreme breccias.
These range in timing from early and locally developed
autoclastic flow-breccias and intrusion carapace breccias to
spatially restricted intermediate-stage diatreme breccias to
ubiquitous syn-high-sulfidation hydrothermal breccias and fault
breccias (Rohrlach, Madera and Watt, 1999).
Table 2 describes the distribution and characteristics of these
breccias.
Post-mineralisation intrusives and extrusives
At least two generations of late-stage diorite and porphyritic
andesite dykes that can represent up to four per cent by volume,
intrude high-sulfidation mineralisation and associated alteration
182 Adelaide, SA, 19 - 22 September 2004
(Rohrlach, Madera and Watt, 1999). Narrow (1 - 10 m)
porphyritic andesite dykes or wider (10 - 30 m) porphyritic
diorite dykes can display chilled or sheared margins and often
contain xenolithic fragments of hydrothermally altered country
rocks. Recent infill drilling has established that these dykes often
occur at the boundary of alteration domains or are intimately
associated with faults and shear zones. Locally some of these
intrusives are affected by weak argillic or propylitic alteration
characterised by chlorite, epidote and magnetite but they are
usually devoid of copper sulfides.
Narrow unaltered dykes that intrude the high-sulfidation
system are thought to be coeval with the extrusion of a much
later andesitic to dacitic flow-dome complex (Logdeck Andesite)
that unconformably overlies the Tampakan deposit south of the
Bangayo Fault. This complex has been dated by Rohrlach (2002)
at 1.47 Ma.
MINERALISATION
Deposit geometry
The delimiting 0.3 per cent copper contour at Tampakan defines a
broad and coherent tabular body of mineralisation that is flat-lying
to gently-dipping towards the southwest. This mineralised zone
generally ranges in thickness from 200 to 400 m and is intersected
from shallow depths at RL1200 m in the northern section of the
deposit (716 190 mN) and extends at depth to RL600 m at the
southern end (714 670 mN). The mineralised zone comprises a
dominant upper high-sulfidation mineralised domain that is
superimposed on an earlier-formed porphyry-copper domain at
depth. The vertical thickness of the high-sulfidation zone
increases adjacent to or around structural corridors (eg the Pula
Bato Fault Corridor and the Lawit Fault). Figure 4 outlines the
current plan projected limits of significant copper mineralisation
at grades 0.5 per cent copper. At this grade contour,
mineralisation extends over distances of 1.75 km N-S and 1.5 km
E-W. The projected western margin of the orebody crops out at
surface along a steep topographic slope. Mineralisation at or
above this grade remains open to the south where the deposit is
downthrown by the Bangayo Fault and unconformably overlain
by the Logdeck Andesite. A possible WNW-trending structure
delimiting mineralisation to the north has yet to be verified by
drilling.
A higher-grade zone of copper mineralisation has been defined
in the western and northern sectors of the Tampakan copper-gold
deposit based on drillhole intersection selection criteria of
100 m per cent copper and 1 per cent copper (Figure 4). This
higher-grade mineralised zone extends in a NNE orientation over
a distance of approximately 1.2 km and is generally coincident
with the Pula Bato Fault Corridor, a prominent lineament in
DTM data and an inferred feeder zone for high-sulfidation
mineralisation. This higher-grade zone broadens to a
quadrangular area of approximately 500 m 500 m to the north
of the interpreted position of the WNW-trending Danlag Fault.
This area is the current focus of an 80 80 m infill and
delineation diamond drilling program as part of a preliminary
feasibility study. Figure 7 portrays a series of stacked,
80 m-spaced assay cross-sections through this area and indicates
that the higher-grade zone can attain a thickness in excess of
300 m and is generally upward flaring. This morphology is
broadly consistent with the extent of high-sulfidation associated
alteration, and suggests that there was a strong underlying
structural control to mineralising fluid flow. Selected
intersections from this higher-grade zone are given in Table 3.
South of the Danlag Fault the higher-grade zone attains a
maximum width of 250 m and gradually tapers southwards.
PACRIM 2004
 A GEOLOGICAL REVIEW OF THE TAMPAKAN COPPER-GOLD DEPOSIT

TABLE 3 encroachment during thermal waning of the hydrothermal system

(Rohrlach, 2000). The distribution and characterisation of
Selected intersections from the Tampakan high-grade zone
( 100 m per cent copper and 1 per cent copper). high-sulfidation mineralisation within the various subtypes of
advanced argillic alteration are given in Figure 5 and Table 2).
Hole No From (m) Interval (m) Cu (per cent) Au (g/t)
TMPD22 65.00 156.00 1.40 0.65 Porphyry-style mineralisation
TMPD27 50.00 142.00 1.24 0.45 Porphyry-style copper mineralisation is most clearly seen in the
TMPD35 168.00 84.00 1.14 0.54 deeper part of the deposit below the high-sulfidation system and
TMPD37 48.00 100.00 1.39 0.55 is generally associated with intermediate argillic alteration (SCC)
of the host andesite sequence and diorite intrusives. Within the
TMPD39 98.00 172.00 1.04 0.40 zone of high-sulfidation overprinting, original porphyry-related
TMPD40 83.00 130.00 1.38 0.54 quartz stockwork development is expressed only as ghost
TMPD41 48.00 194.00 1.16 0.50 veinlets masked by the later alteration assemblages (Figure 6d).
Porphyry-style mineralisation consists predominantly of
TMPD42 125.90 110.10 1.20 0.65 disseminated pyrite and chalcopyrite with minor bornite. Minor
TMPD46 86.00 76.00 2.20 0.50 molybdenite and trace galena have been observed. This
TMPD47 78.25 102.75 1.34 0.31 porphyry-style mineralisation is dominated by low to moderate
density gray-pink coloured, multi-directional, crudely laminated
TMPD51 106.20 94.80 1.09 0.31 quartz veinlets associated with silica-sericite-chlorite-hematite
TMPD53 243.00 138.00 1.08 0.39 alteration. Typical copper grades vary from 0.1 - 0.6 per cent and
TMPD55 62.00 76.00 1.46 0.68 show a tendency to decrease with depth with an increasing
pyrite: chalcopyrite ratio. In the gypsum-anhydrite subzone
TMPD62 89.00 196.00 1.48 0.50 minor copper sulfide selvages to gypsum veins have been
TMPD64 100.00 144.00 0.99 0.70 observed.
TMPD67 165.50 185.50 1.31 0.35
TMPD71 138.00 130.00 1.18 0.37 Oxidation and supergene alteration
TMPD81 235.00 46.00 2.00 0.42 Very limited supergene mineralisation has been observed in the
TMPD85 110.70 134.30 1.24 0.30 Tampakan drill core. This could be attributed to the relatively
thin regolith and high rate of erosion on the steep terrain. Rare
TMPD87 258.00 130.00 0.99 0.57 sooty supergene chalcocite, usually associated with fractures, has
TMPD88 170.00 148.00 1.00 0.31 been observed in a few drillholes near the base of the silicic zone
TMPD90 103.00 92.00 1.15 0.43 (eg TMPD 103). Malachite, azurite and chrysocolla have also
been noted in insignificant quantity from drill cores and during
TMPD92 144.00 196.0 1.55 0.47
surface mapping (Reynolds, 1997). Short intervals of green
TMPD96 56.00 110.00 1.75 0.70 chrysocolla incrustations along chalcocite-pyrite-rich veins have
TMPD97 6.30 117.50 1.72 0.33 been observed in a few recent infill drillholes.
TMPD98 95.00 55.60 2.05 0.81
TMPD100 114.00 374.00 1.15 0.62 Metal zonation
TMPD102 82.00 395.00 1.11 0.53 The majority of drillcore sampling within the deposit has been
TMPD103 35.30 378.05 1.09 0.44 undertaken at 2 m intervals with multi-element analysis for Cu,
Au, Ag, As, Mo, Pb and Zn. This has enabled the study of metal
TMPD104 136.00 290.00 1.20 0.62 distribution and patterns within the deposit and, in particular,
TMPD105 186.00 230.00 1.04 0.45 between the various alteration and mineralogical domains. When
TMPD106 40.00 191.10 1.23 0.27 treated as a single assay data set, several elemental associations
are present but correlation coefficients are poor. This is a
TMPD107 57.60 178.40 1.53 0.75
reflection of the diverse alteration and sulfide mineralogy
TMPD109 57.4 184.60 1.16 0.52 resulting from an earlier porphyry copper event that is strongly
TMPD110 36.00 178.30 1.18 0.44 overprinted by a later high-sulfidation event. Distinct patterns
and improved correlations are obtained when these geological
domains are taken into consideration.

Mineralisation type and zonation Copper

High-sulfidation mineralisation
High-sulfidation mineralisation in the Tampakan deposit is
characterised by the presence of high-sulfidation-state sulfide
assemblages dominated by pyrite, sulfosalts (enargite luzonite),
digenite, bornite and lesser covellite, chalcocite, chalcopyrite,
tenantite-tetrahedrite and molybdenite. Traces of colusite
(Cu-Sn-V-As-Fe sulfide), sphalerite and galena have also been
reported (Madera, 2000). The most common copper sulfides
display a very distinctive replacement sequence from bornite
digenite chalcocite exsolution covellite. This retrograde
alteration sequence formed during the waning stage of the
high-sulfidation hydrothermal system, and is interpreted to
represent high temperature oxidation by gradual ground water
Higher-grade (>0.7 per cent) copper distribution within the
Tampakan deposit is strongly correlated with the morphology of
advanced-argillic alteration. In the north, between sections
715 630N and 716 190N, this higher-grade zone is represented
by a relatively shallow, flat-lying blanket of mineralisation that
thickens considerably within the NNE-trending Pula Bato Fault
zone where deeper mineralisation is more typically disseminated
and hosted by high-sulfidation silica-clay and clay-silica
alteration subtypes (Table 2, mineralisation Types 3 and 4).
South of 715 630N, higher-grade copper mineralisation tends to
be restricted to the Paul Bato structure.
Copper distribution within the silicic zone tends to be quite
erratic. This is interpreted to be due to the irregular nature and
type of sulfide mineralisation associated with vug and fracture

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 183

C MIDDLETON et al

TAMPAKAN COPPER-GOLD PROJECT

D20
E
Drill Cross-Section 715950ND105 D21
D100
Note: D92 D54
D22
D62 / 92 Projected South to North 1200
D5 43.6/0.92 EOH: 101.4m
50.6/1.72
142/1.24
(Metre RL)

.40
6/1
15
36/1.19

4
/1.0
100/1.39

230
1000
EOH: 343.6m
W D62
D23
504600E

506100E
42/1.18

.06
EOH: 351.1m
D27

.4/1
D37

323
D91 38/1.24 EOH: 319.3m EOH: 357.6m

EOH: 402.4m D85

800 EOH: 459.9m EOH: 506.0m

EOH: 420.0m

EOH: 527.40m
EOH: 574.3m

TAMPAKAN COPPER-GOLD PROJECT

E
Drill Cross-Section 715870N D104
4m@28.2 g/t Au D102
D94 D108 ? 1200
D103
D62 31.0/1.22
Base of partial 36.0/2.19 ?
(Metre RL)

oxidation
/0.7
.9
112

130.2/1.68

38.1/1.48
44.0/0.77
1000
2
/0.6

20.0/1.23
W
97.0

?
6

.33
/1.0

40.4/0.76
.5/1
504600E

506100E
? D54
.87
254

EOH: 283.4m 23.4/1.10

261

.4/0
201
?

D91
EOH: 407m
800 EOH: 410.5m EOH: 444.5m

EOH: 479.4m
EOH: 439.5m

D43
TAMPAKAN COPPER-GOLD PROJECT
E
Drill Cross-Section 715790N D42
1200
3.5/1.23 Au g/t
D110
D81 D109
30/1.14 Au g/t
(Metre RL)

20/1.36 Au g/t
101.4/1.00
3

14/2.20 Au g/t
/1.3
130

76/2.20
146

.20
184

/1.3

4/1.15 Au g/t
.99

1/1
.6/1
4/0

.
110

44/1.64 40/1.00
14

.17

1000

W
506100E

46/2.00
00
/1.
504600E

98

D64
D40 D46 800
D82 Approximate limit of
1.0 % copper cut-off
Higher Grade
Metres / Cu (%)
Intersection Approximate limit of
EOH: 523 m. 200 m. 0.5% copper cut-off
EOH: 519 m.
No vertical exaggeration
Upper limit of low As (<150 ppm)

FIG 7 - Stacked assay sections from the higher-grade copper zone in the north-central part of the deposit.

184 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A GEOLOGICAL REVIEW OF THE TAMPAKAN COPPER-GOLD DEPOSIT

fill and to a lesser extent sulfide veining. Strongly mineralised

intervals (>3 per cent Cu) often occur at or near the base of the TABLE 4
silicic alteration zone or in proximity to structures. In the Tampakan resource estimate.
northern sector of the deposit infill drilling has established that
copper grades within the underlying silica-clay and clay-silica Indicated Resources
alteration tend to be much more uniformly distributed (0.7 - 1.5 Cu cut-off Mt Cu Au CuEq Cu Au
per cent Cu) although higher grades are generally developed (%) (%) (g/t) (%) (Mt) (Moz)
proximal to structures or associated with breccia zones. Copper 0.2 663 0.65 0.27 0.81 4.312 5.758
grades in the underlying porphyry-related zone of intermediate
argillic alteration (SCC), typically range from 0.1 - 0.6 per cent 0.3 576 0.71 0.29 0.88 4.092 5.373
copper but generally decrease with the development of 0.4 481 0.78 0.32 0.97 3.754 4.952
gypsum/anhydrite veining. 0.5 384 0.86 0.34 1.06 3.302 4.197
0.6 311 0.93 0.37 1.14 2.889 3.695
Arsenic
0.7 245 1.01 0.39 1.24 2.474 3.071
There is a strong correlation between arsenic distribution and the 1.0 132 1.28 0.48 1.56 1.690 2.039
extent of silicification because enargite is stable in the same acid
Inferred Resources
conditions that generate residual silica. Arsenic enriched zones
(>150 ppm As) are restricted to the silicic lithocap and within or Cu cut-off Mt Cu Au CuEq Cu Au
adjacent to structures in underlying silica-clay and clay-silica (%) (%) (g/t) (%) (Mt) (Moz)
zones. Individual assays in enriched zones commonly exceed 0.2 610 0.54 0.22 0.67 3.292 4.313
1000 ppm As. Median arsenic values for silica, silica-clay and 0.3 528 0.58 0.23 0.71 3.063 3.905
clay-silica advanced-argillic alteration are 196 ppm, 16 ppm and 0.4 396 0.66 0.26 0.81 2.612 3.308
10 ppm respectively. As a consequence of the above relationship,
the high arsenic zone usually has a distinct subhorizontal 0.5 275 0.75 0.29 0.92 2.063 2.565
morphology (Figure 7) but has been dismembered by 0.6 194 0.83 0.31 1.01 1.613 1.937
NW-trending faulting with successive south- block-down 0.7 131 0.92 0.35 1.12 1.202 1.470
displacement. 1.0 19 1.22 0.47 1.49 0.230 0.289
Indicated and Inferred Resources
Gold
Cu cut-off Mt Cu Au CuEq Cu Au
Gold is generally more elevated in the zone of high-sulfidation (%) (%) (g/t) (%) (Mt) (Moz)
alteration where median values of 0.3 ppm, 0.2 ppm and 0.2 1273 0.59 0.25 0.74 7.511 10.232
0.12 ppm are associated with silica, silica-clay and clay-silica
0.3 1104 0.65 0.26 0.80 7.178 9.232
alteration respectively. Higher-grade copper zones (0.7 per cent
Cu) are usually associated with grades of 0.2 - 0.6 ppm Au. 0.4 877 0.72 0.29 0.89 6.315 8.178
Within the primary sulfide zone, intervals of higher gold 0.5 659 0.81 0.32 1.00 5.338 6.780
concentration up to 2 ppm are usually spatially associated with, 0.6 505 0.90 0.35 1.10 4.544 5.682
or in proximity to, the Pula Bato Fault zone, an inferred
0.7 376 0.98 0.38 1.20 3.680 4.588
structural feeder. Intervals of secondary gold enrichment
(1 - 2 ppm) commonly occur within the oxidised part of the 1.0 151 1.27 0.48 1.55 1.920 2.328
silicic caprock but preferentially above the Pula Bato structure.
Note: Copper equivalents are calculated by the addition of
0.58 times the gold grade (g/t) to the copper grade (per cent).
Molybdenum Mt = million metric tonnes, Moz = million oz
Significantly anomalous (100 ppm) molybdenum is commonly Source: Hellman and Schofield, 2004
associated with higher copper grades both in the silicic zone of
alteration or within structures in the underlying silica-clay and
clay-silica zones. TMPD104 intersected 78 m at 0.17 per cent Mo the provisions of the 1999 Joint Ore Reserves Committee Code
from a downhole interval of 292 - 370 m in strongly overprinted (JORC) of The AusIMM. The revised resource estimate only
clay-silica alteration. incorporates mineralised blocks that lie within a conceptual open
pit which has been generated using Whittle software with a pit
slope of 40 and conceptual costs, metallurgical recoveries and
Resource estimation arsenic penalties. A copper price of $A 1.25/lb and a gold price
Details in respect of a previously published resource estimation of $US 380/oz have been assumed for the definition of this pit
by WMC (1998) have been documented by Rohrlach, Madera base. The estimation was undertaken by ordinary kriging with a
and Watt (1999) and Madera and Rohrlach (1998). This previous check estimate in the central part of the deposit by multiple
global resource estimate was based on a 0.2 per cent copper indicator kriging. Four domains were used for the estimation: A
delimiting copper cut-off grade and broad spaced drilling domain of copper depletion immediately above the high
throughout the deposit that generally varied in density from a sulfidation zone (Domain 1): the high sulfidation zone (Domain
320 m 320 m to 160 m 160 m collar spacing. 2); porphyry style mineralisation (Domain 3) and, a peripheral
area of low density drilling (Domain 4) actually excluded from
An ongoing program of detailed 80 80 m infill and the final indicated and inferred resource estimate. Other criteria
delineation diamond drilling of the higher-grade zone in the or information adopted include:
north-central part of the deposit, coupled with a program of
re-logging of some previous drillholes for revised domain a total of 22 027 assays available with most representing a
interpretation, has allowed for a major revision to the previous 2 m sample interval;
estimate. The revised indicated and inferred resource estimate 6 m composite assays;
(Table 4) has been prepared by Hellman and Schofield (2004) in
consultation with Sagittarius geoscientists and is compliant with resource block size 40 m 40 m 12 m (N,E,RL);

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 185

C MIDDLETON et al
search parameters (N,E,RL) of 120 m 120 m 24 m and
160 m 160 m 32 m for indicated and inferred categories
respectively (in Domain three the RLs were 45 m and 60 m
respectively);
indicated and inferred categories required a minimum
number of ten and six composite assays respectively for each
resource block;
variography indicates that a drillhole spacing of 80 m - 120 m
is appropriate for the delineation of the Tampakan resource,
assuming an open pit mining strategy; and
copper equivalent grades presented in Table 4 are calculated
by the addition of 0.58 times the gold grade (g/t) to the
copper grade (per cent). This is based on the relative value of
copper and gold at the time the estimation was made.
A considerable volume of potential mineralisation lying
beneath the base of the pit, or lateral to it has not been included
in the indicated and inferred resource tabulation (Table 4) due to
a lower density of drilling information or where blocks fail to
meet the above selection criteria. As an example, 46 out of 111
drillholes (41 per cent) to date have bottomed in mineralisation
with copper grades 0.3 per cent copper (average of 0.5 per cent
copper and 0.2 g/t gold). The average depth of these holes is
368 m. Further, mineralisation incorporated in the resource has
not been closed off to the west or east in most drill sections and
the deposit remains open to the south of the Bangayo Fault.
Consequently, the resource tabulated in Table 4 is deemed
conservative.
DISCUSSION
The spatial correlation between high-sulfidation copper gold
mineralisation and adjacent or underlying porphyry copper
mineralisation has been widely documented in geoscientific
literature (Arribas, 1995; Holzberger et al, 1996; Hedenquest and
Arribas, 1998; Corbett and Leach, 1998). In some cases, a
temporal and genetic link has been demonstrated between the
two styles of mineralisation and a single evolving magmatic
event. The spatially-coupled high-sulfidation and porphyry
copper mineralisation at Tampakan are temporally separated by
approximately one million years (Rohrlach, 2002). He concludes
that the ~1 Ma age difference requires their formation from
separate magmatic-hydrothermal systems that were established
in the upper crust from different batches of melt. Rohrlach et al
(1999), demonstrate some similarities between the Tampakan
deposit and the Lepanto-Far Southeast deposit pair in the
Mankayan district, northern Philippines.
The high-sulfidation mineralisation at Tampakan is
characterised by many of the features associated with these types
of deposit elsewhere, namely: extensive silicification and
hydrothermal brecciation associated with an areally extensive
advanced-argillic alteration lithocap, the presence of high
temperature and acid-stable alteration mineral assemblages, high
sulfidation state sulfide assemblages (enargite-digenite-covellite
chalcocite), dissected paleo-volcanic setting and a strong
underlying structural control etc (Arribas, 1995). The
identification of such acid-stable alteration assemblages and
associated geochemical anomalism in volcanic settings is an
important exploration vector to the discovery of similar deposits.
Copper mineralisation within the deposit is strongly zoned.
Higher-grade copper zones correlate with the morphology of
advanced-argillic alteration. There is abundant evidence for a
strong underlying stratigraphic and structural control to the
distribution of this high-sulfidation mineralisation event and
associated advanced-argillic alteration. This event is
gradationally superimposed on an underlying porphyry copper
system where original textures and stockwork development are
progressively obliterated or masked with increasing acid
leaching. Given the close juxtaposition of the two mineralisation
styles, we assume that there was significant metal redistribution
186 Adelaide, SA, 19 - 22 September 2004
and upgrading during the overprinting hydrothermal event. The
broad tabular morphology of the mineralised zone at Tampakan
is generally conformable with the host andestic sequence
suggesting that the sequence exerted a strong stratigraphic
permeability control on the distribution of mineralisation and
associated alteration (Rohrlach, Madera and Watt, 1999;
Rohrlach, 2002). A higher-grade core to the porphyry system (if
originally present or preserved) has yet to be located, although
there are some vectors that are currently being investigated.
The Tampakan deposit represents a major site of copper metal
accumulation and is the subject of a current preliminary
feasibility study focused on a higher grade zone in the
north-central part of the deposit. Subject to future mining
methods and process technology to manage arsenic, it is possible
that an operating cut-off grade could be as low as 0.3 per cent
copper. At these grades, the mineralisation is extremely coherent,
thick and shallow, adding general support to the interpretation on
which the recent resource estimate was based.
Little advanced exploration has been undertaken over the
district-scale advanced-argillic and argillic alteration zone
external to the main drilling grid. This augurs well for the future
discovery potential of other high-sulfidation copper-gold and
porphyry copper systems in the district.
ACKNOWLEDGEMENTS
The authors would like to thank and acknowledge Indophil
Resources NL and Sagittarius Mines Inc, for permission to
publish this paper on the Tampakan copper-gold deposit. We
appreciate the support and encouragement of management, in
particular A Roberts and T Robbins in the preparation of this
paper. We acknowledge all previous investigations and
documentation by WMC geoscientists who worked on the
deposit over a ten year period prior to its acquisition by
Sagittarius. In particular we would like to acknowledge the
research work undertaken by one of our fellow authors, B
Rohrlach, in a recent PhD thesis at ANU in Canberra. We are
indebted to our fellow workers at Sagittarius and Indophil for
their considerable support in our ongoing investigations and we
thank D Nihill for his review of the text and useful comments.
The authors also thank the anonymous AusIMM reviewer for
their recommendations in respect of improvements to our initial
draft.
REFERENCES
Arribas, A Jr, 1995. Characteristics of high-sulfidation epithermal
deposits, and their relation to magmatic fluid, in Magmas, Fluids and
Ore Deposits, (Ed: J F H Thompson), Vol 23 (The Mineralogical
Association of Canada).
Balkau, J, Buenavista, A G, Norris, M, Watt R D and Marcos, D, 1998.
The discovery of the Tampakan Cu-Au deposit, Mindanao,
Philippines, in Proceedings 104th Annual Northwest Mining
Association Meeting SEG Giant Ore Deposits, Spokane,
Washington, 2 - 4 December.
Buenavista, A, Ignacio, A and Edris, M, 1995. Geology of the Mt
Matutum quadrangle and vicinity, unpublished WMC internal report
(K3468), pp 1-27.
Corbett G J and Leach, T M, 1998. Southwest Pacific Rim gold-copper
systems: structure, alteration and mineralization, Society of
Economic Geologists, Special Publication No 6.
Domasig, W, Marcos, D and Protacio, A, 1998. Results of mineral
exploration over the Columbio FTAA in Southern Mindanao,
Philippines, unpublished WMC internal report.
Hall, R, 1996. Reconstructing Cenozoic SE Asia, in Tectonic Evolution of
Southeast Asia, (Eds: R Hall and D Blundell) pp 153-184, Geological
Society Special Publication No 106.
Hedenquest, J W and Arribas, A Jr, 1998. Evolution of an
intrusion-centred hydrothermal system: Far Southeast-Lepanto
porphyry and epithermal Cu-Au deposits, Philippines, Bulletin of the
Society of Economic Geologists, 93(4).
PACRIM 2004
 A GEOLOGICAL REVIEW OF THE TAMPAKAN COPPER-GOLD DEPOSIT

Hellman, P and Schofield N, 2004. Tampakan Resource Estimate,
unpublished resource estimate report prepared for Indophil
Resources NL, July.
Holzberger, I R et al, 1996. The Freida River-Nena copper and gold
deposits, discovery and assessment, a case example, in Proceedings
AMF Conference, Cairns, Porphyry related copper and gold deposits
of the Asia Pacific Region.
Madera, A, 2000. Geology and mineral resource potential within the
Columbio FTAA of WMC Philippines, Inc, unpublished WMC
internal report.
Madera, A and Rohrlach, B D, 1998. Geology and resource potential of
the Tampakan Copper Deposit, Southern Mindanao, in Proceedings
Geocon 98 XI Annual Geological Society of the Philippines
Convention.
Reynolds, L, 1997. Geology of the Gintong Bato copper-gold deposit,
unpublished WMC internal report (K3729).
Rohrlach, B D, 2002. Tectonic evolution, petrochemistry, geochronology
and palaeohydrology of the Tampakan porphyry and high
sulphidation epithermal Cu-Au deposit Mindanao, Philippines, PhD
thesis (unpublished), Australian National University, Canberra.
Rohrlach, B D, Loucks, R R and Palin, J M, 2002. Intra-arc compressive
stress promotes development of slow-cooling, multiply-replenished
lower crustal magma chambers parental to porphyry copper deposits,
in Proceedings Geoscience 2002: Expanding Horizons. 16th
Australian Geological Convention, 1 - 5 July, Adelaide. Abstracts
Volume 67, pp 91.
Rohrlach, B D, Madera, A Y and Watt, R D, 1999. Geology, alteration
and mineralisation of the Tampakan copper deposit, in Proceedings
PACRIM 99 Congress, pp 517-525 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
WMC, 1998. Quarterly report to shareholders 31 March.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 187

The Tintina Gold Province, Alaska and Yukon New World-Class
Gold Resources and Their Sustainable Development
S H Mueller1, R J Goldfarb1, C J R Hart2, 3, J L Mair3, E E Marsh1 and C S Rombach1,4
ABSTRACT
The Tintina Gold Province (TGP) is a 150 000 km2 area comprised of
>15 individual gold belts and districts of intrusion-related and orogenic
gold deposit types, extending from southwestern and interior Alaska
(USA) to eastern Yukon (Canada), and bounded by the Tintina-Kaltag
and Denali-Farwell regional fault systems. The development of the Fort
Knox (5.4 Moz at 0.95 g/t), and Brewery Creek (0.85 Moz at 1.44 g/t)
deposits, and discovery of the Pogo (5.8 Moz at 18.86 g/t) and Donlin
Creek (25.4 Moz at 2.5 g/t) deposits, have prompted resurgence in
exploration for bulk-tonnage gold deposits within the TGP. With the
exception of Pogo, most of the economically important deposits within
the province represent new, low-grade resources at the sites of historic
small prospects and placer deposits. The deposit styles can best be
categorised as:
1. sheeted veins in both the apical portion of ca. 90 Ma plutons and in
adjacent wallrock;
2. replacements, disseminations, and breccias located either above or
adjacent to Cretaceous plutons within the hornfelsed aureoles;
3. fault-related ores that range from quartz bodies in ductile shear
zones, to brittle, gouge-filled, sulfide-rich fractures, to disseminated
ores along or adjacent to shear zones these all may occur proximal
or at significant distances from Cretaceous plutons;
4. epizonal stockwork mineralisation and related breccias; and
5. gold-bearing skarns.
The deposits of the TGP have environmental signatures characterised
by anomalous arsenic and antimony concentrations in both ground and
surface water, as well as anomalous concentrations of As, Sb, Hg, Bi, and
W in stream sediments and silts, and in soils. The sustainable
development of the TGP deposits is dependent on the proximity of the
deposit to pre-existing infrastructure, the beneficiation method, and the
land management and social politics associated with the environmental
and economic impacts of mine development in Alaska and Yukon.
INTRODUCTION
The The Tintina Gold Belt was a concept put forward eight
years ago by junior exploration geologists to highlight the 68
million oz (Moz) gold endowment of Yukon and Alaska (Tucker
and Smith, 2000; Hart et al, 2002). This terminology has since
been changed to the Tintina Gold Province (TGP), owing to the
fact that rather than a single gold belt, the region is composed of
numerous belts and districts (Bundtzen et al, 2000). Recent
exploration efforts in the TGP have primarily focused on a new
class of intrusion-related gold deposits associated with
Cretaceous stocks and dikes (Hart et al, 2002). These deposits
account for approximately 35 Moz of in-ground gold resources
and are the source for considerable placer production during the
previous one hundred years. In this paper, the regional geology
of the TGP, as well as more local geology of individual mining
districts and deposits, are reviewed, with emphasis on deposits in
1. United States Geological Survey, MS 973 Dever Federal Center,
Denver CO 80225, USA.
2. Yukon Geological Survey, Box 2703 (K-10), Whitehorse Yukon
Y1A 2C6, Canada.
3. Centre for Global Metallogeny, University of Western Australia,
Perth WA.
4. Anglogold North America, Greenwood Village CO, USA.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
the Selwyn basin, Fairbanks district, Goodpaster district, and
Kuskokwim region. Furthermore, the environmental signatures
of deposits, both in surrounding soil and water, are described and
related to the exploration and sustainable development of these
and, as of yet, undiscovered deposits.
REGIONAL GEOLOGY
The TGP is underlain by multiple and diverse geologic terranes
and, except for eastern Yukon, is bounded by the Denali-Farewell
and Tintina-Kaltag fault systems to the south and north,
respectively (Foster et al, 1994). The province is underlain by
three dominant tectonic regimes, which, from east to west, are as
follows:
1. east-central Yukon, north of the Tintina fault, which is
composed of turbidite sequences, basinal clastic rocks, and
limestones that were deposited as a Neoproterozoic to
middle Paleozoic continental margin sequence of the
Selwyn basin (Gordey and Anderson, 1993);
2. polymetamorphosed Neoproterozoic and Paleozoic
metasedimentary and meta-igneous assemblages of the
pericratonic Yukon-Tanana terrane, which can be divided
into multiple subterranes that underlie much of east-central
Alaska (Foster, Keith and Menzie, 1994); and
3. the amalgamated Paleozoic-Mesozoic basement terranes
that are overlapped by the >10-km-thick, Cretaceous flysch
sequences of the Kuskokwim basin (Bundtzen and Gilbert,
1983; Decker et al, 1994; Miller and Bundtzen, 1994).
Given the diverse geologic terranes and overlap assemblages
underlying the TGP, as well as a complex pre-, syn- and
post-deformational magmatic history between 110 and 70 Ma, it
is not surprising there are multiple styles of gold mineralisation
associated with the orogenic (epizonal through hypozonal) and
intrusion-related deposits that comprise the TGP. In fact, specific
classification of many deposits as intrusion-related or orogenic
(eg Pogo, Donlin Creek) remains controversial (Groves et al,
2003). Mineralisation styles include:
1. sheeted veins in the apical portion of ca. 90 Ma plutons and
in their adjacent wallrocks;
2. replacements, disseminations, and breccias located either
above or adjacent to Cretaceous plutons within their
hornfelsed aureoles;
3. fault-related ores that range from quartz bodies in ductile
shear zones, to brittle, gouge-filled, sulfide-rich fractures,
to disseminated ores along or adjacent to shear zones-these
may be proximal or at significant distances from
Cretaceous plutons;
4. epizonal stockwork mineralisation and related breccias; and
5. reduced gold-bearing skarns (Hart et al, 2002) (Figure 2).
In general, there is an underlying age relationship for the
plutonism and hydrothermal activity, with most reported ages
ranging from 107 - 89 Ma (Selwyn basin, Yukon-Tanana terrane)
in the east to 71 - 65 Ma in the west (Kuskokwim basin). Further
description of the mineralisation styles will be addressed below
in the deposit descriptions.
189
S H MUELLER et al
FIG 1 - The Tintina Gold Province is made up of numerous gold belts and districts throughout central Alaska and Yukon. TB = Tombstone
Belt, GP = Goodpastor District, FB = Fairbanks District, KK = Kuskokwim District. Modified from Hart et al, 2002.
FIG 2 - Schematic representation of the three main styles of gold
ore associated with Cretaceous intrusive rocks in the Tintina Gold
Province. Modified from Hart et al, 2002.
DEPOSITS OF THE SELWYN BASIN
Brewery Creek
The Brewery Creek gold deposit, located approximately 60 km
to the east of Dawson City in the foothills of the Olgilvie
Mountains of central Yukon, is near the northeastern boundary of
the Tintina regional fault and in the hanging-wall of the Robert
Service thrust fault (Diment, 1996; Lindsay et al, 1999) (Figure
1). The deposit is hosted in clastic rocks of the Cambrian to
Lower Devonian Road River Group, Devono-Mississippian Earn
Group, and Neoproterozoic-Cambrian Hyland Group (Diment
and Craig, 1999), in addition to numerous dikes and sills of the
94 - 92 Ma Tombstone Plutonic Suite (TPS). Intrusive rocks
range in composition from high-K diorite to quartz monzonite,
and include a quartz-syenite stock (Diment and Craig, 1999;
Lindsay et al, 1999).
The Brewery Creek deposit is characterised by an epizonal
style of mineralisation that has been dated at 91 Ma (Lindsay et
al, 1999). The orebodies typically occur as quartz-poor veinlets,
with a few per cent arsenopyrite and pyrite with arsenic- and
190 Adelaide, SA, 19 - 22 September 2004
gold-enriched rims, as well as disseminated sulfides in the less
competent metasedimentary rocks (Diment and Craig, 1999).
The mineralisation is structurally controlled, trending
northeasterly along a 12 km zone, with the bulk of the
mineralisation concentrated along faults and shears (Diment and
Craig, 1999). The ore has an Au-As-Sb-Hg geochemical
signature, with stibnite post-dating the main phase of gold
deposition. The gold is refractory and is found in both the
arsenopyrite and pyrite (Diment, 1996; Diment and Craig, 1999).
However the deposit has been oxidised to a depth of 100 m,
forming scorodite and geothite, which allowed for heap leaching
of the gold ore. Hypogene ore remains unmined and a target for
future operations. Brewery Creek was discovered by soil
sampling in a region of weak As-Sb anomalies obtained from a
Yukon government regional stream sediment survey (Hart et al,
2002). The pre-mining resource was 13.3 Mt at 1.44 g/t for a
total of 0.85 Moz Au. During mining of the supergene zone,
from 1997 - 2002, a total of 0.27 Moz Au was recovered (Hart et
al, 2002) (Table 1).
Scheelite Dome
The Scheelite Dome gold deposit is located 25 km north of the
town of Mayo in central Yukon (Figure 1). It was first prospected
for its placer gold occurrences that occur downstream from the
more recently discovered lodes. Like Brewery Creek, Scheelite
Dome also occurs in the hanging wall of the Robert Service
thrust on the northern side of the Tintina fault, and is hosted in
the rocks of the Neoproterozoic-Cambrian Hyland Group
(Murphy, 1997). Scheelite Dome is underlain by a composite
quartz-monzonite stock, and the deposit was named after
tungsten-bearing skarns that are hosted by calcic
metasedimentary rocks on the northern side of the stock (Mair et
al, 1999; ODea et al, 2000). There are multiple styles of
mineralisation at Scheelite Dome, including sheeted gold- and
tungsten-rich quartz-feldspar veins within the stock, and a variety
of reduced gold- and tungsten-rich skarns, breccias, and veins in
the surrounding hornfels. Auriferous quartz veins within both the
stock and hornfels are hosted in E-trending tensional corridors
and NW-trending transpressive faults, (Mair et al, 1999;
Stephens et al, 2004). The mineralisation has been dated at
92 Ma (Murphy, 1997). The dominant sulfide minerals
associated with gold include arsenopyrite, pyrite, and, pyrrhotite,
with native bismuth and telluride minerals also common. Gold
predominantly occurs as free gold in both skarns and veins. The
resource is estimated to be 11.4 Mt at roughly 2.4 g/t (Hart et al,
2002) (Table 1).
PACRIM 2004
PACRIM 2004

TABLE 1
District, deposit type, host rock, resource data, production and elemental associations for Tintina Gold Province deposits.

Premining Production
Resource
District and Deposit Terrane Host Lithology Deposit Type Resource Grade Total Placer Lode Elemental Reference
deposit Age (Ma) (Mt) Association
TOMBSTONE BELT
Brewery Creek 91.40.2 Pz Selwyn Basin, syenite pluton, boitite intrusion-hosted sheeted 13.3 1.44 g/t 0.85 Au-As-Sb Hart et al, 2002; Diment and
Yukon Tanana Terrane monzonite porphyry veins, stockwork, Craig, 1999; Lindsay et al, 1999
sills, Pz Selwyn Basin disseminated, oxide and
Shale sulfide
Clear Creek 91.4-91.5 Pz Selwyn Basin, pluton intrusion-hosted sheeted 40 >300 ppb Au-As-Bi Hart et al, 2002; Marsh et al,

THE TINTINA GOLD PROVINCE, ALASKA AND YUKON: NEW WORLD-CLASS GOLD RESOURCES
Yukon Tanana Terrane veins 1999
Dublin Gulch 93 Pz Selwyn Basin, pluton intrusion-hosted sheeted 99 1.1 g/t 4.1 Au-Bi Hart et al, 2002; Maloof et al,
Yukon Tanana Terrane veins 2001
Scheelite Dome 92 Pz Selwyn Basin, Upper Pz Hyland replacements, stockwork, 11.37 2.4 g/t Au-As Mair et al, 1999; ODea et al,
Yukon Tanana Terrane Group, Tombstone structural control and 2000, Hart et al, 2002
Adelaide, SA, 19 - 22 September 2004

strain zone skarns

Placer 14.3 3.16 Hart et al, 2002
GOODPASTER
Pogo 107 Yukon Tanana Terrane mafic gniess quartz veins, sheeted veins, 9 18.86 5.8 Au-Ag-Bi-As Hart et al, 2002, Smith et al, 2000
shear hosted quartz veins
and quartz stibnite breccias
Richardson Placer 0.1 Hart et al, 2002
FAIRBANKS
Cleary Summit 104 Yukon Tanana Terrane Fairbanks schist stockwork, sheeted veins 1.4 34 1.59 0.301 Au-As-Bi Hart et al, 2002
Ft. Knox 92.4 Yukon Tanana Terrane biotite granite pluton stock work, sheeted veins 169 0.93 5.4 1.8 Au-Bi-Te-W Bakke et al, 2000; Hart et al,
2002
Ryan Lode 90 Yukon Tanana Terrane Fairbanks schist, shear hosted, stockwork 3.6 3.04 0.347 0.03 Au-As-Sb Hart et al, 2002
biotite tonalite pluton
True North 90/102 Yukon Tanana Terrane Chatinika eclogite veins, disseminated sulfides 14.6 1.69 0.79 Au-As-Sb Hart et al, 2002
Placer Yukon Tanana Terrane 8.1 Bundtzen et al, 1996
KUSKOKWIM
Donlin Creek 74.4-70.3 Kuskokwim Basin Kuskokwim basin stockwork, disseminated 122.1 2.91 12.3 Au-As-Sb-Hg Ebert et al, 2000; Goldfarb et al,
flysch sulfide 2004; Hart et al, 2002
Chicken Mtn, 63.4 Kuskokwim Basin pluton stockwork, vein fault 14.5 Au-Ag-As-Sb-Cu Bundtzen et al, 1992; Hart et al,
Golden Horn 2002
Nixon Fork 69 Kuskokwim Basin Ordivician limestone Au-skarn 0.1 42 0.17 0.183 Au-Cu-Bi Bundtzen and Miller, 1997
and dolomite
Vinasale 69 Kuskokwim Basin kuskokwim basin disseminated sulfide, 10.3 2.4 0.8 Au-As-Ag-Sb-Hg DiMiarchi, 1993; Miller and
flysch, quartz breccia, dolomite vein Bundtzen, 1997; Hart et al, 2002
monzonite pluton
Shotgun 67.3 Kuskokwim Basin granite porphyry, stockwork, porphyry 32.8 0.93 1.1 Rombach and Newberry, 2001
Shotgun Hills pluton
191

Placer 3.16 Bundtzen and Miller, 1997

S H MUELLER et al
Clear Creek
Located 120 km southeast of Dawson City, the Clear Creek
prospects are similar in many respects to the Scheelite Dome
occurrences (Figure 1), and considerable placer gold has been
extracted from the creeks draining this region. The ca. 92 Ma
Clear Creek mineralisation is hosted in the metasedimentary
rocks of the Neoproterozoic-Cambrian Hyland Group and in
some of the spatially and temporally associated six
mid-Cretaceous stocks. The igneous bodies range in composition
from diorite through equigranular granodiorite, to porphyritic
quartz monzonite (Murphy, 1997; Marsh et al, 1999). The main
prospects consists of sheeted E- to ENE-trending
quartz-K-feldspar extensional veins, with low sulfide contents
(<1 per cent), which cut five of the six stocks (Marsh et al,
2003). The gold is found as free gold and as refractory phases in
arsenopyrite, bismuthinite, pyrite, and pyrrhotite. In addition,
disseminated gold is found in the hornfelsed country rocks.
Factor analysis by Marsh et al, (1999) determined that the
deposit has an As-Au-BiSb-Te signature. Based on recent
exploration, the resource is estimated to be 40 Mt at >300 ppb
Au (Hart et al, 2002) (Table 1).
Dublin Gulch
The Dublin Gulch property is located 70 km north of the town of
Mayo in central Yukon (Figure 1). Much like the previous two
deposits, Scheelite Dome and Clear Creek, Dublin Gulch is also
associated with a ca. 92 Ma pluton that intrudes Neoproterozoic
to Cambrian Hyland Group metasedimentary rocks (Brown,
Baker and Stephens, 2001), and is also the source for substantial
amounts of placer gold. There are multiple types of
mineralisation at the Dublin Gulch deposit, including subparallel
sheeted quartz vein arrays within the intrusion and contact
aureole, breccias within the more distal metasedimentary rocks,
and tungsten and tin skarn deposits adjacent to the pluton
(Brown, Baker and Stephens, 2001; Maloof, Baker and
Thompson, 2001). The intrusion-hosted sheeted veins, as in the
case of Clear Creek, host the bulk of the resource. Furthermore,
the auriferous veins within and outside the stock also trend E to
ENE, suggesting a low magnitude, broadly E-W-directed
shortening and N-S extension during the emplacement of all the
92 Ma intrusions (Stephens et al, 2004). At Dublin Gulch, gold is
most commonly associated with bismuthinite and lead- and
bismuth-bearing sulfosalts in the sheeted veins within the pluton,
and pyrite, arsenopyrite, and pyrrhotite in the sheeted veins
within the metasedimentary rocks (Brown, Baker and Stephens,
2001; Maloof, Baker and Thompson, 2001). The deposit
metallogeny shows a strong Au-Bi correlation. The resource is
estimated to be 99 Mt at a grade of 1.1 g/t Au (Hart et al, 2002)
(Table 1).
DEPOSITS OF YUKON-TANANA TERRANE
The Fairbanks district, located in the interior of east-central
Alaska, is host to the most economically important gold deposits
within the TGP. The district was discovered more than one
hundred years ago and subsequent historical placer production
exceeds 8 Moz Au. Large-scale exploration and development of
the known lodes did not begin until the 1980s, when low-grade,
high-tonnage bulk minable systems became economically viable
targets. The country rocks of the Fairbanks district are primarily
composed of three medium- to high-grade metamorphic
assemblages of the Yukon-Tanana terrane: the lower amphibolite
facies Fairbanks schist and Muskox sequence, the greenschist
facies Birch Hill sequence, and the eclogite facies rocks of the
Chatanika terrane. These units range in age from Neoproterozoic
(Fairbanks Schist) to Paleozoic. Numerous ca 90 Ma felsic to
intermediate intrusions, emplaced throughout the district, may be
genetically linked to the gold deposits (Figure 1). The
192 Adelaide, SA, 19 - 22 September 2004
Goodpaster district, also within the Yukon-Tanana terrane and
about 150 km to the southeast of Fairbanks, is underlain by
sillimanite-grade, Neoproterozoic to mid-Paleozoic paragneiss,
and Mississippian and Early Cretaceous felsic orthogneiss
(Smith et al, 2000).
Pogo
The Pogo deposit is located 145 km east of Fairbanks and within
the Goodpaster mining district of eastern Alaska (Figure 1). The
Pogo deposit occurs within a 3-km-wide, 15-km-long,
NW-trending zone, which is located a few kilometers south of
the ~93 Ma Goodpaster batholith (Smith et al, 2000). The trend
includes thick, shallowly-dipping auriferous quartz shear veins,
stockwork and sheeted vein zones; steeply-dipping, shear-hosted
veins; and quartz-stibnite veins and breccias (Smith et al, 2000).
Of these, the Liese zone, which comprises two or more
shallowly-dipping, shear-hosted quartz veins averaging 7 m in
thickness, contains most of the ore, with an estimated resource of
8.69 t at 18.86 g/t (Smith et al, 2000) (Table 1). Common
metallic minerals within the ore are pyrite, arsenopyrite,
chalcopyrite, native bismuth, bismuthinite, native gold, galena,
and sphalerite. Gold shows a strong correlation with bismuth.
The deposit was obvious in the 1976 US Geological Survey
regional stream sediment survey data from the creek draining the
Liese zone (anomalous Au, As, and W), but it was another five
years until company survey work located mineralised outcrop
(Hart et al, 2002). The Liese zone is partially hosted in a series
of 107 Ma granitic dikes, and hydrothermal alteration associated
with the gold deposit is cut by a 94.5 Ma (U-Pb) diorite dike
(Smith et al, 2000). The 40Ar/39Ar cooling ages on hydrothermal
biotite and sericite are typically ~91 Ma (Smith et al, 2000), but
recent Re/Os dates on molybdenite from quartz veins are
~104 Ma (Selby et al, 2002). It is still uncertain as to whether the
mineralisation is related to the 107 Ma Goodpaster batholith, but
represents protracted hydrothermal activity; a younger intrusion;
or that it is not intrusive-related (Selby et al, 2002).
Cleary Summit
The Cleary Summit area, located 30 km northeast of the city of
Fairbanks, contains about thirty, small, high-grade (~10 g/t Au)
(Table 1), historical lode mining operations that were active in
the Fairbanks district prior to ~1960. These include Hi-Yu,
Cleary Hill, McCarty, Tolvana, Christina, and Chatham (Metz,
1991; McCoy et al, 1997). The mines of the Cleary Summit area
are all hosted in Proterozoic Fairbanks schist, and are best
described as shear-hosted orebodies, with veins occurring as
open-space fractures dominated by massive white and ribbon
quartz, with variable sulfide content (McCoy et al, 1997).
Gold-related phases include arsenopyrite, stibnite, and
jamesonite (Metz, 1991). The vein systems continue along a
NW-trending strike for about 8 km. The bulk of these deposits
were discovered by placer miners, who previously had mined the
lower elevations of creeks of the Cleary Summit area.
Fort Knox
The Fort Knox mine, located 22 km northeast of Fairbanks, is a
current gold producer and the largest operating gold mine in
Alaska. Gold ore was known at what is presently the Fort Knox
deposit for more than one hundred years. Placer gold was
discovered in Fish Creek in 1901, downstream from what is now
the Fort Knox mine. In 1913, gold-bearing quartz veins were
discovered over what is now the Fort Knox deposit, and bismuth-
and tellurium-bearing mineral phases were reported in the
auriferous veins cutting porphyritic biotite granite (Bakke et al,
2000). Between 1980 and 1995, soil surveys, drilling, and
trenching defined the Fort Knox deposit, with production
beginning in 1996.
PACRIM 2004
 THE TINTINA GOLD PROVINCE, ALASKA AND YUKON: NEW WORLD-CLASS GOLD RESOURCES
The deposit is hosted entirely within a 92 Ma composite,
high-K granodiorite and biotite granite pluton, which intrudes the
Fairbanks schist (Bakke et al, 2000). The ore occurs in early
pegmatitic and steeply-dipping, sheeted-stockwork quartz veins,
and younger W-NW-trending shear veins, much like the sheeted
veins within the deposits in Yukon (McCoy et al, 1997,
Thompson and Newberry, 2000). The ore at Fort Knox has a
relatively low sulfide content (<1 per cent by volume) (Bakke et
al, 2000). Gold is strongly associated with bismuth and
tellurium. The resource is 169 Mt at 0.93 g/t for a total of 5.4
Moz Au. To date, Ft. Knox has produced 1.8 Moz Au (Hart et al,
2002) (Table 1).
True North
True North is located 8 km to the north of Fort Knox and is being
mined for higher-grade ore to blend with the Fort Knox ores. The
True North deposit is hosted within the eclogitic Chatanika
terrane, and has characteristics of an epizonal gold deposit
(Bakke et al, 2000). The ore zones are shallowly-dipping and
variably brecciated, and they appear to be spatially related to
regional thrust faults. Fine-grained gold occurs with pyrite and
arsenopyrite in drusy quartz veins, and in altered and brecciated
schist adjacent to the quartz veins (Bakke et al, 2000). In the
oxidised zone, the pyrite and arsenopyrite have weathered to
geothite and scorodite, respectively. The resource at True North
has been calculated at 16.5 Mt averaging 2.46 g/t, for a total of
1.2 Moz Au (Hart et al, 2002) (Table 1).
Ryan Lode
The Ryan Lode (Ryan Lode and Curlew deposits) property is
located 13 km to the west of Fairbanks and approximately 53 km
west of the Ft. Knox deposit (McCoy et al, 1997). The host rocks
are mica and quartz-mica schist units of the Fairbanks schist
(McCoy et al, 1997). The Ryan Lode deposit consists of
NE-striking, sub-parallel, shear zones. More local faults cut the
main shear zone and enhance both the grade and size of the
orebodies (Bakke et al, 2000). Gold is strongly associated with
arsenic and antimony. The ore is oxide and mixed oxide-sulfide to
~90 m below the surface. Below the mixed zone is a poorly
defined sulfide-rich resource. The Curlew deposit is 500 m to the
south of the main Ryan Lode orebody. The deposit consists of
NE-trending shear zones and breccias hosted within a medium- to
coarse-grain granodiorite. The current resource estimate is 3.6 Mt
at 3 g/t for a total of 0.35 Moz Au (Hart et al, 2002) (Table 1).
DEPOSITS OF SOUTHWESTERN ALASKA
The deposits of the Kuskokwim basin in southwestern Alaska
define the western side of the TGP. The Kuskokwim region is
predominately underlain by rocks of the Upper Cretaceous
Kuskokwim Group (Cady et al, 1955). They were deposited in a
NE-trending, strike-slip basin that subsided between a series of
amalgamated terranes (Miller and Bundtzen, 1994). The older
basement terranes, which were amalgamated by mid-Cretaceous
(Patton et al, 1994; Miller and Bundtzen, 1994), include
Paleozoic-Mesozoic oceanic crust and subduction zone
assemblages, Late Proterozoic to Paleozoic continental and
continental margin rocks, and fragments of Early Proterozoic
continental basement (Decker et al, 1994; Miller and Bundtzen,
1994; Patton et al, 1994). The waning stages of Kuskokwim
Group deposition were accompanied by regional volcanism and
intrusion, including formation of a number of widely distributed
volcanic-plutonic complexes of Late Cretaceous to early Tertiary
age that overlie and intrude the sedimentary rocks of the
Kuskokwim basin. The region contains the western extension of
the Tintina (Iditarod-Nixon Fork faults system) and
Denali-Farewell (Holitna segment) fault systems (Goldfarb et al,
2004) (Figure 1).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Donlin Creek
The Donlin Creek deposit is located 450 km northwest of
Anchorage, in the Kuskokwim Mountains of southwestern
Alaska, approximately 20 km north of the Kuskokwim River.
The ~70 Ma epizonal deposit is hosted within and adjacent to a
NE-trending, 8-km-long, hypabyssal dike-sill complex that
intrudes the Kuskokwim Group sedimentary rocks (Miller and
Bundtzen, 1994; Bundtzen and Miller, 1997; Ebert et al, 2000;
Goldfarb et al, 2004). The gold at Donlin Creek occurs in N- to
NE-striking, steeply-dipping veins and veinlets that fill brittle
extensional fractures in the rhyolite-rhydacite porphyry dikes and
sills (Gray, Gent and Snee, 1997; Szumigala, Dodd and Arribas,
2000). The veins are dominated by gray and clear quartz with as
much as 3-5 per cent sulfides. The gold is refractory in nature
and is found in arsenopyrite, but rarely in pyrite (Szumigala,
Dodd and Arribas, 2000; Goldfarb et al, 2004). Late-stage
stibnite, orpiment, realgar, cinnabar, and native arsenic are also
present, but in smaller quantities (Goldfarb et al, 2004). Current
resource estimates indicate approximately 25 Moz Au at a
cut-off grade of 1.5 g/t (Table 1).
Shotgun
Located 450 km west-southwest of Anchorage, the Shotgun
deposit lies on the southern periphery of the Kuskokwim basin in
the Shotgun Hills region of southwestern Alaska. The shales of
the Kuskokwim Group have been intruded by a 67 Ma granite
prophyry pluton that hosts the deposit (Rombach and Newberry,
2001). The deposit consists of gold-arsenic mineralisation within
quartz stockwork and breccias. The ore typically has <1 per cent
sulfides by volume, with arsenopyrite>pyrite>pyrrhotite>
chalcopyrite. Gold occurs both as inclusions within arsenopyrite,
as well as fine-grained native gold within the quartz veins. The
resource is estimated at 32 Mt at a cut-off grade of 0.93 for a
total of 1.1 Moz Au (Table 1).
Chicken Mountain/Golden Horn
The Chicken Mountain and Golden Horn deposits are located
approximately 30 km northwest of the Donlin Creek deposit.
These gold and polymetallic stockworks and breccias are hosted
within the ca. 70 Ma Chicken Mountain and Black Creek
plutons. The former pluton consists of alkali gabbro,
monzodiorite, and wherlite, which are intruded by inner phases
of quartz monzonite and syentite (Bundtzen et al, 1992); similar
alkaline to subalkine composite systems characterise the Black
Creek intrusive complex (Bull, 1988). The vein ore assemblage
consists of arsenopyrite, pyrite, stibnite, cinnabar, scheelite,
chalcopyrite, molybdenite, and sulfosalts. The sulfide content of
the veins is typically <5 per cent by volume. The gold in both
deposits occurs as fine-grain native gold and as inclusions in
arsenopyrite (Bundtzen et al, 1992; Bundtzen and Miller, 1997).
The resource is estimated to be 14.5 Mt at 1.2 g/t Au (Bundtzen
and Miller, 1997) (Table 1).
Vinasale
Vinasale is a poorly-studied deposit that is located 100 km to the
northeast of Donlin Creek in the Kuskokwim Mountains. The 69
Ma mineralisation occurs as disseminations, breccias, dolomite
veins, and segregations within a composite monzonite, quartz
monzonite, and granite porphyry pluton (DiMarchi, 1993;
Bundtzen and Miller, 1997). The gold is contained within sulfides
and sulfosalts of arsenic and antimony (Bundtzen and Miller,
1997). The resource is estimated to be 10.3 Mt at a 2.9 g/t grade
cut-off for a total of 0.8 Moz Au (Hart et al, 2002) (Table 1).
193
S H MUELLER et al

Nixon Fork The material in Table 2 summarises all publicly available

In contrast to the vein and stockwork styles of mineralisation in
the Kuskokwim Group flysch, the Nixon Fork deposit is a
gold-rich skarn deposit in adjacent carbonate platform rocks. The
Nixon Fork deposit is located 100 km northeast of Donlin Creek
in southwestern Alaska. It is hosted in Ordovician limestone and
dolomite of the Farewell terrane. The deposit is immediately
adjacent to a quartz monzodiorite pluton that is cut by
gold-bearing, quartz and sericitically altered dikes, which appear
to have concentrated the gold in the adjacent skarns (Newberry et
al, 1997). The resource is estimated to be 0.1 Moz at a grade of
4.2 g/t for a total of 0.17 Moz Au (Table 1).
ENVIRONMENTAL SIGNATURES
Understanding the environmental signatures of the gold deposits
of the TGP is critical for several reasons:
1. the environmental signature of unmined deposits provides
exploration companies and land agencies with information
on the baseline concentrations of trace metals in soil and
water at a deposit prior to disturbance, which can then be
used in planning realistic cleanup goals during mine
closure;
2. understanding the behavior of trace elements associated
with the natural weathering and erosion of mineral deposits
can provide valuable information for exploration for similar
type deposits; and
3. understanding the mobility of trace metals at undeveloped,
developed and abandoned mine sites will help in the
remediation of historical mines, and provide information
that will be valuable during the life of the operating mine.
ground and surface water chemistry for the major gold deposits
of the TGP. Regardless of lithology, absence or presence of
permafrost, and style of mineralisation, all deposit-related
ground and surface waters across the TGP are slightly acidic to
slightly basic (ie pH = 5 - 9) and Ca-HCO3--dominated (Mueller
et al, 2002a,b; Verplanck et al, 2003; Mueller et al, 2003a,b)
(Table 2). Acid rock-drainage is generally not an issue at
deposits within the TGP because the metasedimentary rocks
appear to have a high acid neutralising capacity due to their
relatively elevated carbonate content (Mueller et al, 2003b),
which is typically a function of the hydrothermal alteration
associated with the orogenic and intrusion-related gold deposits.
The conductivity measurements of both ground and surface
water throughout the TGP vary greatly (surface water =
185 - 1320 S/cm, ground water = 320 - 2940 S/cm), and
generally elevated conductivity correlates with increased cation
and anion (eg Ca, Mg Fe, As, Mn, Sb, HCO3- and SO42-)
concentrations that may result from interaction with orebodies.
Based upon the element associations in Table 1 and the
summary water chemistry data for five of the deposits in Table 2,
the most characteristic environmental signature of both mined
and unmined gold deposits within the TGP are the enrichment of
arsenic and antimony in ground and surface water caused by the
oxidation of arsenopyrite and stibnite associated with the ore.
Typically the epizonal and shear-hosted deposits within the
metasedimentary and sedimentary units have a higher sulfide (>2
per cent by volume; usually arsenopyrite, pyrite, and stibnite)
content, and, therefore, have higher concentrations of arsenic and
antimony in both ground (As from 20 - 860 g/L and Sb <2 - 40
g/L) and surface water (As from <2 - 810 g/L and Sb <2 - 250
g/L) (Goldfarb et al, 1995; Mueller et al, 2002a, b; Verplanck et
al, 2003; Mueller et al, 2003a, b). A comparison of the

TABLE 2
Summary statistics of selected parameters and elements for filtered and unfiltered ground and surface water from throughout the
Tintina Gold Province.

Surface water (filtered) Surface water (unfiltered) Groundwater (filtered) Groundwater (unfiltered)
Parameter 25% Med 75% Max n 25% Med 75% Max n 25% Med 75% Max n 25% Med 75% Max n
pH 7.4 7.6 7.8 8.7 96 7.1 7.4 7.5 8.0 23
Cond 185 237 324 1323 96 320 506 830 2940 23
(S/cm)
(mg/L)
AlkHCO3 95 118 148 243 96 174 280 402 709 23
SO4 11 19 39 580 90 30 82 160 1800 23
Ca 23 29 40 168 96 23 29 41 180 96 36 62 117 541 23 37 73 120 558 23
Mg 6.6 11 15 94 96 6.6 11 15 101 96 13 26 47 200 23 14 29 48 214 23
Na 1.2 1.6 2.2 5.3 96 1.2 1.6 2.2 5.3 96 2.3 4.3 8.3 220 23 2.5 4.8 8.4 220 23
K 0.3 0.4 0.4 5.4 95 0.3 0.4 0.4 5.9 95 1.8 2.8 5.9 15 23 2.0 3.2 7.2 18 23
Fe <2 <2 0.2 17 53 0.0 0.0 0.4 700 55 0.0 0.1 0.6 8.4 17 0.2 1.4 4.9 42 18
(g/L)
Al 4.4 9.38 18 224 49 48 143.5 394 15300 86 1.7 2.5 5.3 3290 23 5.9 22 754 18500 23
As 3 6.55 12 250 44 2.5 8.2 22 810 47 24 98 198 845 23 29 144 400 862 23
Cr 1.4 1.95 4.4 16 88 1.2 2.1 3.4 20 74 1.6 2.5 6.9 12 17 2 5.7 9.9 59 14
Cu 0.6225 0.775 1.3 3.1 54 0.6 0.815 2.025 21 58 0.81 1.3 2.5 7 16 0.52 1.2 6.7 25 23
Mn 3.85 6.2 18 668 95 3.1 13 31.4 2840 95 23 199 470 1800 23 50 275 623 2230 23
Mo 0.315 0.71 1.2 3.3 28 0.43 0.79 1.1 3.1 31 0.71 1.4 4.9 15 22 0.75 1.4 3.2 15 23
Ni 0.6 1.35 1.725 12 92 1 1.4 2.05 41 91 1.7 2.8 9.8 48 23 1.9 4.3 11 68 23
Pb 0.06 0.095 0.2 11 36 0.1 0.2 0.70 31 67 0.06 0.08 0.1 3.9 13 0.13 0.25 1.2 120 22
Sb 0.3325 0.475 2.7 200 66 0.39 2.1 3.9 250 43 0.48 2.2 5.1 42 22 0.53 0.83 4.4 29 23
Se 1.4 1.8 2 6.8 26 1.2 1.7 2 4.4 34 1.2 1.5 4.2 10 6 3.1 5 8.1 11 3
Zn 2.075 5.25 6.55 1500 80 0.7 1.3 2.8 2200 81 2.7 5.4 11 94 23 3.5 5.6 19 120 23

194 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE TINTINA GOLD PROVINCE, ALASKA AND YUKON: NEW WORLD-CLASS GOLD RESOURCES
hydrogeochemical signatures of undeveloped (Donlin Creek),
active (True North) and partially reclaimed (Brewery Creek)
epizonal deposits shows that low levels of arsenic and antimony
are typical of surface waters draining the deposits regardless of
their state of development (Mueller et al, 2003b). This is also
likely the case for deposits with more deeply formed styles of
mineralisation, which also contain significant volumes of
arsenopyrite and stibnite (eg Scheelite Dome, Dublin Gulch,
Cleary Summit and Golden Horn). For example, waters draining
the Christina adit in the Cleary Summit area had a pH of 7.6,
conductivity of 324 S/cm and dissolved arsenic and antimony
concentrations of 84 g/L and 200 g/L and total arsenic and
antimony concentrations of 810 g/L and 250 g/L, respectively
(Goldfarb et al, 1995). The essentially totally intrusion-hosted
Fort Knox deposit, with an order of magnitude lower volume of
sulfide minerals, does not have notable metal enrichments in
surface water (As<10 g/L, Sb<2 g/L ) or in ground water (As<
40 g/L, Sb<2 g/L) downstream from the mined area.
The presence of elevated concentrations of arsenic and
antimony in ground and surface water above regulatory levels (ie
As<3 g/L and Sb<6 g/L, U.S. Environmental Protection
Agency [EPA]) is a concern surrounding many deposits in the
TGP due to the associated health effects of arsenic and antimony
on human and aquatic life. The documentation of elevated
concentrations of trace elements of concern (TEC) prior to
disturbance, as well as knowledge of the beneficiation process,
have resulted in projects at Fort Knox, True North, Brewery
Creek, and Donlin Creek opting for zero discharge facilities to
avoid future liability and extensive cleanup costs (Hart et al,
2002; Nova Gold Resources, 2002). Furthermore, documentation
of naturally-elevated concentrations of TEC distal from known
mining operations can benefit mining operations during closure,
as was the case at Ryan Lode, where FGMI and Kinross were
able to demonstrate that arsenic concentrations were naturally
elevated at the same or higher concentration levels widely
throughout the Fairbanks region, resulting in a more reasonable
cleanup target level than the <3 ppb as as suggested by the EPA.
Environmental signatures of soils in the TGP are also
important for remediation planning. The natural degradation of
the upper levels of deposits throughout the TGP has resulted in
broad, often regionally extensive, As, Sb, Hg, and Au soil
anomalies, which are obviously the subsequent cause of many
stream sediment anomalies (Table 1). Brewery Creek was
discovered by a stream silt geochemical survey that was
characterised by anomalous mercury concentrations (>200 ppb
Hg). Reanalysis of the anomalous silts for additional metals
showed that they contained 38 ppb Au, 27 ppm As, and 16 ppm
Sb, effectively targeting the deposit (Hart et al, 2002). While
other deposits, such as Ft Knox and Donlin Creek, were
discovered in the upstream reaches of placer mines, geochemical
anomalies from soil sampling were nevertheless used to delineate
target areas prior to drilling (Hart et al, 2002).
SUSTAINABLE DEVELOPMENT
The sustainable development of gold deposits within the TGP is
dependent on a number of intercorrelated factors, such as climate,
politics, environmental concerns, governmental permitting
policies, access, infrastructure, and beneficiation costs. All of the
deposits within the TGP lie between 60 and 66N, with
temperatures ranging between -40C in the winter to 30C during
the summer. Much of Alaska and Yukon is unspoiled sub-arctic
wilderness and tundra, which makes mining activities politically
sensitive issues. With arsenic and antimony as common
high-concentration trace elements, particularly in ground water,
the use of cyanide in a number of milling operations, and growing
public concern over the detriment mining can cause to the
environment, almost all present mining projects have chosen zero
discharge (Table 3). Furthermore, the 13 Alaskan Native
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Corporations, which were formed by the 1971 Alaskan Native
Claims Settlement Act (ANCSA), have the largest public land
holdings in the state of Alaska (both surface and subsurface), and,
therefore, their use and management of the land for subsistence
and economy must be considered.
The location of the deposits will play a key role in the
sustainable development of the gold resources in the TGP. The
two largest operations in the TGP, Brewery Creek and Fort
Knox/True North, have benefited from having work forces that
can be housed in nearby towns (Dawson City and Fairbanks,
respectively). Development of these deposits required relatively
short spur roads, and additions of short power lines to support
mine and mill operations. However, development of more remote
deposits, such as Vinasale, Shotgun, Donlin Creek, and Dublin
Gulch, would require construction of roads and transmission
lines that could be hundreds of kilometers long. On-site diesel
power generation would not be economic for deposits where the
ores have a high work index. Remote areas are often expensive
and difficult to access and generally less well explored.
Furthermore, mines in remote regions, such as the Kuskokwim
basin, would require transportation of supplies along seasonally
open rivers, which would increase both operating cost and risk
(Table 3).
Both capital and operating costs are substantially affected by
the beneficiation method. The ore from deposits such as Ft Knox,
Dublin Gulch, Pogo, and Shotgun are amenable to cyanide
leaching, regardless of surface oxidation. Other deposits with
refractory gold, where surface oxidation is relatively deep, may
be economically viable because the oxidised rocks have a low
work index; for example, at Brewery Creek, run-of-mine ores
were taken to the leach pad without crushing. However, the
beneficiation of sulfide and mixed oxide-sulfide ores at deposits
such as Donlin Creek require a high work index, making the
recovery of gold at these deposits more expensive and more
complex (Table 3).
In general, support for exploration and development of gold
resources in the TGP by the Alaskan and Yukon governments has
been positive, particularly when compared to attitudes in other
states within the United States and provinces in Canada. In the
Yukon, exploration benefits from government-funded exploration
programs, exploration tax credits, and federal tax credits. In
Alaska, exploration expenditures can be written off against future
taxes on gold production, state-secured loans can assist and
provide new infrastructure, and regulatory and taxation policy is
more positive than in other places within the United States (Table
3).
SUMMARY
The TGP is defined by multiple gold deposit styles, which are
spatially and temporally associated with mid- and Late
Cretaceous intrusions, hosted in metasedimentary rocks across
the interior of Alaska and Yukon, in the North American
Cordillera. Together, they account for a lode resource of 37 Moz
Au and are likely they source of an additional 30 Moz of placer
gold that has been historically recovered. The deposit styles can
best be categorised as:
1. sheeted veins in both the apical portion of ca. 90 Ma
plutons and in adjacent wallrock;
2. replacements, disseminations, and breccias located either
above or adjacent to Cretaceous plutons within the
hornfelsed aureoles;
3. fault-related ores that range from quartz bodies in ductile
shear zones, to brittle, gouge-filled, sulfide-rich fractures,
to disseminated ores along or adjacent to shear zones
these all may occur proximal or at significant distances
from Cretaceous plutons;
195
S H MUELLER et al

TABLE 3
Summary of key points for sustainable development throughout the Tintina Gold Province.

Sustainability summary
Ore characteristics Access/infrastructure Land and environment
Tombstone Belt
Brewery Creek Oxide w/undetermined sulfide Highway 60 km from Dawson City, short spur As,Sb, Hg signature in soil and water,
resource, oxide amenable to heap road, established power transmission lines Brewery Creek was a heap leach facility,
cyanide leach zero discharge and bioleach facility
Clear Creek Free Au, oxide/sulfide mix, cyanide Highway access from Dawson City, network As, Bi in soil
leach? of exploration roads, no established power
transmission lines
Dublin Gulch Free Au, oxide/sulfide mix, cyanide Highway from Mayo, spur roads, no Au, Bi, W in soils
leach established power transmission lines
Scheelite Dome Free Au, oxide/sulfide mix Highway from Mayo, spur roads, no Au, As, in soils
established power transmission lines
Goodpastor
Pogo Free Au, oxide/sulfide mix 50 km spur road being built from Delta Au, Ag, Bi, As in soils
Junction, power transmission line being
installed, facilites under construction
Fairbanks District
Cleary Summit Free Au in quartz veins Highway access 30 km from Fairbanks, old Au, Bi, Te, W
spur roads
Ft. Knox Free Au in quartz veins, crush float Highway access 30 km from Fairbanks, well Au, Bi, W, Te, As in soils, little or no
and cyanide leach maintained spur road, established power As, in waters, zero discharge facility,
transmission lines, modern mill and mine
facilites
Ryan Lode Mixed oxide/sulfide, historical cyanide Highway access 12 km from Fairbanks, 5 km As, Sb in soils, surface water and
heap leach, undetermined sulfide dirt spur road, established transmission lines ground water, located within 3 km of
resource residential housing, would need to be
zero discharge
True North Oxide, crush, flotation, cyanide leach, Highway access 30 km from Fairbanks, well As, Sb in soils, surface water and
undetermined sulfide resource maintained spur road, established power ground water, zero discharge
transmission lines, modern mill and mine
facilites
Kuskokwim
Donlin Creek Sulfide, floatation, cyanide, oxidation Requires 17 km access road from town of As, Sb, Au, Hg in soils, As, Sb in waters
Crooked Creek, access to Crooked Creek by and ground water, surface and
barge on Kuskokwim River, established air subsurface rights owned by Calista
strip and camp Native Corp, zero discharge facility,
local community
Chicken Mountain, Free Au, mixed oxide/sulfide Access by winter trail and air only, no Au, As, Sb in soils
Golden Horn transmission lines
Nixon Fork Oxide Access by air Au, W, in soils
Vinasale Sulfide Access by air, barge via Kuskokwim river, no Au, Sb, Bi in soils, on lands owned by
transmission lines, no established facilites Doyon Ltd Native Corp
Shotgun Free Au, inclusions in sulfide, cyanide Access by air only, no transmission lines Au, As, Cu, in veins, on lands owned by
leach Calista Native Corp

4. epizonal stockwork mineralisation and related breccias; and REFERENCES

5. gold-bearing skarns.
The deposits of the TGP have environmental signatures
characterised by anomalous arsenic and antimony concentrations
in both ground and surface water, as well as anomalous
concentrations of As, Sb, Hg, Bi, and W in stream sediments and
silts, and in soils. These have been used with success, to explore
for new deposits within the TGP, as well as to help plan required,
post-mining clean-up levels. The sustainable development of
current and future mineral deposits within the TGP is dependent
upon positive attitudes in the community and government,
proximity to pre-existing infrastructure or the support to create
new infrastructure, and the responsible development of mineral
resources in an environmentally and culturally sensitive manner.
Bakke, A A, Morrel, B, Odden, J, Bergstrom, T and Woodman, J, 2000.
Kinross Gold USAs activities in the Fairbanks mining district, in
The Tintina Gold Belt: Concepts, Exploration, and Discoveries (Eds:
T L Tucker and M T Smith), pp 89-98 (British Columbia and Yukon
Chamber of Mines).
Brown, V S, Baker, T and Stephens, J R, 2001. Ray Gulch tungsten
skarn, Dublin Gulch, central Yukon: Gold-tungsten relationships in
intrusion-related ore systems and implications for gold exploration,
in Yukon Exploration and Geology 2001 (Eds: D S Emond, L H
Weston and L L Lewis), pp 259-268 (Exploration and Geological
Services Division, Yukon Region, Indian and Northern Affairs:
Canada).
Bull, K F, 1988. Genesis of the Golden Horn and related mineralisation in
the Flat area, Alaska, MS thesis, (unpublished), 149 p, University of
Alaska.

196 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE TINTINA GOLD PROVINCE, ALASKA AND YUKON: NEW WORLD-CLASS GOLD RESOURCES
Bundtzen, T K and Gilbert, W G, 1983. Outline of geology and mineral
resources of upper Kuskokwim region, Alaska, Geological Society
Journal, 3:101-117.
Bundtzen, T K and Miller, M L, 1997. Precious metals associated with
igneous rocks of southwestern Alaska, in Mineral Deposits of Alaska
(Eds: R J Goldfarb and L D Miller), Econ Geol, Monograph 9, pp
242-286.
Bundtzen, T K, Bouley, B A, Noyes, H J and Nokleberg, W J, 2000.
Regional metallogenesis of central Alaska, in Proceedings Society
for Mining, Metallurgy, and Exploration Annual Meeting, Salt Lake
City, Utah, 28 February - 1 March, pp 1-20.
Cady, W M, Wallace, R E, Hoare, J M and Webber, E J, 1955. The central
Kuskokwim region, Alaska, 132 p, US Geological Survey
Professional Paper 268.
Decker, J, Bergman, S C, Blodgett, R B, Box, S E, Bundtzen, T K,
Clough, J G, Coonrad, W L, Gilbert, W G, Miller, M L, Murphy, J
M, Robinson, M S and Wallace, W K, 1994. Geology of
southwestern Alaska, in Geology of Alaska (Eds: G Plafker and H C
Berg), Geological Society of America, The Geology of North
America, No G-1, pp 285-310.
DiMarchi, J J, 1993. Geology, alteration, and mineralization of the
Vinasale Mountain gold deposit, west central Alaska, in Alaska
Division of Geological and Geophysical Surveys Professional Report
113, pp 17-29.
Diment, R, 1996. Brewery Creek gold deposit, in Yukon Exploration and
Geology (Ed: C F Roots), pp 57-64 (Exploration and Geological
Services Division, Yukon, Indian and Northern Affairs: Canada).
Diment, R and Craig, S, 1999. Brewery Creek gold deposit, central
Yukon, in Yukon Exploration and Geology, 1998 (Eds: C F Roots and
D S Emond), pp 225-230 (Exploration and Geological Services
Division, Yukon, Indian and Northern Affairs: Canada).
Ebert, S, Miller, L, Petsel, S, Dodd, S and Kowalczyk, P, 2000. Geology
mineralization, and exploration at the Donlin Creek project,
southwestern Alaska, in The Tintina Gold Belt: Concepts,
Exploration, and Discoveries (Eds: T L Tucker and M T Smith), pp
89-98 (British Columbia and Yukon Chamber of Mines).
Foster, H L, Keith, T E C and Menzie, W D, 1994. Geology of the
Yukon-Tanana area of east-central Alaska, in The Geology of Alaska:
Geological Society of America, Geology of North America, (Eds: G
Plafker and H C Berg), v G-1, pp 205-240.
Goldfarb, R J, Taylor, C D, Meier, A L, dAngelo, W M and OLeary, R
M, 1995, Hydrogeochemistry of mine-drainage waters associated with
low-sulfide, gold-quartz veins in Alaska, in Geological Studies in
Alaska by the US Geological Survey, 1995, (Eds: J A Dumoulin and J
E Gray), US Geological Survey Professional Paper, 1574, pp 3-17.
Goldfarb, R J, Ayuso, R, Miller, M L, Ebert, S W, Marsh, E E, Petsel, S
A, Miller, L D, Bradley, D, Johnson, C and McClelland, W, 2004.
The Late Cretaceous Donlin Creek gold deposit, southwestern
Alaska: Controls on epizonal ore formation, Econ Geol, 99:642-671.
Gordey, S P and Anderson, R G, 1993. Evolution of the northern
Cordilleran miogeocline, Nahanni map area (1051), Yukon and
Northwest Territories, Geological Survey of Canada, Memoir 428.
Gray, J E, Gent, C A and Snee, L W, 1997. Epithermal mercury-antimony
and gold-bearing vein lodes of southwestern Alaska, in Mineral
Deposits of Alaska (Eds: R J Goldfarb and L D Miller), Econ Geol,
Monograph 9, pp 287-305.
Groves, D I, Goldfarb, R J, Robert, F and Hart, C J R, 2003. Gold
deposits in metamorphic belts: overview of current understanding,
outstanding problems, future research, and exploration significance,
Econ Geol, 98(1):1-29.
Hart, C J R, McCoy, D T, Goldfarb, R J, Smith, M, Roberts, P, Hulstein,
R, Bakke, A A and Bundtzen, T K, 2002. Geology, Exploration and
Discovery in the Tintina Gold Province, Alaska and Yukon, in
Integrated Methods for Discovery: Global Exploration in the
Twenty-First Century. Society of Economic Geologists, Special
Publication 9 (Eds: R J Goldfarb and R L Nielsen), pp 241-274.
Lindsay, M J, Baker, T, Oliver, N H S, Diment, R and Hart, C J R, 1999.
The magmatic and structural setting of the Brewery Creek gold mine,
central Yukon, in Yukon Exploration and Geology, 1999 (Eds: D S
Emond and L H Weston), pp 271-279 (Exploration and Geological
Services Division, Yukon Region, Indian and Northern Affairs:
Canada).
Mair, J L, Hart, C J R, Goldfarb, R J, ODea, M and Harris, S, 1999.
Geology and metallogenic signature of gold occurrences at Scheelite
Dome, Tombstone gold belt, Yukon, in Yukon Exploration and
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Geology, 1999 (Eds: D S Emond and L H Weston), pp 165-176
(Exploration and Geological Services Division, Yukon Region,
Indian and Northern Affairs: Canada).
Maloof, T L, Baker, T and Thompson, J F H, 2001. The Dublin Gulch
intrusion-hosted gold deposit, Tombstone plutonic suite, Yukon
Territory, Canada, Min Dep, 36(6):583-593.
Marsh, E E, Hart, C J R, Goldfarb, R J and Allen, T L, 1999. Geology
and geochemistry of the Clear Creek gold occurrences, Tombstone
gold belt, central Yukon Territory, in Yukon Exploration and Geology,
1999 (Eds: D S Emond and L H Weston), pp 185-196 (Exploration
and Geological Services Division, Yukon Region, Indian and
Northern Affairs: Canada).
Marsh, E E, Goldfarb, R J, Hart, C J R and Johnson, C A, 2003. Geology
and geochemistry of the Clear Creek gold occurrences, Tintina Gold
Province, Yukon, Canada, Can J of Earth Sci, 40(5):681-699.
McCoy, D T, Newberry, R J, Layer, P, DiMarchi, J J, Bakke, A A,
Masterman, J S and Minehane, D L, 1997. Plutonic-related gold
deposits of interior Alaska, in Mineral Deposits of Alaska (Eds: R J
Goldfarb and L D Miller), Econ Geol, Monograph 9, pp 191-241.
Metz, P A, 1991. Metallogeny of the Fairbanks Mining District, Alaska
and Adjacent Areas, 370 p (Mineral Industry Research Laboratory,
School of Mineral Engineering, University of Alaska: Fairbanks).
Miller, M L and Bundtzen, T K, 1994. Generalized geologic map of the
Iditarod quadrangle, Alaska, showing potassium-argon, major-oxide,
trace-element, fossil, paleocurrent, and archeological sample
localities, US Geological Survey Miscellaneous Field Studies Map
MF-2219A, scale 1:250 000, 48 p.
Mueller, S H, Goldfarb, R J, Farmer, G L, Sanzolone, R, Adams, M,
Theodorakos, P M, Richmond, S A and McCleskey, R B, 2002a.
Trace, minor and major element data for ground water near
Fairbanks, Alaska, 1999 - 2000, US Geological Survey Open-File
Report 02-00090, 12 p.
Mueller, S H, 2002b. A geochemical characterization of ground water near
Fairbanks, Alaska, with emphasis on arsenic hydrogeochemistry, MSc
thesis (unpublished), University of Colorado: Boulder.
Mueller, S H, Goldfarb, R J, Miller, M L, Munk, L A, Sanzolone, R,
Lamothe, P J, Adams, M, Briggs, P H, McClesky, R B and
Theodorakos, P M, 2003a. Surface and ground water geochemistry
near the Donlin Creel gold deposit, southwestern Alaska, US
Geological Survey Open-File Report 03-492, 37 p.
Mueller, S H, Hart, C J R, Goldfarb, R J, Munk, L A and Diment, R,
2003b. Post-mining hydrogeochemical conditions, Brewery Creek
gold deposit, central Yukon, in Yukon Exploration and Geology, 2003
(Eds: D S Emond and L L Lewis), pp 271-280 (Yukon Geological
Survey).
Murphy, D C, 1997. Geology of the McQuesten River region, northern
McQuesten and Mayo map areas, Yukon Territory. Exploration and
Geological Services Division, Yukon, Indian and Northern Affairs,
Canada, Bulletin 6, 122 p.
Newberry, R J, Allegro, G L, Cutler, S E, Hagen-Levelle, J H, Adams, D
D, Nicholson, L C, Weglarz, T B, Bakke, A A, Clautice, K H,
Coulter, G A, Ford, M J, Myers, G L and Szumigala, D J, 1997.
Skarn deposits of Alaska, in Mineral Deposits of Alaska (Eds: R J
Goldfarb and L D Miller), Econ Geol, Monograph 9, pp 355-395.
NovaGold Resources Inc, 2002. Technical report: preliminary assessment
Donlin Creek Project, Alaska, 109 p.
ODea, M, Carlson, G, Harris, S and Fields, M, 2000. Structural and
metallogenic framework for the Scheelite Dome prospect, Yukon
Territory, in The Tintina Gold Belt: Concepts, Exploration, and
Discoveries (Eds: T L Tucker and M T Smith), pp 115-129 (British
Columbia and Yukon Chamber of Mines).
Patton, W W Jr, Box, S E, Moll-Stalcup, E J and Miller, T P, 1994.
Geology of west-central Alaska, in The Geology of Alaska:
Geological Society of America, Geology of North America (Eds: G
Plafker and H C Berg), v G-1, p 205-240.
Rombach, C S and Newberry R J, 2001. Shotgun deposit: granite
porphyry-hosted gold arsenic mineralization in southwestern, Alaska,
Min Dep, 36(6):607-621.
Rhys, D, DiMarchi, J J, Smith, M, Friesen, R and Rombach, C S, 2003.
Structural setting, style, and timing of gold mineralization at the
Pogo deposit, east-central Alaska, Min Dep.
Selby, D, Creaser, R A, Hart, C J R, Rombach, C S, Thompson, J F H,
Smith, M T, Bakke, A A and Goldfarb, R J, 2002. Absolute timing of
sulfide and gold mineralization: A comparison of Re-Os molybdenite
and Ar-Ar mica methods from the Tintina Gold Belt, Alaska,
197
S H MUELLER et al
Geology, 30(9):791-794.
Smith, M, Thompson, J F H, Moore, K H, Bressler, J R, Layer, P,
Mortensen, J K, Abe, I and Takaoka, H, 2000. The Liese Zone, Pogo
property: a new high-grade gold deposit in Alaska, in The Tintina
Gold Belt: Concepts, Exploration, and Discoveries (Eds: T L Tucker
and M T Smith), pp 131-144 (British Columbia and Yukon Chamber
of Mines).
Stephens, J R, Mair, J L, Oliver, N H S, Hart, C J R and Baker, T, 2004.
Structural and mechanical controls on intrusion-related deposits of
the Tombstone Gold Belt, Yukon, Canada, with comparisons to other
vein-hosted ore-deposit types, J Structural Geol, 26:1025-1041.
Szumigala, D J, Dodd, S P and Arribas, A Jr, 2000. Geology and gold
mineralization at the Donlin Creek prospects, southwestern Alaska,
Alaska Division of Geological and Geophysical Surveys Professional
Report 119, p 91-115.
198 Adelaide, SA, 19 - 22 September 2004
Thompson, J F H and Newberry, R J, 2000. Gold deposits related to
reduced granitic intrusions, in Reviews in Economic Geology volume
13: Gold in 2000, pp 377-400 (Society of Economic Geologists).
Tucker, T L and Smith, M T, 2000. The Tintina Gold Belt: Concepts
Exploration and Discoveries, 224 p (British Columbia and Yukon
Chamber of Mines).
Verplanck, P L, Mueller, S H, Youcha, E K, Goldfarb, R J, Sanzolone, R
F, McClesky, R B, Briggs, P H, Roller, M, Adams, M and
Nordstrom, D K, 2003. Chemical analyses of ground and surface
waters, Ester Dome, central Alaska, 2000-2001, US Geological
Survey Open-File Report 03-244, 40 p.
PACRIM 2004
Data Metallogenica Around the Pacific Rim and Across the
World
K N OSullivan1 and A D T Goode2
ABSTRACT
Data Metallogenica (DM) is unique as a global mineral sample collection
and deposit database with over 70 000 rock samples and a wide range of
descriptive, photographic and measurement data. Subscribers have access
to approximately four gigabytes of on-line data with many examples
included on the public pages. DM commenced its life on the Pacific rim
in 1971 and much recent growth has focused on countries of the Pacific
margin, particularly Australia, Fiji, Chile, Indonesia and Peru.
Approximately 600 separate deposits and mineral occurrences in this
region are represented in DM. The sets of rock samples have been
photographed at high resolution and included on the web site, together
with unpublished descriptions, deposit cross-sections, bibliographies,
field and mine photographs, petrology and reflectance mineralogy.
Figures and abstracts from PhD theses are a new addition and the
donation of proprietary databases by companies and individuals is an
encouraging trend. DM increasingly is being viewed as an acceptable and
important repository for fragile and irreplaceable personal and company
data, it is also being recognised in its true light as a high level information
system, not solely a collection of rock samples. Data Metallogenica is a
not-for-profit venture of AMIRA International that requires self-funding
from December 2004. Growth and development must be funded entirely
out of subscription and sponsorship revenues because no government
subsidy is provided. The Pacific rim is in the mainstream of mining and
exploration activities worldwide and those involved have much to gain
from a dynamically sustaining DM. The home page at
www.datametallogenica.com contains much example data and details of
subscription and membership rates.
INTRODUCTION
Data Metallogenica (DM) is unique as a global mineral sample
collection and deposit database. It has a core holding of some
70 000 rock samples assembled over a period of 33 years. It is
supported by a wide range of descriptive, photographic and
measurement data which is expanding rapidly. DM includes two
collections of rock materials which strongly complement each
other. The most extensive and best documented collection is
named Lithotheque; this contains approximately 65 000 small
samples (6 - 8 cm) fixed to aluminium plates in sets of
(nominally) 20 at a time. The Macrotheque collection contains
nearly 5000 samples and differs in two main ways:
1. the samples are loose, fist-sized and often accompanied by
thin or polished sections; and
2. samples in any one set may be drawn from widely scattered
localities representative of a particular environment of
mineralisation, unlike Lithotheque, where each sample set
has been collected from one site or one district.
The web database can be found at
www.datametallogenica.com. To this point, development has
concentrated on entering data from the main Lithotheque
collection; Macrotheque will be added in a subsequent phase of
work. Subscribers have access to approximately four gigabytes
of on-line data and a powerful and fast search facility. Files have
been optimised for both dial-up and broadband users. The data
that is available includes high quality photographs of every
Lithotheque rock set, together with extensive plan, section and
1. Data Metallogenica, 63 Conyngham Street, Glenside SA 5065.
2. MAusIMM, AMIRA International, Level 2, 271 William Street,
Melbourne Vic 3000.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
field photographic data, and interpreted PIMA mineralogy where
the sample types are suitable. Much example data has been
included on the public pages of the web site, where it may be
browsed at any time.
It is of historical note to PACRIM delegates that DM started
life on the Pacific rim. It came into being in 1971 when a young
Peter Laznicka, while waiting for his doctoral dissertation to be
assessed by the University of Manitoba, took employment with
Seltrust Australia. Tasked with traversing large tracts of country
along the eastern seaboard of Australia, he began mounting small
samples on cards, which were then used extensively in working
meetings with the company. The system of organisation and
storage was refined in years following, and remains a unique
strength that underpins the DM database. It has lent itself to
efficient digital data capture, and makes any personal visit an
effective use of time spent: empirical comparisons can be made
between many deposits in different locations in a matter of
minutes.
BACKGROUND TO AMIRAS OWNERSHIP OF
DATA METALLOGENICA
In the late 1990s, nearing retirement as professor of mineral
deposits at the University of Manitoba, Peter Laznicka sought to
secure the future of DM beyond his lifetime. He wanted it to
remain available to the public and continue to be developed to
serve future generations of explorers and educators.
In 1997, he approached AMIRA International for assistance;
this resulted in an agreement whereby ownership transferred to a
joint venture established between AMIRA International and the
Australian Mineral Foundation (AMF). Funding came from a
consortium of companies and agencies with special assistance
from the South Australian Government. The initial three-year
programme set out to move the collection from Canada to
Australia, relocate the Laznicka family and engage Peter and his
geologist wife Sarka on the AMIRA project. The work of the
project aimed to fill significant gaps in the collection, digitally
image sample sets for the Internet, and build a web site to serve
the data. This was completed successfully by mid 2002, and a
further two-year extension project (P554A) floated to extend
achievements of the first project; notably by adding supporting
deposit data, especially cross-sections, maps, field photographs
and petrology. The much lamented closure of the AMF in
December 2001 created significant management and
infrastructure problems, which have been worked through but
which have left their financial legacy. Full ownership was
assumed by AMIRA International, which now runs data
Metallogenica as a not-for-profit enterprise on behalf of the
world-wide minerals industry and education providers.
DATA METALLOGENICA AND THE PACIFIC RIM
The Pacific rim is particularly well served by sample sets and
mineral deposit information in Data Metallogenica. A number of
major sampling trips during the life of the AMIRA projects have
deliberately sought to add deposits from the wave of discoveries in
recent times; nowhere more extensively than in South America.
Field visits have been made to the countries of Chile, Peru, Fiji
and Indonesia, plus the Australian states of Queensland, New
South Wales, and Victoria. Samples and data from approximately
199
K N OSULLIVAN and A D T GOODE
FIG 1 - Distribution of circum-Pacific mineral deposits for which samples and data are held by DM.
600 separate deposits and mineral occurrences bordering the
Pacific can be searched and viewed via the DM website or, even
more profitably, visited at the DM Centre, located at 63
Conyngham Street, Glenside (Adelaide), South Australia. Figure
1, centred on the Pacific basin, shows the location of
deposit-related information held by DM for this region.
Mineral exploration across the world is highly competitive and
those geologists who have seen most rocks in important settings
are likely to be most effective in the discovery process. Andrew
White wrote, The key component to the success of any
exploration programme is information, coupled with the
cognitive and conceptualising skills of the explorationist
(Management of Mineral Exploration, White 1997). DM is
unique in its ability to assist development of those critical
cognitive and conceptualising skills.
To give one instance: DM contains approximately 4000
samples from more than 200 porphyry systems worldwide which
can be viewed and compared at any web-connected location. Not
unexpectedly, not all alteration in porphyry coppers is perhaps as
expected; well illustrated by examining DM samples collected
from the Island Copper porphyry deposit on Vancouver Island. It
also happens that this deposit is long worked-out and cannot be
visited in the field, so that samples held by DM are able to add an
important, and unique, piece to the cognitive jigsaw.
Table 1 provides a more extensive listing of deposits, selected
to highlight the range and diversity of sample sets and supporting
data for countries around the Pacific basin. It does not include all
mineralised sites from this area in the database, nor does it take
account of the many sets of representative lithologies to be found
there, but it does provides an accurate snapshot.
200 Adelaide, SA, 19 - 22 September 2004
DM is a research tool that caters for all levels of capability,
including experienced geologists. In particular, the deposit
Legend that accompanies each suite of samples contains an
unpublished graphical cross-section compiled from a
combination of published, company and field visit information.
Also included is a summary description of the deposit samples
themselves and the host lithologies. Unsolicited strong support
for DMs approach came earlier this year from Dr Michael
Gunning, a geologist with considerable industry experience, who
now holds the position of Mineral Deposits Research geologist
with the Saskatchewan Ministry of Industry and Resources in
Canada. He wrote:
The strength of DM is the sample suites, and the
power of DM is global access to the suites via
high-res photos. The legend/cross-section format
for describing the suites makes the collection
more than just a bunch of shiny rocks by having
host rocks as well as ore samples, and by
outlining the broad spatial relationships of units
and rock types outlined on the legend, DM is
truly a research tool.
A further strength not immediately obvious is that most
samples have been collected by only one person, using a rigorous
empirical and non-genetic strategy of representing variation as
seen in the field. Thus, samples have been selected to include
mineralisation, alteration, host lithologies and weathering effects.
Minor mineralised showings have not been neglected either,
providing a very complete insight into the expression of
mineralisation in any given metallotect.
PACRIM 2004
 DATA METALLOGENICA AROUND THE PACIFIC RIM AND ACROSS THE WORLD

TABLE 1
A selection of deposits from the Pacific Rim found in Data Metallogenica.

PACRIM deposits in Data Metallogenica

Australia NSW: Cadia Cu, Captains Flat VMS, Cargo Au, Copper Hill Cu, Gidginbung Au, Hill End Au, Junction Reefs Au, Lake Cowal
Au, Lewis Ponds VMS, Mineral Hill Au Cu, Northparkes Cu Au, Peak Hill Au, Torrington W Bi, Woodlawn VMS, Wyalong
Au, Yeranderie Pb Zn. Qld: Cracow Au Ag, Gympie Au, Hadleigh Castle Au, Kidston Au, Moonmera Cu Mo, Mt Chalmers Cu,
Mt Coolon Au, Mt Leyshon Au Ag, Mt Morgan Cu Au, Mt Rawdon Au, Red Dome Au Cu, Pajingo Au Ag, Wirralie Au Ag,
Yandan Au. Vic: Ballarat Au, Bendigo Au, Castlemaine Au, Fosterville Au, Heathcote Au, Stawell Au. Tas: Hellyer VMS,
Hercules VMS, King Island Sn, Mt Bischoff Sn, Mt Lyell Cu, Renison Bell Sn, Rosebery VMS, Zeehan VMS
Central America Mexico: Alamos Cu, Cananea Cu Mo, Cerro de Mercado Fe, Concepcion del Oro Cu Au, Copala Ag, Durango Fe, El Arco Cu
Mo, El Mercurio Hg, Hidalgo Au, El Triunfo Au Ag, Fresnillo Ag, Guadalcazar Hg, Guanajuato Ag Au, Huitzuco Hg. Costa
Rica: Esparza Au
North America Canada: Afton Cu, Ajax Cu, Babine Lake Cu, Highland Valley Cu, Bluebell Zn Pb, Boss Mtn Mo, Brenda Cu Mo, Britannia Cu
Zn, Myra Falls VMS, Cariboo Bell Cu, Horn Ag, Magnum Cu, Copper Creek Hg, Copper Mtn Cu Au, Blue Grouse Cu,
Craigmont Cu, Duncan Lake Zn Pb, Endako Mo, Ferguson Zn Pb, Gibraltar Cu, Goldstream Cu Zn, Bodie Au Ag, Campo Seco
Cu Zn, Crescent Mn, Darwin Pb Zn, Mother Lode Au, Riverside Au Ag, Island Mtn Cu, McLaughlin Au, Pend'Oreille Zn Pb,
Mountain Pass REE, Nabesna Au, New Almaden Hg, Red Dog Zn Pb, Red Mtn Ag, Republic Au Ag, Willow Creek Au. USA:
Bisbee Cu, Boriana W, Castle Dome Cu Mo, Chloride Pb Zn, Copper Basin Cu, Copper Creek Cu Mo, Dundee Cu,
Esperanza-Sierrita Cu, Four Metals Cu, Globe-Miami Cu, Iron King Zn Pb, Ithaca Peak Cu, Magma Cu Zn, Inspiration Cu,
Mineral Park Cu Mo, Mission, Cu Mo, Morenci-Metcalf Cu, Oatman Au, Ray Cu, Safford Cu, San Manuel-Kalamazoo Cu Mo,
Silver Bell Cu, Twin Buttes Cu, United Verde Cu Au
Oceania Fiji: Emperor Au, Faddys Au, Tuvatu Au, Wainivesi Zn Pb. Nauru: phosphate. New Caledonia: Kopeto Massif Ni, La Ballade
Cu, Marais Kiki Cr, Tiebaghi Cr. New Zealand: Bendigo Au, Waihi Au Ag, Macraes Flat Au, Ohakuri Au Ag, Reefton Au,
Rotakawa As Au, Martha Hill Au Ag, Waikato Heads Fe, Waiotapu As Sb. PNG: Lihir Au, Ok Tedi Cu Mo, Panguna Cu Au,
Porgera Au. Vanuatu: Forari Mn
South America Peru: Antamina Cu Zn Ag, Antapaccay Cu Au, Casapalca Pb Zn, Cerro de Pasco Pb Zn, Colquijirca Ag Pb, Cuajone Cu Mo, El
Galeno Cu Au, Hualgayoc Cu Au, Huancavelica Hg, Mala Cu, Michiquillay Cu, Milpo Pb Ag, Minas Conga Cu Au, Morococha
Zn Pb, Pierina Au Ag, Quellaveco Cu Mo, Rio Pallanga Zn Pb. Chile: Adrianitas Fe, Aguas de las Falda Au, Andacollo Au,
Candelaria Cu Au, Cerro Iman Fe, Cerro Negro Cu, Chanarcillo Ag, Chuqicamata Cu Mo, El Abra Cu, El Hueso Au, El Penon
Cu, El Romeral Fe, El Salvador Cu, El Soldado Cu, El Teniente Cu Mo, El Tesoro Cu, El Tofo Fe, Escondida Cu, Inca del Oro
Au, La Africana Cu, La Coipa Ag Au, La Estrella Cu, La Reina Cu, La Union Cu, Lo Aguirre Cu, Los Pelambres Cu Mo,
Mantos Blancos Cu, Mantoverde Cu, Maria Christina Zn Pb, Ovalle Fe, Panulcillo Cu, Potrerillos Cu Mo, Punitaqui Hg, Punta
del Cobre Cu, Quebrada Marquesa Mn, Rio Blanco Cu Mo, Sierra Gorda Cu, Teresita Cu, Tocopilla Cu, Ujina Cu, Zaldivar Cu
Western Pacific Rim Indonesia: Batu Hijau Cu Au, Cikatomas Mn, Cirotan Au Ag, Ertsberg Cu Au, Gosowong Au Grasberg Cu Au, Kelian Au,
Pongkor Au. Japan: Akenobe Cu Sn, Hanaoka Zn Pb, Hitachi Cu Zn, Hosakura Zn, Ikuno Sn, Kamaishi Fe Cu, Kamioka Zn,
Kosaka Zn Pb, Matsuo Fe, Osarizawa Cu, Shimokawa Cu Zn, Toyoha Pb Zn, Yanahara Cu, Yoichi Mn Pb. Malaysia: Bau Au
Ag, Gambang W, Mamut Cu Au, Sungei Besi Sn, Sungei Lembing Sn. Philippines: Acupan Au, Antamok Au, Dizon Cu Au,
Kennan Au Cu, Mabuhay Au, San Antonio Cu, Tapian Cu, Masinlock Cr, Paragon Au, Surigao Au, Santo Nino Cu Au, Santo
Tomas Cu Au, Carmen Cu Au. Taiwan: Chinkuashih Cu Au. Thailand: Pa Daeng Zn, Phuket Sn Ta, Song Tho Zn Pb

DM should be considered as an essential companion to that
which is published in technical journals; whether it be a paper on
epithermal deposits at Pongkor, Indonesia as in the May-June
2002 volume of Economic Geology, the Waihi epithermal system
of New Zealand as in the March-April 2002 volume of Economic
Geology, or the Hillgrove gold-antimony deposit in NSW
Australia as in the April 2000 volume of Mineralium Deposita.
In all these cases, as is not uncommon, there is a dearth of
photographs because of the cost of publishing. Colour
photographs are extremely rare, and photo-quality reproductions
totally absent. DM fills this void by providing high resolution
photographs of rock samples and a range of field shots and other
data not previously published. To illustrate this with two
examples, selected at random from a deliberately old, 1978,
volume of Economic Geology: neither the definitive paper on the
Panguna porphyry copper deposit on Bougainville Island in
Papua New Guinea (Baldwin, Swain and Clark 1978), nor one on
the Kosaka (Kuroko) mine in Honshu, Japan, include any field or
mine photograph. Both mines are long closed and currently
inaccessible; Panguna due to insurrection, and Kosaka because it
was an underground operation. The DM web site, however,
contains 39 rock samples from Panguna, 20 rock samples from
Kosaka, interpreted PIMA mineralogy of the samples in both
deposits, and extensive field and mine photographs for Panguna.
Additional information and photographs for Kosaka are held on
file and will be added later. All this data is in addition to the
cross-sections and sample descriptions contained in Legends
which accompany every sample set.
A NEW DM DATA DIMENSION ON THE WEB
This year has seen very rapid expansion of DM data content,
which includes the addition of many new high resolution
photographs of individual DM samples and the growth of
supporting data for deposits and settings. Supporting data includes
descriptions of deposits and settings, plans and cross-sections,
mine and field photographs, much of which has never been
published and which has been generously provided by mining
companies or from the personal files of consultants and long-time
explorers. Text figures and abstracts from post graduate
dissertations are an important and growing new component. A
particularly good example is Steve Garwins PhD dissertation on
the Batu Hijau porphyry copper-gold system in Sumbawa,
Indonesia. Many excellent figures from this thesis can be viewed
on the public pages of DM by clicking the Example Data button
on the DM home page and following the Batu Hijau links.
Also new and available, although not yet for all sample sets, are:
selective bibliographies, and
photogalleries of plans, sections, field photographs and
petrology.
Examples of Pacific basin deposits supported by this new
information are shown in Table 2.
Users have noted that on-line searches in Data Metallogenica
are fast and effective both on dial-up and broadband connections.
Anyone may visit and use the search screen (Figure 2) located

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 201

K N OSULLIVAN and A D T GOODE

TABLE 2
Examples of Pacific basin deposits supported by photogalleries
and selective bibliographies.

Examples of deposits supported by new types of data

Country Commodity Deposit name
Australia Cu Cadia East, NSW
Au Fosterville, Vic
Au Gidginbung (Temora), NSW
Au Mt Leyshon, Qld
Zn Pb Cu Woodlawn
Chile Cu Mo Los Pelambres
Fiji Au Emperor
Indonesia Cu Au Batu Hijau
Cu Au Grasberg-Ertsberg
Au Pongkor
Papua new Au Lihir
Guinea Cu Au Ok Tedi
Cu Au Panguna
Peru Cu Zn Ag Antamina
Au Ag Pierina
Taiwan Cu Au Chinkuashih
USA Cu Silver Bell

FIG 2 - DM search screen with new Deposit Type drop-down

under the Search Catalogue button on the DM home page. It is
identical to the search screen presented to subscribers when they
log-in, except that forward links are disabled when results appear.
The search screen now contains a drop-down Deposit Type
menu which recently has been completely redesigned to provide
a much more powerful and comprehensive search aid. It may be
employed either on its own or in conjunction with other search
fields to obtain very precise outcomes; boolean searches permit
highly refined queries to be carried out. For example: a search
for Au, not Cu, not porphyry-style in British Columbia, will
result in 11 deposit matches, compared to 22 matches obtained in
searching simply for Au in British Columbia. In the near future it
is planned to add similar drop-down menus for alteration type
and rock-type association together with a free-text keyword
search facility.
THIRD PARTY DATABASES
Proprietary databases provided by individuals and companies
offer opportunities for DM to expand its information capabilities
particularly quickly. Three of direct relevance to the Pacific rim
are:
1. the Andean Gold Database compiled by David Shatwell;
2. the Codelco Petrological Database of Andean Rocks and
Minerals, compiled by Enrique Tidy; and
3. the Newcrest Gold Database compiled by David Jones.
These have not yet been included on the DM web site and are
currently under examination to determine the most effective
methods of incorporation. The Andean Gold Database was
compiled in 2000 and contains descriptions of 70 deposits in the
Andean countries of Colombia, Ecuador, Peru, Bolivia, Chile and
Argentina. The Codelco Petrological Database of Andean Rocks
and Minerals is an inventory of more than 15 000 specimens
submitted by Codelco exploration teams in Chile, Mexico, Brazil
and Peru for petrological or other analysis, including ore
menu.
microscopy, XRD, microprobe and age dating. It contains more
than 5000 photographs and associated technical reports. The
Newcrest Gold Database covers global gold production from
1847 - 2002. It contains world gold production by year and
country, gold production for 316 companies (~75 per cent of total
world production), available data on 1316 individual mines (also
grouped by district).
CONCLUSION
It is exciting that DM is increasingly being viewed in more than
one country as a permanent repository for fragile and
irreplaceable personal and company data. It is very important
that DM is now being recognised in its true light, which is as a
high level information resource of considerable importance to
both industry and education. Data Metallogenica is much more
than a collection of rock samples; however, the photographs of
those samples are without peer on the web in terms of their
quality and resolution and no geoscientist, particularly in
mineral-related fields, should be without access to it. That DM
has come so far and delivered so much is a tribute to Peter
Laznickas lifetime of dedication, to the vision of AMIRA
International, and to the financial and moral support of the
mining community around the world. From December 2004, the
maintenance, growth and development of DM will depend
entirely on the level of subscription revenues that are generated,
because DM receives no government grant or subsidy. By taking
up subscriptions, companies institutions and individuals can help
ensure the continued rapid growth of information available on
the web through DM, and also provide a safety net to capture this
fragile sample and mineral deposit data which otherwise would
be lost. The Pacific rim is in the mainstream of mining and
exploration, and those involved have much to gain from a
dynamically sustainable DM. Details of subscription and
sponsorship rates are available on the DM home page at
www.datametallogenica.com.

202 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DATA METALLOGENICA AROUND THE PACIFIC RIM AND ACROSS THE WORLD

ACKNOWLEDGEMENTS
The authors are grateful to AMIRA International for permission
to publish this paper, and to the many companies, institutions and
individuals around the world whose support in funds and
information through AMIRA Projects P554 and P554A made it
possible for Data Metallogenica to become an acknowledged
global electronic resource, and at the same time expand the
collection of physical samples and data. Particular recognition is
extended to Peter Laznicka, his geologist wife Sarka, and their
twin sons for their efforts in building Data Metallogenica as a
family endeavor over the period of some 28 years prior to its
transfer to AMIRA International and relocation to Australia. The
special assistance of the government of South Australia has been,
and continues to be, of great importance; the relationship grows
closer and increasingly productive with the passage of time.
Lastly, and with considerable sadness, the role of the Australian
Mineral Foundation (AMF) should be recognised. The AMF
provided both operations management and premises for Data
Metallogenica in a joint venture established with AMIRA
International; until the AMF finally closed its doors in December
2001.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 203

Discovery Techniques and Regolith

Seeing Through Alluvial Gold Fine Fraction Stream Sediment Sampling M C Arundell and 207
in the Sofala Area, Central New South Wales S G Gatehouse

HoistEM A New Airborne Electromagnetic System G W Boyd 211

Groundwater Geochemistry as a Tool for Mineral Exploration Under Cover P de Caritat and D Kirste 219
The Curnamona Province

The FalconTM Airborne Gravity System as an Exploration Driver in P Harman, D Isles and 221
Australia C Anderson

Automated Mineralogical Logging of Core From the Emmie Bluff, Iron J Huntington, A Mauger, 223
Oxide Copper-Gold Prospect, South Australia R Skirrow, E Bastrakov,
P Connor, P Mason,
J Keeling, D Coward,
M Berman, R Phillips,
L Whitbourn and
P Heithersay

New Ways to Explore Through the Regolith in Western New South Wales K G McQueen and 231
A McRae

Prospectivity Mapping Using GIS With Publicly Available Earth Science G A Partington and 239
Data A New Targeting Tool Being Successfully Used for Exploration in M J Sale
New Zealand

Recent Developments in Electrical Exploration T J Ritchie 251

Weathering of Base Metal Deposits and Implications for Exploration in the K M Scott 257
Lachlan Fold Belt

A Spatially Enabling Relational Geologic Database of Java Island, Western L D Setijadji and 263
Sunda Arc of Indonesia A Step to the Geoscience and Mining Data K Watanabe
Model

The Controls on Element Distribution in the Regolith at the Portia and K P Tan, P de Caritat, 273
North Portia Prospects, Curnamona, South Australia K M Scott, R A Eggleton
and D Kirste

Compositional Variations of White Mica in the Footwall Hydrothermal K Yang, J F Huntington, 283
Alteration System of the Hellyer Zn-Pb Deposit, Tasmania K M Scott and P Mason
Seeing Through Alluvial Gold Fine Fraction Stream Sediment
Sampling in the Sofala Area, Central New South Wales
M C Arundell1 and S G Gatehouse2
ABSTRACT
Extensive gold mining activity dominantly alluvial has taken place in
the Sofala-Wattle Flat district since the 1850s. This work will have
located most, if not all, of the visible sources for the gold in current
stream systems. Also, there are many gold bearing elevated Tertiary
alluvial terraces in the district which may confuse possible gold signals
from primary hard rock sources in the streams. Associated element
signatures eg As and other chalcophiles, will discriminate non-alluvial
signatures but will not necessarily distinguish non gold bearing
mineralisation. Both trace elements and conventional gold analyses will
dominantly re-identify known occurrences.
Conventional BCL sampling is considered misleading as experience
from the Oberon area to the south of Sofala demonstrated that BCL
sampling was very sensitive perhaps too sensitive. All occurrences
major or minor of gold will be detected (Gatehouse, 1992).
To improve on BCL sampling, a small, -120# sample with selective
duplication was recommended to screen coarse-grained gold anomalism
and a program of sampling covering an area thought to be prospective
was enacted. Samples returning anomalous Au were analysed in duplicate
and if the results demonstrated reproducibility, indicating fine gold only,
the sample site was assigned a high priority. It was found that most of the
reproducible samples clustered in one area Spring Gully.
The results of this sampling program led to immediate identification
and, ultimately, the successful delineation of the Spring Gully gold
deposit near Wattle Flat.
INTRODUCTION
The Turon river alluvial goldfield and later discovered hardrock
vein sources were the focus of Australias second gold rush in the
1850s. It was considered likely that the Sofala area had been
extremely well prospected for hydraulically active (coarse and
visible) gold. During and since the rush all streams would have
effectively been panned and any traces of gold would have been
followed up with a collective rigour that modern day exploration
is unlikely to match. With any of our conventionally available
gold exploration techniques, the majority of the gold anomalies
detected would represent sources previously located by
prospectors.
Analysis of gold associated elements, eg As, Bi, Sb, Mo, and
base metals should give a better basis for locating hard rock gold
sources than any conventional gold technique but that there will
be significantly anomalous trace elements with all prospective
hard rock gold sources. Significant savings in analytical costs
and field collection rates could be achieved by not collecting or
analysing samples for gold in such a situation.
There are at least two risks with this approach. One is that
there can be metal anomalism that is not related to gold sources
and these element assemblages will not distinguish these false
anomalies. The second risk is that there could be significant hard
rock gold deposits which do not have a detectable associated
metal signature.
1. Arundell Geoscience, PO Box 170, Belmont WA 6984.
2. Hellman and Schofield, Suite 6, 3 Trelawnley Street, Eastwood NSW
2122.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
STREAM SEDIMENT GEOCHEMISTRY
Drainage patterns and sampling densities
The topography of the Sofala area varies moderately, with a local
relief of less than 100 m. Drainages are well developed with third
and fourth order streams often having strong geological control
on scales of tens of kilometres. First order streams are often
small gutters which flow directly into third or higher order
streams. Consequently, the natural range of catchment sizes
appears to be biased towards either 0.1 - 0.3 km2 or 2 - 5 km2.
Neither of these catchment sizes was considered suitable for a
representative stream sediment program. Sampling of all the
small basins was considered unjustifiably expensive while
sampling of the larger catchments would have biased results
towards coarse (probably alluvial) gold sources and signals
would be significantly confounded by both nugget problems and
hydraulic effects.
It was considered that sampling should be conducted at a
density of one per 3 - 4 km2. Where first order gutters drained
into third to fourth order streams, proportionate sampling of each
gutter and bulking into one sample representing about 0.25 km2
was considered to be preferable to both not sampling or sampling
all gutters individually.
Sampling and analytical technique
A number of observations indicated that a fine fraction gold
technique would be advantageous for the Sofala catchments.
Generally, the basic volcanic and volcaniclastic rocks of the
targeted Sofala Volcanics weather recessively when compared
with the younger surrounding felsic volcanics, sandstones,
argillites and other clastic units. The volcanics produce fine clays
and iron oxides whereas the sandstones and argillites produce a
detritus of coarser lithic aggregate. Fine fraction screening would
thus bias samples towards basic volcanic detritus and possibly
towards finer gold sources. Assay values in fine fractions were
expected to be higher than 5 kg Bulk Cyanide Leach (BCL) as
less coarse debris would be included as a dilutent in the sample.
Sampling would therefore be more consistent and representative
of the basic volcanics in the catchments and will lessen the effect
of the clastic and felsic lithologies in mixed drainage basins.
Despite the optimised nature of this sampling, it was
considered that the potential for misleading anomalism might
remain high for a BCL gold analysis. Perched alluvial workings
are present on the flanks of a many third to fourth order
catchments in the Sofala area. BCL gold analysis alone would
not distinguish these workings from hard rock gold sources.
Further, sampling conducted by Homestake near Sofala
illustrated this point (Homestake, 1992). There, a one per 5 km2
BCL survey yielded 80 per cent anomalous samples at a 1 ppb
level. At a higher anomalous level (10 ppb), it was clear that
anomalism merely outlined areas in which ground was heavily
disturbed by historic gold diggings. It is considered probable that
much of the gold detected was hydraulic in origin, perhaps
recycled from now remnant alluvial deposits in perched Permian
conglomerates.
207
M C ARUNDELL and S G GATEHOUSE

Because of the remnant alluvial problem, it was considered

that gold analyses alone would be far less useful (possibly even
misleading) than chalcophile trace metal analyses. Arsenic is
known to be associated with gold in a number of the Sofala
deposits and thus was to be used as a guide to follow-up. Other
elements such as Sb, Te, Bi, Hg, Ag and Mo were also used.

OPTIMUM SIZE FOR SAMPLING AT SOFALA

The answer to this question is dependent on how much gold,
which is hydraulic (alluvial) and therefore could have been seen
by past prospectors, is to be avoided. A balance between locating
previously unseen fine gold occurrences and relocating
unrecorded gold occurrences is required. With a single sample
type there is not a high level of confidence in doing both. If past
discoveries, recorded or unrecorded, are relocated, the technique
will necessarily underrate fine gold sources. It was decided that
known prospects should be avoided.
A stream sediment technique for Sofala should concentrate on
gold less than 50 micron (becoming very difficult to discern with
the unaided eye) but with lesser tolerance above that level to
around a visible single grainsize (100 micron). For the purpose
of modeling, it was also assumed that gold was spherical to a
reasonable degree.
A sample simulation system was developed in which the
salient parameters include the expected grade of the stream, the
grain size distribution gold and the sample size. Other parameters
included gold shape and density, debris shape and density.
A series of models relating sample reproducibility to sample
mass assuming changing gold grainsizes at a constant stream
concentration is shown in Figure 1. A complimentary series of
models relating sample reproducibility to sample mass using
changing stream concentration at a constant gold grainsize is
shown in Figure 2. In these figures, only those streams which
survive the following criteria are plotted. A total of 400 streams
have been simulated in each of the diagrams.
From these figures an indication can be deduced of what mass
of sample will give reproducible results from particular stream
gold grade and grainsize. Assuming a stream concentration of
20 ppb, Figure 1 shows that assays become stable above about
20 - 30 grams when grainsizes of gold exceed around 50 micron,
near the visibility limit of a single grain. Assuming gold is all
below 50 micron, Figure 2 shows that gold assay stabilise when
stream concentration goes above about 20 ppb in using about
20 g samples.
Samples below 30 g from a course gold bearing stream have
an high probability of not detecting the gold or more rarely
getting an extreme value. Moreover, if a reassay is conducted in
the event of an extreme value it will probably not be repeated. At
higher gold concentrations in these streams, eg 20 ppb, a smaller
sample, say 20 g would give similar results.
Figure 1 also shows that 20 ppb streams can give assay results
of 100 ppb. The question then arises whether such a result from a
single sampling is due to a coarse gold particle or a consistent
high fine gold concentration. It is not possible to distinguish this
from a single determination. If a second assay is done on the
same sample, then a fine gold stream would persist in its
anomalism while a coarse stream would have a high probability
of disappearing. A strategy for collecting sufficient material for
duplicate assaying will enable re-assay of a proportion of the
samples, which on first assay are anomalous, to determine the
persistence and reproducibility of the anomalous result.
The conclusion is that samples of 20 - 30 g are optimum for
distinguishing between coarse and fine gold in samples which
are known to be finer than about 80#. Coarse gold will be rarely
FIG 1 - The effect of changing grainsize on the detectability of
seen, and if seen will give extreme results and will not reproduce.
20 ppb gold by different sample sizes (spherical grains).

208 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 SEEING THROUGH ALLUVIAL GOLD FINE FRACTION STREAM SEDIMENT SAMPLING
FIG 2 - The effect of changing stream concentration on the
detectability of 20 micron gold by different sample sizes
(spherical grains).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
A sampling strategy selective for fine gold
Optimisation of sample sizes was therefore based on the sample
size which will give the most reproducible results for fine gold
and least reliable on sampling of coarse gold.
Criteria are: collect stream samples, analyse for gold, discard
drainages below 5 ppb, assay another sample from the residues
of the drainages exceeding 5 ppb and discard drainages whose
second assays are below 5 ppb.
The drainages which remain are anomalous in relatively fine
gold. In all probability, close agreement between assay pairs
reflects finer gold or high concentration or both. Gold which is
coarse or in low concentrations or both will, in all probability,
disappear on re-assay. There is a particular sample size which
will optimise detection of gold which is finer than a nominated
grainsize.
The feature to note is the low to absent number of successful
outcomes from small samples of streams which contain
coarse-grained gold. In the rare event that a coarse gold bearing
stream passes the proposed test, the difference between assay
pairs is often large. These simulations suggest that a sample size
of about 25 g should be selective for fine gold even in quite rich
coarse gold bearing streams.
FIELD PROCEDURES
As visible gold is to be avoided, there is little point in analysing
any material greater than 100 - 150 micron. If +100 - 150 micron
material contains no gold then it would effectively dilute the gold
concentration in the -100 - 150 micron material. Analytical
results of such samples would be lower than for -100 - 150
micron samples and consequently uncertainty is increased as the
detection limit of the analytical technique would be more likely
tested. If the +100 - 150 micron material contains gold particles
such samples would be rejected by the sampling protocol.
By definition, sub-hydraulic gold will not concentrate in the
streams. It should flush through the stream systems with fine
lithic debris. Provided debris is fed into stream systems in
proportion to drainage basin area (an unlikely situation but a
necessary assumption) a rough calculation suggests that an area
of 100 25 m averaging 1 ppm fine gold will dilute to 10 ppb in
a drainage representing 0.25 km2. If these figures remotely
reflect reality, they indicate that a high sampling density is
necessary for confident assessment of fine gold sources. A
prioritising of area on the basis of geological prospectivity and
sampling at high density is preferable to blanket sampling of
large areas at low density under such assumptions.
Devonian argillites and siltstones appear to weather into sand
sized lithic aggregates while the Ordovician Sofala Volcanics
mostly weather to sesquioxides and clays. In drainage basins of
mixed lithologies, fine fraction sampling should bias sampled
debris to volcanic sources. Given the Sofala Volcanics are
considered more prospective, this is advantageous. Therefore, it
was suggested that -125 micron (-120#) material be collected for
assay.
RESULTS
A program of fine fraction drainage sampling was carried out
primarily over an area south of the Turon River. This area was
delineated from mapping and geophysical interpretation as
having high geological prospectivity. The Big Oakey fault, a NW
trending structure was targeted since this is considered to be the
primary control of gold mineralisation at a number of Au
prospects in the Sofala area (Figure 3).
A total of 405 samples were collected. Samples were wet
sieved by the laboratory (ALS Orange) to -120# and analysed
without further pulverising for Au (PM219 FA/CR) in
duplicate if the original was >10ppb and for a suite of elements
by ICP (IC591 As, Sb, Bi, Hg, Ag, Mo, Se, Te, Cu, Pb, Zn, Fe,
Mn, Ca, Cd, Co, Ni, Cr, V). In later surveys, samples were dry
sieved by the field crew.
209
M C ARUNDELL and S G GATEHOUSE
FIG 3 - Mapped geology and location of sample points.
FIG 4 - Au versus duplicate Au. Fine reproducible Au samples and
Coarse non-reproducible Au samples highlighted.
The results from this program were initially analysed by
plotting original Au values versus duplicate Au for all the
+10 ppb Au samples (Figure 4). Reproducible Au samples
greater than 50 ppb were assigned a high priority. Reproducible
samples were defined as samples where the initial Au value was
in agreement with the duplicate Au value (Figure 4). These
anomalous samples cluster in what became known as the Spring
Gully prospect.
Subsequent soil sampling at Spring Gully defined a 1200 m
long, 50 - 250 m wide +100 ppb Au anomaly. The soil anomaly
straddles the contact of the Sofala volcanics and Silurian
210 Adelaide, SA, 19 - 22 September 2004
Chesleigh Formation. Field relationships indicate that the Sofala
Volcanics were thrust over the Chesleigh Formation at this
location.
Initial drilling of the soil anomaly encountered significant
mineralisation -48 m @ 1.52 g/t Au. Further drilling delineated a
low grade sub-economic gold resource of 3.55 Mt @ 1.01 g/t Au
(RGC quarterly report, March 1995). The style of mineralisation
encountered was considered to be analogous to sediment hosted
Carlin style mineralisation (Rowley, 1994).
The success of the fine fraction stream sediment technique is
shown by the style of mineralisation encountered at Spring
Gully. Spring Gully is a low-grade sediment hosted disseminated
Au deposit.
LESSONS
The following lessons can be drawn from the exploration of the
Sofala area and the discovery of mineralisation at Spring Gully.
Potential problem identified very early in exploration
program.
Conventional stream sediment sampling would have found
the same anomalies that either previous companies or past
prospectors had located.
Innovative (counter intuitive) approach to solving a specific
problem.
Taking a small sample in Au exploration would initially
seem a good way of missing gold. However, to some extent
this was the objective of the sampling program to miss
coarse-grained gold.
Mineralisation targeted (Volcanic hosted) was not
mineralisation found (Sediment hosted).
The gold mineralisation at Spring Gully is mostly hosted by
the Chesleigh formation not the Sofala Volcanics. The stream
sampling program was not confined to the targeted sequence.
An approximate 500 m buffer was created such that all
material draining from the targeted volcanics would be
sampled.
CONCLUSIONS
The discovery of significant mineralisation at Spring Gully
demonstrates that exploration in historic goldfields can be
successful. Not all gold was discovered by the old timers
fine-grained gold mineralisation (<50 micron) is unlikely to have
been detected by panning and dollying.
Spring Gully also illustrates that fine-grained Au anomalism
can be discriminated from coarse-grained Au anomalism.
Perhaps of more importance to exploration in other areas, fine
Au may be masked by coarse Au.
REFERENCES
Gatehouse, S, 1992. An investigation into the feasibility for small
sample stream sediment surveys for fine gold in the Sofala
Volcanics, NSW, RGC Internal Report, Cr 92/3/3.
Homestake Australia Ltd, 1992. Second and final progress report on
exploration licence 3571 (Millah Murrah) for the period ending
December 31, 1991, Geological Survey of New South Wales
Unpublished Exploration Report, GS1991/048.
Rowley, D, 1994. The nature and origin of gold mineralisation at the
Spring Gully Gold prospect near Sofala, NSW, BSc Honours thesis
(unpublished), University of Canberra.
PACRIM 2004
HoistEM A New Airborne Electromagnetic System
G W Boyd1

ABSTRACT THE HOIST

HoistEM is a helicopter-towed time domain electromagnetic system. This
paper presents details of the towed array structure (the Hoist) as well as a
description of the instrumentation.
A comparison between HoistEM and ground TDEM (from which
HoistEM was derived) is given.
Over 50 000 line kilometres of HoistEM have been completed to date
and datasets for structural mapping, nickel, base metals, manganese,
ground water and bathymetry are presented and discussed.
1. Geosolutions Pty Ltd, 16 Centre Way, Belair SA 5052. E-mail:
gwb@senet.com.au
In order for a time domain electromagnetic system to be effective
the transmitter must be as large as possible. Experimentation
showed that a transmitter loop greater than 20 m in diameter was
required. The current HoistEM system consists of a 24 m
diameter loop of bare stranded aluminium conductor (Figure 1).
To support this loop in flight a Hoist using commercially
available windsurfer masts and specially fabricated carbon
fibre/kevlar components is used (Figure 2).
The Hoist uses high drag to ensure flying stability and
direction. The design survey speed is 45 knots airspeed and the
Hoist is tested to 60 knots to give a safety factor for unexpected
turbulence.

FIG 1 - Current HoistEM system consisting of a 24 m diameter loop of bare stranded aluminium conductor

FIG 2 - Hoist with commercially available windsurfer masts and specially fabricated carbon fibre/kevlar components.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 211

G W BOYD

THE RECEIVE LOOP

In any EM system the receive loop is a critical component. All
noise sources that the receive loop picks up degrade the signal
from the ground. Most airborne EM systems use a towed bird to
house the receive loop(s). The major source of noise is the
motion of the receive loop through the earths magnetic field.
HoistEM uses a multiturn receive loop that is directly attached
to the Hoist. The aim is to use the inertia of the Hoist to restrain
the movement of the receive loop and minimise the effect of the
receive loop moving through the earths magnetic field.
A particular problem with airborne receiver coils is
microphonic buffeting. This buffeting occurs because of the size
of the towed bird and the speed it is towed at. The buffeting FIG 3 - HoistEM transmitter waveform.
causes significant noise in the receive coil. By using a small
diameter coaxial cable for the receive loop and attaching it
directly to the Hoist this buffeting is eliminated. In order to power the transmitter a 25 horsepower Kohler
petrol motor is attached to the skids of the helicopter. Figure 4
There are three further advantages gained by having the shows this attachment.
receive loop attached to the Hoist. The system has a true
IN LOOP geometry. Also a large receive loop moment with a After the received signals have been collected by the receive
very high bandwidth is easily obtained. loop on the Hoist they are routed up to the helicopter via a
purpose built shielded cable. They are then digitally sampled by
There are also some disadvantages having the receive loop the receiver also attached to the rear floor of the helicopter. This
close to the transmit loop. An unavoidable system response is receiver is a specially modified version of the Normandy/
generated when the receive and transmit loops are in close Poseidon POSEM ground Time Domain EM receiver. The
proximity. The very large magnetic field caused by the receiver has a 20 bit self calibrating A-D converter that reads true
transmitter pulse overloads the electronics in the receiver and microvolts. It has NortWest Cape VLF rejection. Sampling is 128
high quality electronic design and componentry are required to channels with each channel being 112.7 microseconds wide.
minimise this effect.
Navigation uses a purpose built pilot guidance and position
acquisition system. A very high quality NovAtel RT20 GPS
INSTRUMENTATION IN THE HELICOPTER boardset is employed which can be differentially corrected using
The transmitter that generates the electromagnetic pulse is the Fugro Omnistar system (2 m accuracy) or a NovAtel RT20
attached to the floor behind the pilot. This transmitter produces a base station (20 cm accuracy).
five millisecond pulse of approximately 340 amps which is After initial switch-on no instrumentation in the helicopter
routed down to the Hoist via heavy duty copper welding cables. requires operator intervention until the end of the flight. The
This pulse has a very rapid switch off of 40 microseconds. The pilot is capable of switching the equipment on and off,
transmitter waveform is shown in Figure 3. eliminating the need for an operator in the aircraft.

FIG 4 - A 25 horsepower Kohler petrol motor attached to the skids of the helicopter to power the transmitter.

212 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


COMPARISON BETWEEN HOISTEM AND
GROUND TDEM
Figure 5 shows images at three delay times for both HoistEM
and Sirotem data collected over the same prospect. The HoistEM
data were collected in 2003 while the Sirotem data were
collected in 1998. The surveys are from the Eastern Goldfields of
Western Australia. The HoistEM was collected on 200 m spaced
lines, at a flying height of 35 m with data presented at 10 m
intervals along the lines. Sirotem data was collected using 100 m
loops at 100 m intervals. The Sirotem survey used 100 m line
spacing but only lines that correspond with HoistEM data at
200 m have been used in the comparison.
FIG 5 - Comparison between Sirotem and HoistEM data.
At the two earliest delay times the images are virtually
identical. At the latest delay time presented some differences are
apparent in the images and this is due to the HoistEM signal
levels decaying below system noise levels over most of the
survey area.
Figure 6 shows a comparison between HoistEM and Sirotem
decays directly over the most conductive feature in the area. The
Sirotem decay has been normalised so as to lie close to but not
directly over the HoistEM decay. Both decays show virtually
identical shapes out five milliseconds when the HoistEM decay
falls below system noise levels. It is apparent from the above
comparisons that low amplitude late time anomalies will only be
detected by a ground TDEM system which has a significantly
lower system noise level.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
HOISTEM A NEW AIRBORNE ELECTROMAGNETIC SYSTEM
FIG 6 - Comparison between Sirotem and HoistEM decays.
STRUCTURAL MAPPING
A small HoistEM survey was flown in the Forrestania area of
Western Australia in April 2004. The aim of the survey was to
provide structural information to assist in the location of the
portal for the decline into a new underground mine. The area
surveyed was a one by 2 km block.
The survey area was flown at 50 m line spacing with NS and
EW flight lines. Figure 7 shows an image of NS flight lines on
the left and EW flight lines on the right. The upper section is at a
delay time of 1.01 ms with the lower section at 1.92 ms. These
images are of raw data from the aircraft after standard
processing.
This data demonstrates how the HoistEM system geometry
minimises the effect of flight line direction.
SALLY MALAY NICKEL DEPOSIT
During June 2002 a HoistEM survey was conducted through the
Sally Malay region in the Kimberley of Western Australia. The
survey was flown on NNE oriented lines spaced 75 m apart with
a 30 - 40 m terrain clearance. Figure 8 shows the HoistEM
response in profile form of a flight line over the known
mineralisation. The same line was flown in opposite directions.
The deposit is clearly observed at 8 081 500 mN. The shape of
the response is consistent between the two flight lines with no
asymmetry to the response again highlighting the advantage of
HoistEMs symmetric configuration.
Figure 9 shows a conductivity depth section for one of the
previous lines. The deposit is clearly resolved compared to
background. The mineralisation has a conductivity contrast of
well over 10:1 to the surrounding host.
213
G W BOYD

FIG 9 - CDI image Sally Malay, Western Australia.

BASE METALS/NICKEL PLAY CENTRAL

AUSTRALIA
Figure 10 shows profiles and a CDI from a HoistEM survey
conducted in 2003 in the Arunta block of Central Australia. The
survey consisted of 200 m spaced NS lines at a nominal flying
height of 35 m. A very strong late time EM response is clearly
evident in both the profiles of raw data and the CDI.

FIG 10 - HoistEM profiles and CDI Arunta Block.

FIG 7 - Comparison between NS and EW flight lines.
Figure 11 presents a conductivity depth slice at 30 m as well as
the available magnetics which was collected at 500 m line
spacing. A 100 nT contour has been placed on the magnetics.
Figure 11 clearly shows that conductivity and magnetics map
different geological components. Some possible structural
interpretations have been placed on the EM data to further
highlight the differences between the EM and the magnetics.
Drilling has shown a vertical pipe of weathered material whose
depth extent has not been reached but exceeds many hundreds of
metres. The pipe has a conductivity of 800 mS/m which is
approximately one sixth that of sea water. The pipe is
approximately 1 km in diameter. Because of the size and depth
of the pipe a very strong EM anomaly is generated. This is
illustrated in Figure 12 where two images of raw data are
presented at the delay times of 3.2 ms and 12 ms. Raw data in
image form as shown in Figure 12 is available within one hour of
completion of flying in the field.

WOODIE WOODIE MANGANESE PROJECT

FIG 8 - HoistEM profiles Sally Malay, Western Australia.
A HoistEM survey was flown over the Woodie Woodie
manganese mine corridor in May 2002. This survey also
incorporated some test lines to allow further evaluation of the
HoistEM system by researchers at Curtin University. Survey line
spacing was 80 m with a flying height of 35 m. Some lines were
reflown at 70 m height.

214 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 HOISTEM A NEW AIRBORNE ELECTROMAGNETIC SYSTEM

Figure 13 shows a CDI created using EMFlow for the

discovery line over the ChrisD orebody. Manganese
mineralisation shows up as high conductivity in discrete CDIs
calculated in the Woodie Woodie Mine corridor. The two million
tonne ChrisD orebody sits under 20 m of Permian clay. Figure 14
shows a comparison between ground TDEM and HoistEM data
over the ChrisD orebody.

FIG 13 - CDI image Line 3270 Woodie Woodie, Western

Australia.

FIG 11 - HoistEM conductivity and magnetics Arunta Block.

FIG 14 - Comparison between HoistEM and ground TEM Woodie

Woodie, Western Australia.

Figure 15 shows CDIs generated from a repeat line which was

flown in the opposite direction. Excellent correlation with little
difference because of flight line direction is evident. Figure 16
shows the effect of flying at different survey heights. The upper
CDI is for the 40 m survey height. In order to maximise signal to
noise in an airborne EM system the lowest possible flying height
must be utilised.

FIG 15 - HoistEM repeat lines Woodie Woodie, Western Australia.

MAPPING SEA WATER INCURSION

During 2003 a long HoistEM line was surveyed in the Onslow
area of Western Australia. Figure 17 shows the location of the
25 km long HoistEM line on a landsat image of the area. The aim
of the survey was to map the extent of sea water incursion so as
FIG 12 - Images of raw HoistEM data Arunta Block.
to guide drilling for water bores.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 215

G W BOYD

FIG 16 - Effect of different survey heights Woodie Woodie, Western Australia.

FIG 17 - HoistEM line - Onslow, Western Australia.

A conductivity depth image is also presented in Figure 17. The
extent of the sea water incursion is obvious. The CDI shows a
resistive near surface layer overlying the sea water incursion.
This correlates with the landsat image which shows no surface
indication of saline water.
BATHYMETRY SPENCER GULF, SA
Figure 18 shows CDIs for two repeat lines flown out into
Spencer Gulf in South Australia. One line was flown from land
out to sea while the repeat line was flown back to land. The aim
of the survey was to assess the ability of HoistEM to map the
depth of seawater. The HoistEM navigation system was fitted to
a chartered fishing boat and depth soundings were made along
the survey line. These are shown as white bars superimposed on
the CDIs. Not only do the CDIs present the correct conductivity
for seawater but they also show that the depth of the water can be
inferred from the CDIs. The HoistEM system has been used for a
number of bathymetry surveys around Australia and its accuracy
in measuring sea water depth is currently being assessed by
DSTO research scientists.

216 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 HOISTEM A NEW AIRBORNE ELECTROMAGNETIC SYSTEM
FIG 18 - HoistEM CDIs Spencer Gulf, South Australia.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 217
G W BOYD

ACKNOWLEDGEMENTS
The author would like to thank the organisations who have given
permission for the public release of the data used in this paper.

218 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

Groundwater Geochemistry as a Tool for Mineral Exploration
Under Cover The Curnamona Province
P de Caritat1 and D Kirste2

ABSTRACT

The Broken Hill region is host to a single known supergiant Pb-Zn-Ag orebody and numerous smaller
deposits, which occur in the limited outcrop regions of the Paleo- to Neoproterozoic Curnamona
Province basement. Those areas of exposed basement have undergone mineral exploration for over a
century, yet additional significant mineral discoveries have been limited. Approximately 90 per cent
of the Province is concealed by the Cenozoic sedimentary sequence of the Callabonna Sub-basin and
soils. Thus, the high mineral potential of the Curnamona Province for Broken Hill style and Fe-oxide
Cu-Au style mineralisation is likely to be realised under cover. Traditional mineral exploration
methods used in areas of outcrop face challenges when applied to areas of transported cover,
especially where it is greater than a few metres thick, and new exploration tools are needed.
Hydrogeochemistry is one of these novel tools, and here we demonstrate its potential to fingerprint
groundwater-mineralisation interaction through the application of major and trace element
concentrations, stable and radiogenic isotope signatures and reaction-transport modelling.
We have collected about 350 groundwater samples from existing boreholes in the Curnamona
Province, both in areas of outcrop in the ranges and in areas of cover in the surrounding basins. The
samples have been analysed for a comprehensive suite of major, minor and trace elements as well as
for several stable and radiogenic isotopes (oxygen and deuterium on water; sulfur and oxygen on
sulfate; carbon on bicarbonate; chlorine, strontium and lead). The results indicate that the
groundwater composition is affected by a number of processes: evaporation, evapotranspiration,
mixing, precipitation/dissolution and oxidation/reduction. The impact of each of these major
processes is teased out using conservative tracers (chloride, bromide), isotopes and geochemical
modelling. To develop hydrogeochemistry into a useful tool for exploration requires a number of
steps in the analysis of the data.
1. We calculate an index of sulfur excess to gain knowledge of samples that contain more sulfur
than can be accounted for by evaporation or mixing.
2. The sulfur isotopic composition of dissolved sulfate enables us to discriminate between the
samples containing sulfur of meteoric origin versus sulfur from a Broken Hill type sulfide
mineralisation. Further, the oxygen isotopic composition of the dissolved sulfate can shed light
on where sulfide oxidation occurred within the hydrogeological system.
3. Reactive transport modelling is used to quantify the decay in trace element concentrations
released by oxidation of sulfides (chalcopyrite, sphalerite, galena) as water moves through an
aquifer containing clay minerals and iron oxyhydroxides, thus giving insight into how far from a
sulfide body one may expect trace element concentrations of a given level.
4. Finally, for those samples where we have obtained lead isotope composition, we can evaluate the
specific mineralisation type (Broken Hill type, Thackaringa type, etc) that is compatible with the
signature in the groundwater or make inferences regarding mixing with background lead.

This approach can help delineate areas of interest for subsequent mineral exploration work under
sedimentary cover (area selection), rank geophysical anomalies and vector towards mineralisation.

1. CRC LEME, C/O- Geoscience Australia, GPO Box 378, Canberra ACT 2601.
2. CRC LEME, C/O- Department of Earth and Marine Sciences, Australian National University, Canberra ACT 0200.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 219

The FalconTM Airborne Gravity System as an Exploration Driver
in Australia
P Harman1, D Isles2 and C Anderson3

ABSTRACT

The Falcon airborne gravity gradiometer system is a new and unique exploration tool developed by
BHP Billiton in the late 1990s. It commenced full operational exploration in diamond and base metal
applications overseas during 2000 and Australia-based surveys began in late 2002. The system
provides resolution in gravity data approaching that of detailed ground surveys, but covers much
larger areas than practically achievable on the ground, more rapidly and more economically. Because
of this precision, speed and cost efficiency, Falcon makes gravity a front line exploration tool,
particularly in exploration for diamonds, iron ore and certain types of base metal deposit.
BHP Billitons Falcon implementation strategy involves applying Falcon in joint venture
arrangements where BHP Billiton retains some form of buy-in option. In Australia, Gravity Capital
and associated companies have arrangements with BHP Billiton which allow Gravity to form joint
ventures based around Falcon applications. To date, joint venture Falcon surveys have been
completed at Broken Hill, in the Kimberley region of WA, in the Gawler Craton, in the Northern
Territorys microdiamond field, in the Victorian slate-belt and in the Gippsland Basin, where the
system has proved invaluable in defining structural targets for hydrocarbon exploration.
At Broken Hill, through the initiatives of the CRC for Predictive Mineral Discovery and the
Geological Survey of NSW, a survey of 1000 km2 was conducted and more than ten excellent base
metal targets were identified, in addition to a wealth of new geological information. The data is now
publicly available through the NSW Geological Survey and analysis and integration of the new and
detailed gravity information with pre-existing data is providing new leads in geological
understanding and exploration. More importantly, the process of testing the exploration targets is
well under way.
In the Gawler Craton, during BHP Billitons involvement in the joint venture covering the
Prominent Hill discovery, Falcon was flown and compared to the substantial pre-existing sets of
detailed ground gravity. The Prominent Hill survey illustrated the value of Falcons rapid coverage
and its significant other advantages over ground-based surveying. Gravity has also flown a large part
of its Tarcoola project area and the Cultana area in joint venture with Eagle Bay Resources.
In the northern Australian diamond fields, Gravity, on behalf of its associated company, Diamond
Mines Australia, has implemented surveys over ten areas spread over four different joint ventures.
Excellent results have been achieved over the major known kimberlite and lamproite pipes and
numerous high ranking target anomalies have been identified for testing during the 2004 field season.
In the Victorian slate belt, Gravity has formed a JV with a private company to fly Falcon over
the inferred northern extension of the Bendigo goldfield. This area of thin cover shows a number of
structural aspects which make it highly amenable to gravity surveys. In addition to the primary gold

1. MAusIMM, Managing Director, Gravity Capital Limited, Level 7 Exchange Tower, 530 Little Collins Street, Melbourne Vic 3000.
2. MAusIMM, Consultant, PO Box 229, West Perth WA 6005.
3. FAusIMM, C/O- Gravity Capital Limited, Level 7 Exchange Tower, 530 Little Collins Street, Melbourne Vic 3000.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 221

P HARMAN, D ISLES and C ANDERSON

potential, the area is likely to host palaeochannels which could contain significant alluvial gold.
Palaeochannels are known to be readily detectable from Falcon data.
In the Gippsland Basin, BHP Billiton has covered the Snapper and Marlin fields and Gravity, in
conjunction with Lakes Oil, has conducted a number of surveys, both offshore and onshore. The data
gathered has rapidly provided a new and quite detailed view of the structural environment to the
extent that a number of wells are planned to test targets based mainly on the gravity.
Falcon continues to prove its capabilities in breaking new ground in both mature exploration
provinces and in grass roots applications. Its prime quality is its ability to provide exploration focus
in the context of the new geological information it provides.

222 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

Automated Mineralogical Logging of Core From the Emmie Bluff,
Iron Oxide Copper-Gold Prospect, South Australia
J Huntington1, A Mauger2, R Skirrow3, E Bastrakov3, P Connor1, P Mason1, J Keeling4,
D Coward1, M Berman5, R Phillips1, L Whitbourn1 and P Heithersay2
ABSTRACT
A new rapid spectroscopic core and chip logging and imaging system
developed by CSIRO was evaluated by PIRSA and GA on a wide variety
of mineralised properties in South Australia, including the Emmie Bluff,
iron oxide copper-gold-style prospect on the Stuart Shelf. During a
three-month campaign in South Australia in 2003 the system logged over
50 000 metres of core from the State Core Library, company prospects
and mines.
The new system uses automated core tray handling, continuous visible
and infrared spectroscopy, and digital imaging, to characterise and
identify dominant mineral species on core, chips and pulps, at spatial
resolutions of ~1 cm (spectral data) and ~0.1 mm (image data). The
system can log up to 1000 m of core per day and several thousand metres
of drill chips. Mineralogy is pre-interpreted using specialised
identification software trained on a selected suite of minerals showing
characteristic absorption features within the measured spectral range.
On core from the Emmie Bluff iron oxide copper-gold prospect,
470 km NW of Adelaide and 80 km SSE of the Olympic Dam Cu-U-Au
deposit, the spectral mapping system was able to characterise a number of
alteration zones enveloping the sub-economic Cu-Au mineralisation.
These consist of distal white mica phases showing decreasing Al content
towards mineralisation, a zone of increased iron oxide content in the
hangingwall to mineralisation, and a zone proximal and coincident to
Cu-Au mineralisation that is enriched in chlorite with variable Mg/Fe
chemistry. Additionally the results show background spectral
characteristics that mimic the magnetite distribution.
The speed and objectivity offered by such a system, combined with
assay data and visual log descriptions, provide tremendous scope to
increase the knowledge gained from, and to add value to, the expense of
drilling, particularly at the prospect evaluation stage. Geologists may log
samples with increased confidence and efficiency by focussing attention
on issues that demand their particular skills, whilst leaving routine
mineralogical interpretation of hard-to-recognise alteration phases to the
new system. All these factors may lead to higher quality logs and reduce
the need for expensive re-logging. Early mapping of alteration zones,
some often larger than previously detected, may thus provide more
targeted and hence more cost effective drilling programs, as well as
providing an early indication of geotechnical and/or metallurgical issues
in an ore system, where these have a mineralogical expression (eg
deleterious swelling clays or talc).
INTRODUCTION
With a substantial proportion of South Australias prime
exploration terrain covered by regolith or consolidated
sediments, increased attention is being directed towards drilling
and to extracting maximum information and value from
pre-existing and newly drilled samples.
1. CSIRO Exploration and Mining (CEM), PO Box 136, North Ryde
NSW 1670.
2. Primary Industry and Resources South Australia (PIRSA), GPO Box
1671, Adelaide SA 5001.
3. Geoscience Australia (GA), GPO Box 378, Canberra ACT 2601.
4. MAusIMM, Primary Industry and Resources South Australia
(PIRSA), GPO Box 1671, Adelaide SA 5001.
5. CSIRO Mathematical and Information Sciences (CMIS), PO Box
136, North Ryde NSW 1670.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
With this in mind, PIRSA, GA and the Cooperative Research
Centre for Landscape Environments and Mineral Exploration
(CRC LEME), in collaboration with CSIRO, used a new
spectroscopic core and chip logging and imaging system (the
HyLogger), developed by the CSIRO, to investigate a wide
variety of mineralised properties and regolith settings in South
Australia. This included the Emmie Bluff, iron
oxide-copper-gold-style prospect in the eastern Gawler Craton
(Figure 1). Emmie Bluff is located 80 km SSE of the world-class
Olympic Dam Cu-U-Au mine.
The HyLogging system automatically images and spectrally
maps the alteration and host rock mineralogy of the core,
bringing to light potential new vectors to mineralisation and
knowledge to refine ore system models.
As part of the Australian Mineral Industries Research
Association (AMIRA International) Project P685, eight diamond
drill holes from Carpentaria Explorations 1989 Emmie Bluff
drilling program were scanned with the prototype HyLogger in
April and May 2002. This represented a total of 2157 m of various
sized core from 347 trays and collected 319 899 spectra. During a
three-month repeat campaign in September 2003 the improved
system was again used to log over 50 000 metres in three months
from the State Core Library and company prospects and mines.
The 2003 campaign included a second scanning of one of the
Emmie Bluff drill holes, SAE6, described here as typical of the
core logging process and the alteration characteristics of many
iron oxide Cu-Au systems in the eastern Gawler Craton.
CORE AND CHIP LOGGING TECHNOLOGY
The HyLogging System (Figure 2) comprises an integrated suite
of spectroscopic, imaging, lighting and materials handling tools
for the semi-automatic scanning, logging and interpretation of
the mineralogy from drill core, drill chips and blast hole
powders. Core is continuously sampled spatially and spectrally
from end-to-end, generating a rich, statistically-robust sample
set. This is achieved by moving core and chip trays beneath the
fixed spectrometer using an x/y robotic table. Depending on core
diameter and tray size one tray can be scanned in three to five
minutes. Over 1000 m of core have been measured in one day,
though a more typical working average is 600 m. The core
remains in its original trays and requires minimum sample
preparation, other than being dry and clean. Six chip trays can be
handled in the place of one core tray, permitting 120 m of 1 m
composite samples to be measured in five minutes.
Key characteristics of the system are shown in Table 1. The
current version of the HyLogger is suited to the recognition of
iron oxides, hydroxyl, sulfate and carbonate-bearing minerals
and chemical variants of many of these. The system is not
designed for ore bearing minerals per se, though when massive
some phases can be mapped. A next generation system is being
designed for non-OH-bearing silicates, such as feldspars, quartz,
pyroxenes, olivines and garnets.
The HyLogger currently uses three bore-sighted grating
spectrophotometers spanning the visible and near infrared (VNIR)
and shortwave infrared (SWIR) spectral regions. These measure
calibrated radiance that is converted to spectral reflectance via
calibration to a Spectralon standard. The spectrometer views a
single spectral profile along the length of the core, straddling
approximately one third of the width of the core.
223
J HUNTINGTON et al

0
0
13500 13630 0
13800
0
29

Cairn Hill
MT WOODS 100 km
INLIER
Manxman
Joes Dam
Prominent Hill
0
30

OLYMPIC DAM

Acropolis
Hiltaba Suite granitoid Wirrda Well
(1595-1570 Ma)
MESOPROT.

Mafic intrusions (> ?1570 Ma) 310

Emmie Bluff
Gawler Range Volcanics WOOMERA
(mainly felsic, ~1590 Ma)

Mafic Gawler Range Volcanics

Wallaroo Gp - metasedimentary
PALAEOPROT.

& metavolcanic rocks; Moonta

Porphyry, ~1750 Ma)
Donington Suite granitoids, 0
~1850 Ma 32
Hutchison Gp and other Palaeo-
proterozoic metasedimentary PORT
rocks; BIF AUGUSTA
Undifferentiated Archaean to Menninnie
Palaeoproterozoic rocks Dam Pb-Zn
Weednanna Au
Fault
330
Coastline

Alteration & Cu-Au

Magnetite - calcsilicate - alkali
feldspar and/or magnetite -
biotite + sulfides
Hematite - sericite - chlorite - Wallaroo 0
carbonate + Cu-Fe sulfides 34
0
13630
Magnetite assemblages overprinted Moonta
by hematite assemblages

Other deposit or prospect

0
13800
FIG 1 - (A) Location of the Gawler Craton and Olympic Cu-Au province, (B) Basement geology and iron oxide Cu-Au systems of the
Olympic Cu-Au province (modified from Skirrow et al, 2002).

224 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 AUTOMATED MINERALOGICAL LOGGING OF CORE FROM THE EMMIE BLUFF, IRON OXIDE COPPER-GOLD PROSPECT
FIG 2 - View of the prototype HyLogging system.
TABLE 1
Nominal HyLogging system characteristics used at Emmie Bluff.
Spectrometer spectral range 400 - 2500 nm
Nominal spectral resolution 16 nm
No of spectral channels 196
Spectrometer spatial resolution 8 mm * 8 mm
No of observations (spectra)/metre 125
Linescan camera spatial resolution 0.1 mm
Robotic table scanning rate 60 mm/second
Typical daily acquisition 600 m
Laser profilometer resolution 0.25 mm
Light source Quartz halogen
The core imaging system is a digital three-colour (red, green,
blue), linear array, linescan camera that builds up a continuous
image of the core, line by line, as the tray passes beneath. The
full width of the core and some of the adjacent core tray section
is imaged. The camera is calibrated on each core tray pass to
ensure consistency with time. This digital imaging system
provides a very high-resolution historical record of the drilled
material suitable for textural and some geotechnical
interpretation and long-term archiving. The image data are
compressed and subsequently decompressed in real-time during
analysis and visualisation. Multiple views of the core imagery
are provided, for example the high-resolution core imagery is
used to create standard format virtual core tray images,
equivalent (though of much higher resolution) to those often
taken of core trays with a conventional digital camera. The
virtual core tray images are also reformatted into single mosaics
of a complete drill hole.
The system incorporates a laser profilometer used to measure
the physical condition of the core as well as fractures and breaks
to assist in geotechnical assessment and for control of other
aspects of the system.
The spectrometer, image and profilometer data are all captured
simultaneously with a single traverse of the core tray. Though the
spatial resolution of the various data sets differs, all are registered
in software and can be interrogated on the analysis screen.
The HyLogging hardware is complemented by special
software (TSG-Core) for the analysis, mineralogical
interpretation and simultaneous visualisation of the spectral,
laser and image data. Mineralogical interpretations of those
minerals that are spectrally active over the range of the
spectrometer are calculated using an updated version of The
The Spectral Geologist (TSG)
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Spectral Assistant (TSATM) Algorithm (Berman et al, 1999),
along with their relative proportions and an interpretation error.
User defined functions for the mapping of selected minerals
species can be readily computed and saved for future use. All
data, spectral, mineralogical, core images, imported assays, etc
can be manipulated and viewed simultaneously on screen, or
exported into third party mine planning and visualisation
packages.
Though the HyLogging system comprises a sophisticated mix
of new and existing technologies, the cost of using the system is
targeted at less than ten per cent of average drilling costs for
diamond core, while providing very high resolution continuous
sampling at a fraction of that used by most other methods. We
suggest that considerable value can be added to drilling programs
using this technology through a mix of increased objectivity,
time and tasks saved, reduced re-logging, and new knowledge
leading to improved ore system models.
DATA ACQUISITION PROCEDURE
Among the ancillary information acquired for SAE6 were the
original geologists log recording lithology and some alteration,
assay data for the following elements: Cu, Zn, Au, Pb, Ba, U3O8,
and petrophysical information of magnetic susceptibility and
specific gravity.
Sections of the core comprise very dark material with
apparently low SWIR active mineral content. Over these
intervals the high proportion of magnetite, quartz and feldspar
make for a less than ideal suite for VNIR-SWIR hyperspectral
measurement. However, against initial predications, significant
mineralogical patterns that can be correlated with a
mineralisation model exist in the recorded spectra.
SAE6 passes through some 840 m of sedimentary cover
(Pandurra Formation) and was thus HyLogged only from 837.6
to 1200.0 m, generating some 58 890 samples.
Prior to scanning the Emmie Bluff core was vacuumed, as it
had been in storage for some time, and also all tray dimensions
and core depths as recorded on and in the core trays were logged
into an MS Access database. These were subsequently used to
geocode every spectral observation. Efficient materials handling
was paramount to optimum workflow and as a consequence over
1000 m was logged on some days, though 600 m was more
typical. It was found that the preparation processes could keep
ahead of scanning.
Since the logger measures the spectra of everything in a core
or chip tray, including tray materials (galvanised iron in the case
of Emmie Bluff), wooden or plastic depth markers, spectra of
these are included in the analysis software so they can be masked
out during analysis.
GEOLOGICAL SETTING OF EMMIE BLUFF
(as previously determined)
The Emmie Bluff iron oxide Cu-Au prospect was discovered by
drilling of coincident gravity and magnetic anomalies, in basement
beneath 800 - 900 m thickness of Mesoproterozoic,
Neoproterozoic and younger cover. This Cu-Au prospect is one of
many in the Olympic Cu-Au province (Skirrow et al, 2002), a
>500 km long belt containing iron oxide Cu-Au (IOCG)
mineralisation and extending from the Moonta-Wallaroo district
through the Olympic Dam region to the Mt Woods Inlier (Figure
1). The northern extent of the province is unknown. The principal
basement lithological units in the province (Jagodzinski, 2004) are
Palaeoproterozoic granitoids of the Donington Suite (~1850 Ma)
and metasedimentary rocks of the Wallaroo Group (<1850 Ma to
~1740 Ma), and early Mesoproterozoic Hiltaba Suite granitoids
and Gawler Range Volcanics (~1575 - 1590 Ma). All of these
lithologies are hydrothermally altered, to varying extents, and
three main alteration assemblages are recognised that formed at
~1575 - 1590 Ma (Skirrow et al, 2002; Raymond et al, 2002).
225
J HUNTINGTON et al

These are: containing chalcopyrite-pyrite-bornite-covellite, to a lower zone of

CAM Magnetite-bearing assemblages with calcsilicates
(actinolite, clinopyroxene), alkali feldspar (dominated by
K-feldspar in the Olympic Dam region; albite elsewhere),
and minor pyrite, apatite, chalcopyrite, allanite and rare
pyrrhotite and titanite;
MB Magnetite-bearing assemblages with biotite and minor
pyrite, chalcopyrite, albite, titanite, pyrrhotite, which are
known mainly from the Moonta-Wallaroo district and Mt
Woods Inlier; and
HSCC Hematite-sericite-chlorite-carbonate assemblages, in
places associated with significant chalcopyrite-pyrite bornite
chalcocite mineralisation (eg Olympic Dam, Prominent
Hill); this assemblage generally overprints magnetite-bearing
assemblages.
In the Emmie Bluff area hydrothermal alteration and
mineralisation are hosted by Donington Suite granitoids,
Wallaroo Group metasedimentary rocks, and Gawler Range
Volcanics. The Wallaroo Group in this area comprises an upper
unit of thin-bedded argillite and carbonate-bearing strata, and an
underlying unit of meta-arkosic (quartz-K-feldspar-muscovite)
rocks.
Two major zones of Fe-oxide rich alteration have been located
about 3 km apart, as follows (Gow et al, 1994):
The northern alteration zone, intersected by SAE6 and
comprising stratabound magnetite-hematite-chlorite-sericite
(CAM overprinted by HSCC), is up to ~150 m thick, ~3 km
diameter, and is hosted mainly by the thin-bedded unit of the
Wallaroo Group and by brecciated granitoid.
The southern, discordant alteration zone contains veins and
replacements within rhyodacite of the Gawler Range
Volcanics, intersected at more than 1 km depth. Veins contain
skarn-like, high-temperature mineral assemblages (CAM)
with combinations of clinopyroxene, amphibole, magnetite
(<10 per cent vol), garnet, quartz, calcite, K-feldspar, pyrite,
and allanite. Copper-iron sulfides are rare.
ALTERATION AND MINERALISATION IN DRILL
HOLE SAE6 (as previously determined)
SAE6 was selected from the eight Emmie Bluff drill holes
scanned by HyLogger as representative of the range of alteration
and mineralisation and typical of their hyperspectral
characteristics. Some details of this hole are held on open-file at
PIRSA, and this and other holes at Emmie Bluff have been part
of recent metallogenic studies by Geoscience Australia (eg
Skirrow and Bastrakov, 2002).
Grades of up to 2.8 per cent Cu and 0.6 g/t Au are reported from
the northern alteration zone intersected by SAE6. The iron oxide
Cu-Au deposit in this area is crudely zoned vertically from an
upper zone of intense hematite-chlorite-quartz or hematite-
sericite-quartz alteration (with the occurrence of chlorite or
sericite controlled by the particular lithology of the protolith) and
magnetite-chlorite-pyrite altered metasedimentary rock with lower
abundances of Fe and Cu sulfides. Gow et al (1994) interpreted
the breccias as tectonic in origin, related to reverse reactivation of
a nearby regional fault. Hematisation of pre-existing magnetite,
and accompanying Cu-Au mineralisation, were also envisaged to
have occurred during the fault reactivation.
The paragenetic sequence of hydrothermal minerals observed
in drill hole SAE6 is similar to the generalised two-stage
sequence observed in many of the prospects in the Olympic
Cu-Au province. It commenced with magnetite and minor pyrite-
chalcopyrite quartz chlorite titanite deposition, followed by
hematite-chlorite/sericite-rutile alteration accompanied by
further precipitation of chalcopyrite. Unlike the southern zone,
however, no amphibole or feldspar alteration has been observed
in SAE6.
In the thin-bedded Wallaroo Group metasedimentary rock unit,
semi-massive to massive magnetite is preserved in discontinuous
bands and laminae, which appear to have been tectonically
disrupted. Some discordant Fe-oxide veins have Fe-oxide-rich
alteration aureoles that extend along compositional layering in
the host rocks. We interpret this magnetite as selective
replacements of laminated metasedimentary rocks. Remaining
parts of the rock are predominantly quartz and chlorite. Minor
pyrite and chalcopyrite are intergrown with the hydrothermal
magnetite, quartz and chlorite. The underlying meta-arkosic unit,
and overlying meta-granitic rocks, were less intensely Fe-oxide
altered in SAE6.
The second paragenetic stage involved martite replacement of
pre-existing magnetite, and chalcopyrite replacement of pyrite,
accompanied by chlorite, within the thin-bedded
metasedimentary unit. In the meta-arkosic and meta-granitic
rocks this alteration resulted in sericitisation of feldspars and
only minor chlorite development. In the upper part of SAE6,
where almost all magnetite was oxidised, the martite was
recrystallised to fine grained anhedral hematite. Additionally in
this zone a subordinate amount of fine-grained platy hematite,
red dusty hematite, and vughs were developed probably late
within the second paragenetic stage. Highest abundances of
chalcopyrite are broadly spatially associated with the zones of
intense hematite development, but in detail chalcopyrite appears
to be less abundant in vughy domains. Most bornite and covellite
occurs as replacements of chalcopyrite.
MINERALOGICAL RESULTS
Mineralogical interpretation was carried out using the TSG-Core
software and manual spectral interpretation, backed up by
reference to the core imagery. Those mineral phases found to
provide the most meaningful geological information in this
environment included the distribution of AlOH white micas,
muscovite and phengite, MgOH phyllosilicates Fe- and
Mg-chlorites, along with iron oxides hematite and goethite, and
magnetite. Additional signatures of fluid inclusion-rich quartz
and minor carbonate were also noted. Selected mineralogical
results are shown in Figure 3.

FIG 3 - Alteration mineralogy of SAE6 as defined by a suite of spectrally defined mineralogical scalars. For details see text. All plots show
depth along the x axis from 820-1200 m. Boundaries of the five main rock/alteration types are shown by vertical dashed lines and
numbered at the top and bottom: 1 - Conglomerate (Pandurra formation), 2 - Meta-granite (Donington Suite), 3 - Thin-bedded,
hematite+chlorite + sulfide-rich meta-sediments, 4 - Thin-bedded magnetite+chlorite+pyrite-rich meta-sediments, 5 - Meta-arkose (3 - 5 all
Wallaroo Group). (A) The relative absorption intensity of iron oxides, showing their maximum in the hanging wall. Each dot represents one
10 mm sample. (B) Histogram of all white mica-bearing samples per 3 m interval. (C) Histogram of only phengitic mica samples per 3 m
interval. (D) AlOH absorption wavelength plot showing the progression from muscovite (blue to light green; samples) to phengite (orange to
red samples) approaching the mineralised zone. (E) Histogram of all chlorite-bearing samples per 3 m interval. Note the three peaks.
(F) Moving average of sample albedo. Brightest intervals are felsic rocks, intermediate are hematite-rich zones and darkest is the magnetite
zone. (G) Log of the magnetic susceptibility. (H) Copper assays coloured from 0 (blue) to 9000 ppm (red). (I) Fe2+ spectral index mapping
the chlorite subgroup that best correlates with the highest copper assays. Coloured by Cu assay value. (J) 765/951 spectral ratio
highlighting the magnetite and hematite zones coloured by rock type. Highest ratio values correspond to higher magnetic susceptibility.

226 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 AUTOMATED MINERALOGICAL LOGGING OF CORE FROM THE EMMIE BLUFF, IRON OXIDE COPPER-GOLD PROSPECT

A Iron Oxide F Albedo

1 2 3 4 5 1 2 3 4 5

B Total Mica G Magnetic Susceptibility

C Phengite H Copper

D Mica Wavelength
I Fe 2+ Index

E Chlorite J 760/950 Ratio

1 2 3 4 5 1 2 3 4 5

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 227

J HUNTINGTON et al
Iron oxides
Figure 3a shows the relative strength of the iron oxide absorption
features (Figure 4d) due primarily to hematite. These
progressively increase in their intensity through the meta-granitic
and reach their maximum development near 950 m in the
immediate hanging wall to the mineralised zone. Iron oxides
(principally hematite) are present within the mineralised zone as
discussed above, but as they are less continuous and mixed with
sulfides their relative expression is not as spectrally obvious.
White micas
Figure 3b illustrates that total white mica is concentrated in the
meta-granite and meta-arkose units, but generally is
progressively depleted towards the contacts of these units with
the sulfide-rich central units, where micas are entirely absent.
Figure 3c however separates out only the phengitic micas (Figure
4b), those depleted in aluminium relative to magnesium or iron.
Significantly these show a progressive increase relative to total
white mica towards the central altered and mineralised zone.
This is best illustrated in Figure 3d that maps, on a
sample-by-sample basis, the specific and progressive wavelength
shift that is caused by the Fe/Mg phengitic substitution of
aluminium. Analysis of the core imagery and this index indicates
that this alteration style extends beyond the meta-granite and
meta-arkose some way into the central meta-sedimentary units.
Chlorites
Figure 3e indicates that chlorite is inversely correlated with white
mica. It occurs primarily in the central thin-bedded
meta-sedimentary units, but overlaps into the adjacent
meta-granite and meta-arkose on either side. The chlorite
distribution forms three pronounced peaks on Figure 3e. The gap
between the second and third peaks coincides with the zone of
strongest magnetic susceptibility (magnetite development) shown
on Figure 3g.
Variations in chlorite composition are present in the copper
mineralised zone, ranging from Fe-rich, through intermediate
compositions to Mg-rich. When compared to the magnetic
susceptibility it is apparent that the chlorite higher in the hole
(the first two chlorite peaks on Figure 3e), which is associated
with the higher Cu values, correlates with the hematite-dominant
section of the mineralised zone. Where magnetite dominates
chlorite is absent but reappears lower down where the magnetic
susceptibility is still elevated and magnetite is the dominant
iron-ore species.
Detailed examination of chlorite spectra indicates differences
between those in the hematite zone above the magnetite zone
(and correlated with the copper zone) and those below the
magnetite zone. This distinction involves the slope of the
reflectance below 2050 nm normally regarded as a measure of
Fe2+ content of chlorites (Figure 4c), biotites and ferroan
carbonates. A spectral index that measures this Fe2+ slope
(Figure 3i) produces an index that, above a certain threshold,
corresponds closely with the copper-rich, hematised interval
(Figure 3h). Future studies will focus on unravelling further the
relationships between the chlorite compositional variations and
the Fe-oxide minerals.
Magnetic susceptibility
It was found that a variety of spectral indices correlate with
magnetic susceptibility (Figure 3g). Not surprisingly the highest
magnetic susceptibility equates to the lowest albedo rocks
(Figure 3f) and intermediate magnetic susceptibility to
intermediate albedo. The zone of intermediate magnetic
susceptibility also equates to the zone of highest copper grades.
A spectral ratio of 765/951 nm (Figure 3j) was also found to
highlight variations in the magnetite/hematite ratio.
228 Adelaide, SA, 19 - 22 September 2004
DISCUSSION
The Fe-oxide normalised intensity plot indicates maximum
Fe-oxide contents in the hangingwall of the copper mineralised
zone, whereas maximal magnetic susceptibility and magnetite
content occurs in the footwall to mineralisation. We have
visually and petrographically confirmed that the zone of
Fe-oxide normalised intensity maxima is dominated by hematite
as martite and other forms described above. We propose that
magnetite was once extensively developed in the upper parts of
the hydrothermal system intersected by SAE6, but was intensely
hematised down to a depth of approximately 1010 m in SAE6.
Gow et al (1994) used oxygen isotope geochemistry to
demonstrate a major involvement of meteoric waters in
hematisation at Emmie Bluff. Bastrakov et al (2003)
demonstrated the sedimentary origin of sulfur for the bulk of
sulfide mineralisation within SAE6. These data are consistent
with introduction of copper by fluids of exogenic origin (Gow et
al, 1994). An alternative model, including introduction of copper
in the meteoric fluid, and leaching and redistributed of
pre-existing high-temperature copper mineralisation, was
presented by Bastrakov et al (2003), and will be discussed in
detail elsewhere.
The HyLogger results for SAE6 are significant because this
drill hole is representative of the locally ore-grade (but
sub-economic) copper prospects in the central Olympic Cu-Au
province. The results provide insights into the zoning and
mineralising processes of such iron oxide copper-gold systems.
Copper mineralisation in SAE6 is best developed in
hematite-chlorite zones in the hangingwall of magnetite-chlorite
alteration zones, flanked by phengitic white mica alteration. To a
significant extent the broad distribution of chlorite and white
micas is determined by host rock composition, with chlorite
dominant in thin-bedded metasedimentary rocks and muscovite
the dominant hydrothermal white mica in quartzo-feldspathic
host rocks. The systematic increase in mean mica wavelength
(Al-depletion) as well as subtle chemical variations within the
chlorite populations towards copper mineralisation nevertheless
appears to be independent of host rock composition, and shows
promise as a possible vector-to-mineralisation.
The other seven holes scanned from the Emmie Bluff district
display similar patterns to SAE6 (Mauger et al 2002) with regard
the chemical trends in the white micas on approach to
mineralisation, and the absence of AlOH micas from mineralised
sections coinciding with the appearance of chlorites. Natural
variability and lack of assay data mean that some of the patterns
were not as clear as those displayed by SAE6, however there are
indications of systematic variations in chlorite chemistry. Further
work is required to confirm these findings with other analytical
techniques.
CONCLUSIONS
As dependence on drilling subtly expressed or blind mineralised
systems increases, the automated HyLogging system has
demonstrated that increased objectivity, increased value and new
geological knowledge can be rapidly acquired to offset the high
cost of modern exploration.
The spectrally determined mineralogy, and indeed many
spectral indices, correlate directly or inversely with known
lithological, petrological, petrophysical and geochemical
properties of the rocks in SAE6. Furthermore new trends in
mineral chemistry have been brought to light, specifically the
spatial trends in white mica, chlorite and iron oxides that may not
be totally explained by lithological variations and will certainly
direct further research. The trends indicated hold promise for new
ways to evaluate these types of iron oxide, Cu-Au systems.
PACRIM 2004
 AUTOMATED MINERALOGICAL LOGGING OF CORE FROM THE EMMIE BLUFF, IRON OXIDE COPPER-GOLD PROSPECT

FIG 4 - Typical HyLogger core images, reflectance spectra and key spectral characteristics of the main mineral phases
encountered at Emmie Bluff.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 229

J HUNTINGTON et al
ACKNOWLEDGEMENTS
The prototype core logger was developed in 2001/2 with the
grateful support of nine company and state sponsors through
AMIRA International. The assistance of Brian Logan, Georgina
Bradley and the staff of the Glenside Core library in Adelaide is
gratefully acknowledged.
REFERENCES
Bastrakov, E N, Skirrow, R G, Barovich, K and Davidson, G, 2003. Nd-
and S-isotope signatures of South Australian Fe-oxide Cu-Au
systems: fingerprinting rich cousins vs poor cousins, in
Proceedings International Geochemical Exploration Symposium,
Dublin, Ireland, August, pp 46-47.
Berman, M, Bischof, L and Huntington, J F, 1999. Algorithms and
software for the automated identification of minerals using field
spectra or hyperspectral imagery, in Proceedings 13th International
Conference on Applied Geologic Remote Sensing, Vancouver, Vol 1,
pp 222-232.
Gow, P A, Wall, V J, Oliver, N H S and Valenta, R K, 1994. Proterozoic
iron oxide (Cu-U-Au-RRE) deposits: Further evidence of
hydrothermal origin, Geology, 22:633-636.
230 Adelaide, SA, 19 - 22 September 2004
Jagodzinski, E A, in press. Compilation of SHRIMP U-Pb
geochronological data for zircons from the Olympic Domain, Gawler
Craton, South Australia, 2001-2003, Geoscience Australia Record,
2004.
Mauger, A J, Huntington, J F, Mason, P, Scott, K, Coward, D, Connor, P,
Heithersay, P, Keeling, J and Bradbury, G, 2002. Emmie Bluff,
CSIRO Exploration and Mining Investigation Report No 999R,
Unpublished AMIRA Report, 21 p.
Raymond, O, Fletcher, I and McNaughton, N, 2002. Copper-gold mineral
systems in the south-eastern Gawler Craton another Mt Isa eastern
succession? Geological Society of Australia Abstracts, No 67, p 69.
Skirrow, R G and Bastrakov, E, 2002. Olympic Cu-Au province
Regional alteration and prospects, in Drill Core Display, Gawler
Craton 2002: State of Play Conference, Adelaide, Primary Industries
and Resources South Australia Report Book 2002/028 (Compilers: G
M Ferris and S Daly).
Skirrow, R G, Bastrakov, E, Davidson, G, Raymond, O L and Heithersay,
P, 2002. The geological framework, distribution and controls of
Fe-Oxide Cu-Au mineralisation in the Gawler Craton, South
Australia: Part II: alteration and mineralisation, in Hydrothermal Iron
Oxide Copper-Gold and Related Deposits: A Global Perspective,
(Ed: T M Porter) Vol 2, pp 33-47 (PGC Publishing: Adelaide).
PACRIM 2004
New Ways to Explore Through the Regolith in Western
New South Wales
K G McQueen1 and A McRae1
ABSTRACT
Much of western New South Wales is characterised by an extensive
regolith formed during more than 60 million years of exposure,
weathering and erosion. This cover includes intensely weathered in situ
regolith and superimposed transported materials of variable thickness,
commonly in complex palaeochannel systems. Mineral exploration in this
region has been considerably hampered by the cover, which can
complicate or mask geochemical anomalies and also obscure geophysical
signatures. A detailed multi-disciplinary study of part of this region,
centred on the Cobar-Girilambone area has demonstrated a methodology
for exploring through this regolith. The approach involves
regolith-landform mapping to establish the main landscape and
palaeolandscape elements and to determine the nature and spatial
occurrence of different regolith materials. By combining knowledge of
the surface distribution of different regolith-landform units with available
3D information on the regolith and geophysical data it is possible to
construct simplified derivative maps or exploration go maps, which can
be used to advise appropriate exploration strategies or help interpret
acquired exploration data. This study has also established the major
regolith controls on geochemical dispersion and determined key
background element associations. These regolith-related element
associations can result in highly variable background levels for many
trace elements and need to be understood and factored into the analysis of
geochemical data for different regolith sampling media. With better
knowledge of the regolith it is possible to select the most appropriate
exploration techniques and sampling media for a given area, as well as
apply appropriate normalisation procedures to the acquired data.
Exploration success can be greatly improved by using many existing
techniques in regolith smart ways.
INTRODUCTION
Most discovered ore deposits in western New South Wales have
been found at or close to the surface in erosional landform
settings. A small number have been discovered obscured by
relatively thin transported cover (eg Elura) or in areas of deeper,
partly leached in situ regolith (eg Tritton), usually by using
geophysical techniques. A small number have also been
discovered as deeper extensions to known, smaller outcropping
deposits (eg the Peak deposit at Cobar) or adjacent to deeply
mined deposits (eg Perseverance near the Peak and QTS near the
CSA deposit at Cobar). There are large tracts of prospective
terrain both in deeply weathered erosional settings and
depositional settings where no discoveries have been made. The
challenge is to develop methodologies and techniques to explore
these areas effectively.
Successful exploration in regolith-dominated terrain requires a
good understanding of the regolith-landform setting, particularly
knowledge of the origin, thickness and physio-chemical
architecture of the regolith. To apply geochemical exploration
techniques it is also important to understand the history of
weathering and degree of chemical leaching, dispersion
pathways of target and pathfinder elements in the regolith and
the likely host components for dispersed elements. A detailed
study has been conducted in the Cobar-Girilambone area of
western New South Wales to examine these various parameters
and devise some guidelines for improved mineral exploration in
the region. In common with much of western New South Wales,
the Cobar-Girilambone area contains a wide variety of regolith
types of highly variable thickness (up to 200 m). There are
1. CRC LEME, Australian National University, Canberra ACT 0200.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
significant areas of transported regolith with shallow to moderate
thickness (up to 60 m). The area also contains a number of
known polymetallic base metal and gold deposits providing the
opportunity to examine regolith characteristics and geochemical
dispersion processes at both mineralised and background sites.
Research in this area was part of a collaborative project between
the Cooperative Research Centre for Landscape Environments
and Mineral Exploration and the New South Wales Department
of Mineral Resources. A key feature of the project was the use of
shallow air core drilling along road traverses, designed to gain
3D regolith information, including petrographic, mineralogical
and geochemical data down the profiles.
THE LANDSCAPE HISTORY AND NATURE OF
THE REGOLITH
Much of the area of western New South Wales has been a
surficial or near surficial environment since at least the
Mesozoic. It is surrounded and onlapped by the Mesozoic-
Cainozoic Eromanga, Surat and Murray Basins (Figure 1) and
much of the preserved history of erosion and deposition in the
region has been controlled by events in these basins, particularly
the timing of basin initiation and subsequent changes in base
level. Palaeozoic basement rocks are variably covered by Jurassic
to Early Tertiary fluviatile, lacustrine and shallow marine
sediments. In erosional and elevated areas, such as the Cobar
Uplands (Duk-Rodkin, Chan and McQueen, 2004), this cover
typically forms thin remnants preserved at different levels in the
present landscape. There is also extensive cover by younger
transported regolith, particularly in palaeochannel systems (up to
60 m deep), colluvial fans and plains and modern drainages.
Much of the post-Miocene cover is backfilling older more deeply
incised palaeovalleys. The geological evidence indicates a
continuously evolving landscape with concomitant erosion,
deposition and recycling of surficial materials through changing
landscape levels. Palaeomagnetic dating of weathering profiles
S
D
N
LA
H
IG
SURAT BASIN
H
EROMANGA BASIN Mesozoic (1600m)
Mesozoic (1200m)
WESTERN
NEW SOUTH WALES
N
R
TE
S
CANOBOLAS DIVIDE
EA
MURRAY BASIN 0 1000 km
Cainozoic (600m)
KMf005-04
FIG 1 - Location and regional geomorphic setting of western New
South Wales. The approximate maximum thickness of preserved
sediments in the surrounding Mesozoic-Cainozoic basins is shown
(in parentheses).
231
K G McQUEEN and A McRAE
across the region indicates preservation of ferruginous materials
from two major episodes of deep oxidation and hematite fixation
at 6010 Ma and 123 Ma (McQueen, Pillans and Smith, 2002).
Older, Carboniferous and Jurassic weathering/oxidation ages
have been found at some ore deposits (OSullivan et al, 2000;
McQueen, Pillans and Smith, 2002).
The regolith of the region can be conveniently divided into in
situ, transported, indurated and lag materials. These materials are
a mixture of rock-derived, extraneous and biogenic components.
In situ regolith
In situ materials can be subdivided on the basis of landform
setting and degree of weathering. Preserved weathering profiles
show great variability in structure and thickness (typically <3 m
to >100 m thick). This variability reflects a combination through
time of variable depth of weathering, controlled by the interplay
of lithological variation, permeability and availability of water,
and variable erosion, related to differential tectonic uplift and
incision. Well-developed weathering profiles show considerable
ferruginisation in the upper parts and ferruginous mottling and
veining in the saprolite zone. However, they cannot be described
as classical laterite profiles. This may be due to the relatively
low iron content of many of the common lithologies in the region
(siliclastic sediments) and greater erosional modification during
profile development. In many profiles it can be difficult to pick
the boundary between fresh rock and saprolite (eg in drill
cuttings) because the abundant primary minerals in the common
parent rocks (quartz-muscovite-clays) are stable into the
weathered zone. The breakdown of muscovite and colour
changes related to oxidation are the best guides in these
weathered rocks. We have found the PIMA a very important aid
to rapid mineralogical analysis of regolith materials, particularly
for the phyllosilicates, and for distinguishing in situ from
transported regolith.
Transported regolith
Transported materials include alluvial, lacustrine, colluvial and
aeolian components. Alluvial materials are probably the most
abundant and occur at all levels of the landscape. Old inverted
palaeochannels contain quartz-rich sands and gravels and some
cobble-boulder conglomerates. Clay-rich sediments with sand
and minor gravel lenses are preserved in deeply incised
palaeochannels that developed through the Tertiary. Pisoidal,
ferruginous clasts are abundant in some of these palaeochannel
sediments. This ferruginous component contains maghemite,
resulting in a strong expression of the host palaeochannels in
aeromagnetic imagery (a problem when conducting magnetic
surveys for basement features). Currently active alluvial systems
are dominated by sandy loams (largely sourced by sheetwash
from recent soil and aeolian deposits) with minor channel
gravels. Clays, silts and fine sands are common in alluvial
swamp and lake deposits that developed as a result of drainage
disruptions. Marginal lunettes contain coarser, well-rounded
sediments. Colluvial materials are widespread in generally thin,
bedrock-masking deposits. These are typically coarse and
angular on upper slopes and decrease in grainsize downslope.
The colluvium is generally dominated by lithic clasts in a silty
loam matrix. Colluvial regolith occupies depositional and
erosional plains, slope deposits on rises and less commonly
colluvial fans adjacent to rangefronts and higher relief
landforms. Aeolian materials mainly occur as a silt-sized
component in the red silty-loam soils of the region. They also
include well-rounded and sorted sands and silts in stabilised
longitudinal dunes on the western edge of the area and in some
source-bordering dunes adjacent to lakes and the modern
drainage system.
232 Adelaide, SA, 19 - 22 September 2004
Indurated regolith
Indurated materials have formed during the weathering history of
the region by the introduction and precipitation of silica, iron
oxides and carbonates. This has occurred in bedrock, transported
regolith and soils. The indurated materials are generally more
resistant to further weathering and erosion and commonly
become inverted in the landscape. Silicified gravels (silcretes)
originally formed in channels now occur as cappings on hills.
These silcretes appear to be early to mid Tertiary in age.
Silicification of more porous (typically already highly siliceous)
bedrock is also a feature of the region. Ferricretes (ironstones)
occur in thick, horizontal caps and linear masses and probably
also formed low in the landscape along drainage channels or
groundwater seeps. Regolith carbonate (calcrete) is common
throughout the region occurring mainly as nodules in the lower
part of the soil, as more massive, in some cases laminated,
hardpans at the soil-bedrock/saprolite interface and as coatings
and veinings on bedrock. Calcrete occurs within palaeochannels
but does not appear to be present in the active, modern drainage
systems The carbonate includes calcite, dolomite, and rarely
magnesite. The calcretes have formed under arid climatic
conditions (probably during the last 1 Ma).
Lag materials
Surface lag is common and widespread in the present landscape
and has been derived from the full range of underlying regolith
materials. Thus it may reflect the immediately underling bedrock
(in areas of outcrop), a local source (in some colluvium) or a
distant source (overlying alluvial regolith). Lag is dominated by
chemically and physically resistate minerals and rocks. It
includes a macroscopic component of quartz, lithic fragments,
ferruginious lithic fragments and highly ferruginous materials.
The latter includes residual and surface deposited-cementing iron
oxides (predominantly hematite with some maghemite). There is
also a micro-lag component of resistate minerals including
zircon, rutile and other heavy minerals (probably including gold
local to eroding gold mineralisation) which has concentrated at
the surface by the winnowing effects of wind and sheetwash.
ELEMENT DISPERSION PROCESSES IN THE
REGOLITH
The regional weathering history has been strongly influenced by
major climatic variations through the Cainozoic. In general terms
this has resulted in deep chemical weathering under
predominantly warm humid conditions in the Early-Mid Tertiary
and superimposed drier chemical weathering under increasingly
arid conditions from the Late Tertiary. In detail the picture is
more complex with fluctuations to at least two cooler-dry
episodes prior to the Oligocene (McGowran and Li, 1998).
Element dispersion processes need to be considered in terms of
these two quite different chemical weathering regimes.
Warm and wet climatic conditions, with high water table
levels, would have favoured hydration/hydrolysis reactions and
mobility of reduced species, particularly Fe2+. Groundwater pH
conditions would have been neutral to acid, the latter particularly
where sulfides were oxidising. It is also likely that high organic
content would have favoured organo-complexing of many
elements. Arid climatic conditions, with falling and fluctuating
water tables, favoured oxidation reactions and a change to more
complex groundwater compositions, particularly with higher
salinity, increased activity of carbonate and sulfate and regional
neutral to alkaline pH. Under these superimposed arid conditions
the solubility and mobility of some elements, particularly Au was
increased where chloride and thiosulfate complexing occurred,
whereas other elements, especially Pb, Ag, Ba and Hg became
relatively fixed as insoluble chlorides and sulfates. Marked pH
gradients around sulfide deposits that continued to weather
PACRIM 2004
 NEW WAYS TO EXPLORE THROUGH THE REGOLITH IN WESTERN NEW SOUTH WALES
resulted in dispersion of elements such as Cu and Zn to form
broad anomalies. During exploration it is important to know
what portion of the weathering history is preserved in a particular
profile. In many cases the earlier dispersion patterns are
overprinted by patterns developed during arid conditions. This
combination can result in very low concentrations of many target
and pathfinder elements over mineralised sites, particularly in the
non-ferruginous saprolite.
Iron oxides/oxyhydroxides exert a major control on element
dispersion/fixation and the sequence of iron mobility and
precipitation of the major stable phases of goethite, and hematite
is critical to the distribution of dispersed elements in the regolith.
These iron phases typically show different distributions around
weathering mineralisation and have different preferences for
hosting trace elements. Studies of weathering profiles over
mineralisation in the Cobar region (eg Scott, Rabone and
Chaffee, 1991; Cairns, McQueen and Leah, 2001; McQueen and
Munro, 2003; Leverett, McKinnon and Williams, 2004) indicate
a progressive change in mineral hosts for target and pathfinder
elements from primary sulfide and specific secondary minerals
(including supergene sulfides, arsenates, sulfates, carbonates,
chlorides and oxides) in the lower part of the profile to more
generic Fe- and Mn-oxides/oxyhydroxides towards the top.
Within the weathering profile element dispersion is largely by
chemical processes. However after elements become fixed in the
stable ferruginous component near surface they can be dispersed
by mechanical processes in the eroding landscape. This can
result in markedly different dispersion patterns in different parts
of the regolith. For example, Pb and As can show limited
chemical dispersion within the weathering profile compared to
Cu and Zn which may form a broad chemically dispersed halo,
generally concentrated in goethite. However at the surface, Pb
and As (as well as Sb, Bi and Ba) become strongly fixed in
hematite which may be widely dispersed into colluvium,
alluvium and soil by mechanical processes.
The role of biogeochemical processes in element dispersion
through the regolith of this region is hardly known. However
there have been some initial studies (eg Hill and Hill, 2003) and
research is continuing. The root systems of plants can clearly
play a role in transferring elements in solution up the weathering
profile and onto the surface (Brooks, 1998). Plants can thus
potentially lift geochemical anomalies in a continuous and
ongoing process through transported, depositing and weathering
cover. The hugely abundant micro-biota of the regolith must play
a significant role in mineral weathering, biochemical dissolution
and fixation of major and trace elements (Gilkes, 1998).
Bioturbation of the upper parts of the regolith, has played a
major role in mechanical redistribution of regolith components
and their contained element concentrations.
REGOLITH-RELATED ELEMENT ASSOCIATIONS
Geochemical investigations of regolith materials from
background/baseline sites across the Cobar-Girilambone region
have identified important regolith-related associations of
elements. Some of these give rise to highly variable background
levels for ore and pathfinder elements in different parts of the
regolith, which may be confused with ore-related anomalies. The
main associations are:
an evaporitic association of Ca-MgAu, in some cases with
Ba-Sr, related to regolith carbonate and sulfate accumulation
in the near-surface regolith and at the base of palaeochannels
and transported regolith;
an association of Mn-Co-ZnNi-CuAu developed in redox
boundary accumulations of manganese oxides/oxyhyroxides
(particularly lithiophorite), commonly at around 20 - 30 m
and the present, deeper water table;
an association of Fe-Cu-Zn with goethitic accumulations in
the regolith; and
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
an association of Fe-As-PbSbBi with hematite,
particularly in ferruginous lag, paleochannel sediments
containing ferruginous lag and in hematite-rich mottles in the
upper saprolite.
In mineralised environments there is a weathering-related
fractionation of most target and pathfinder elements to different
host minerals in different parts of the regolith profile (see
McQueen and Munro, 2003; Leverett, McKinnon and Williams,
2004). Key host minerals for target and pathfinder elements in
the upper, strongly oxidised part of the regolith over
mineralisation include: hematite (Pb, As, Sb, Ba, Bi Cu);
goethite (Cu, Zn Pb, As); alunite-jarosite group minerals (Pb);
manganese minerals, particulary lithiophorite (Ni, Co, Cu, Zn)
and coronadite-cryptomelane-hollandite group minerals (Pb, Ba);
hidalgoite (Pb, As, Ag); hindsalite (Pb, Cu, Zn); anglesite (Pb);
chlorargyrite (Ag); azurite-malachite (Cu) and cassiterite (Sn).
Some clay minerals, particularly kaolinite, also appear to host
elevated concentrations of trace elements such as Cu in
mineralised environments (Cairns, McQueen and Leah, 2001).
GEOCHEMICAL TECHNIQUES AND SAMPLING
MEDIA
A range of regolith materials is available for geochemical
sampling, but the trick is to select the most appropriate,
depending on:
sampling convenience and cost;
the level of anomaly definition required;
the nature of the regolith (eg in situ versus transported,
thickness, degree of leaching); and
the type of dispersion pathways active in the particular
regolith-landform setting.
Many of the appropriate techniques and sampling media are
already widely used in geochemical exploration, but integrating
these into a better regolith framework can greatly improve their
effectiveness. For example, a clearer understanding of the 3D
architecture of the regolith can indicate what level of the profile
is being sampled at any particular location and its likely
dispersion history. Fundamental knowledge of mineral hosts for
target and pathfinder elements in particular regolith settings can
assist in media selection, analytical approach and data
interpretation.
In the Cobar-Girilambone region the predominant soil is a
relatively homogeneous red silty loam with a typical pH range of
5 to 8.5 (pH commonly increases with depth). The soil contains a
major aeolian component of mainly quartz, clay and other
resistate minerals such as zircon. This component is mainly in
the 30 - 80 m size fraction, depending on longitudinal position
in the region (Tate et al, 2004). To reduce the diluting effect of
this extraneous component, the >100 m fraction is the most
appropriate for geochemical sampling. Deeper sampling
(>0.3 m) or sampling the C horizon are other options, as the
abundance of aeolian additions decrease with depth. Despite
aeolian dilution and soil reworking/redistribution there is a
significant bedrock-derived component in soils over in situ
regolith related to bioturbation. In these settings soils are an
appropriate and convenient sampling medium.
Lag is widespread in the region, is easy to sample and has
been widely used as a sampling medium (generally the
3 - 15 mm size fraction), particularly at the reconnaissance level
of exploration. However, in many situations it has a complex
transport history and variable mineralogy that can affect its
usefulness. Bulk lag containing lithic and ferruginous fractions
with a high goethite content is most useful where locally derived
from in situ regolith. Lag transported directly from in situ
regolith is useful for regional anomaly detection. Significant ore
233
K G McQUEEN and A McRAE
and pathfinder element fractionation has occurred during lag
maturation and transport and some elements require
normalisation to iron (hematite) content before interpretation.
Micro lag (<250 m), which includes concentrated dense and
resistate mineral grains, is also a useful sampling medium,
particularly for gold.
Regolith carbonates are widespread in the lower part of the
soil profile and as coatings on saprock/bedrock. This material
accumulates gold and other elements and is useful for both
regional and local anomaly detection (McQueen, Hill and Foster,
1999). Investigation of calcrete geochemistry in the
Cobar-Girilambone area indicates a regional threshold of 4 ppb
Au and values above 19 ppb close to known gold-bearing
deposits. As many of the styles of mineralisation in western New
South Wales are polymetallic with associated gold, calcrete
sampling is a potentially useful technique for locating these
deposits using gold as a pathfinder. Regional anomalies have
been detected in transported regolith down the palaeodrainage
from known gold mineralisation and calcrete sampling
(combined with knowledge of the transport path) could be used
in a similar way to regional stream sediment sampling.
In areas with thin saprolite the upper generally more
ferruginous zone below the soil or transported regolith can be a
good geochemical sampling medium. This is due to the control
that goethite and hematite have in retaining many target and
pathfinder trace elements released and dispersed during
weathering. Templates based on ratios of trace elements to iron
can be useful in separating background concentrations of these
elements from truly anomalous values. In deeper, strongly
weathered profiles or where there has been marked erosional
stripping saprolite geochemistry requires more careful
interpretation. The active redox front, generally close to the
water table, is typically a zone of accumulation for dispersed
trace elements. Perched iron oxide/oxyhydroxide concentrations
in deeper saprolite also commonly retain ore and pathfinder
elements. Fossil and present redox boundaries in well-developed
weathering profiles commonly show a regolith generated
association of Mn and Co in some cases with Zn, Cu and Au
accumulation (as described above).
The regolith biota offer a range of potential sampling media,
but this component of the regolith is still poorly understood and
the advantages over other media types still need to be
established. In the Cobar-Girilambone area studies have been
conducted on metal uptake (particularly Au) in white cyprus pine
(Callitris columellaris, Cohen et al, 1998; Cohen, Dunlop and
Shen, 2000). There have also been some studies by postgraduate
students (Huang, 1998; Hill and Hill, 2003) of other plant species
such as mulga (Acacia aneura), black cyprus pine (Callitris
endlicheri), river red gums (E. camaldulensis), golden wattle
(Acacia pycnantha), yellow box (E. melliodora) and red box (E.
polyanthemos). Preliminary investigation of calcareous (lichen)
soil crusts suggests that these may concentrate some metals
(including Au) and could represent an alternate sampling
medium to soils in areas of in situ regolith.
The relationships between common regolith-landform
components and element dispersion pathways in the
Cobar-Girilambone area are summarised in Figure 2. Regolith
profile sampling models are shown for some of the different
types of regolith-landform settings.
It is essential that geochemical exploration programs take a
multi-element approach utilising a range of target and pathfinder
elements as well as selected major elements. The spatial
association of anomalies for a number of key ore-related
elements provides a more robust indicator of mineralisation than
single element anomalies. This is particularly the case in the
regolith where some elements may be strongly depleted by
weathering processes or where anomalies are very subtle or fall
within the range of highly variable regolith-related background
levels. By analysing for some of the major elements, particularly
234 Adelaide, SA, 19 - 22 September 2004
Fe, Ca, Mg, K, Mn, it is possible to monitor the main host
components for trace elements in the regolith. If necessary, trace
element data can be normalised for varying abundances of the
hosting minerals, for example Fe and Mn oxides/oxyhydroxides
or regolith carbonates. Major element data can also aid
interpretation of profiles and guide consistency of sampling.
Modern methods of analysis are highly suited to multi-element
measurement at low cost and the extra expense of analysing
additional elements is usually far outweighed by the benefits of
the additional knowledge.
From our regional baseline studies in the Cobar-Girilambone
region we have found that multi-acid (total) or strong acid
(aqua regia) digest followed by analysis using ICP OES/MS are
good first pass methods for geochemical exploration of the
regolith. With this approach the maximum information is
obtained on the overall geochemical nature of the regolith
(particularly from drill profiles) at reasonable cost. For anomaly
detection it is important to select the most appropriate sampling
medium. Partial leach methods are useful where the element
dispersion controls and regolith host sites are well understood (eg
from an orientation study). These methods target loosely bound
elements or elements in particular phases and have the advantage
of lowering detection limits by excluding from solution the
non-selected sample component. Some of these methods are
significantly affected by pH variability in the sampling media,
particularly related to the presence of regolith carbonates (eg
Dronseika and Evers, 2000; Fidler, McGeogh and Hannan,
2004). Information on the distribution and landscape setting of
various regolith components can assist in planning and
interpreting such surveys. Where true total analysis data is
required, for example to assist with interpretation of parent rock
type, or apply element ratio techniques to detect alteration
patterns, analysis by XRF or INAA is generally necessary. Some
elements, notably Au, require specific analysis techniques
generally on larger samples to reduce the nugget effect.
INTEGRATING REGOLITH KNOWLEDGE INTO
EXPLORATION
Better knowledge of the regolith can significantly improve
exploration strategies and success. Mapping the 3D distribution
of different regolith materials is a first step for unravelling the
regolith puzzle. A simple subdivision into in situ, shallow
transported and deep transported regolith is commonly sufficient
to dramatically improve sampling and data interpretation
strategies. More control can be obtained by understanding the
degree of chemical leaching, chemical enrichment, source and
direction of transport of different regolith materials.
Understanding the landscape (and palaeo-landscape) setting of
regolith materials can help in determining their nature origin and
history of modification, and consequently suitability as
geochemical sampling media. An approach which integrates
knowledge of the regolith and landscape setting, direct 3D
information on the regolith, geophysical data on the surface and
near-surface regolith and geochemical techniques is summarised
schematically in Figure 3. A number of exploration companies
working in western New South Wales are now incorporating a
regolith-landform knowledge layer into their exploration
strategies (eg Stockton, 2004; Skirka, David and Mackenzie,
2004).
Historically, much geochemical sampling in areas of regolith
cover has involved Rotary Air Blast (RAB) or Aircore drilling to
fresh bedrock or to refusal and sampling of material at this
position. Most explorationists now realise that this is a very
restricted and inconsistent method of sampling as refusal could
be a wide range of materials ranging from bedrock to
components of the regolith profile (eg silcrete, hard pans, gravel
layers in transported regolith). Sampling at a particular depth
within the regolith is also fraught with uncertainties, given the
PACRIM 2004
 NEW WAYS TO EXPLORE THROUGH THE REGOLITH IN WESTERN NEW SOUTH WALES

FIG 2 - (A) Diagram showing the common regolith-landform components and element dispersion pathways in the Cobar-Girilambone
region of western New South Wales. (B) Profile models for different regolith-landform settings and suggested geochemical sampling
media for surface and subsurface (drilling) sampling methods.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 235

K G McQUEEN and A McRAE
FIG 3 - A schematic model for mineral exploration in regolith-dominated terrains showing how regolith-landform knowledge
can be integrated with other exploration techniques.
considerable variability in regolith profiles and the established
heterogenous uptake of dispersed elements in different regolith
host phases. We recommend sampling of the whole profile at 1 m
intervals for at least a small number of drillholes (eg one in ten)
that are representative of the different types of regolith in a give
area. This type of orientation will indicate how the chemical
composition of the regolith varies through the profile and assist
with sampling of the most appropriate horizons and materials.
Good visual, textural and spectral logging of regolith cuttings is
essential for appropriate 3D knowledge of the regolith and to
identify consistent and appropriate sampling media.
CONSTRUCTING EXPLORATION GO MAPS
Work in the Girilambone-Cobar region has demonstrated an
approach for producing reliable regolith-landform maps (suitable
for application by the NSW Geological Survey). These maps can
provide the basis for derivative maps or exploration go maps,
which highlight particular regolith attributes (eg in situ versus
transported origin or thickness of transported cover). Exploration
go maps can be used to advise the most appropriate exploration
strategies in different parts of a regolith dominated terrain and
also to improve interpretation of existing data from these areas.
The basic exploration go map shows areas of in situ regolith,
transported regolith with thickness less than 5 m and transported
regolith of thickness greater than 5 m. This simple subdivision is
useful for advising which geochemical exploration techniques
can be used over these three types of area, based on the premise
that more than 5 m of transported cover can effectively mask
most secondary geochemical dispersion patterns. The current
technique for producing these maps is outlined below and an
example of the different stages of construction is shown in Figure
4.
Following detailed regolith-landform mapping the
regolith-landform unit polygons are grouped into in situ and
transported categories. The areas of transported regolith can then
be subdivided into the two depth categories of less than 5 m and
greater than 5 m. This is done by first plotting any available 3D
data on the thickness of the transported regolith (usually from
drillhole data) and interpolating this information to produce a
first pass thickness map. For this we have used ArcGIS Spatial
Analyst which employs an inverse distance weighted function to
estimate thickness between points of known thickness.
236 Adelaide, SA, 19 - 22 September 2004
Depending on the density of 3D information this generally gives
a very coarse estimate of thickness categories. The next step is to
refine this plot using thickness estimates interpreted from the
detailed regolith landform units and their spatial relationships.
For example, areas of colluvial regolith with abundant angular
clasts surrounding areas of bedrock or shallow in situ regolith are
likely to be composed of transported regolith less than 5 m thick.
Major alluvial systems are more likely to have transported
regolith thicker than 5 m. This plot can be further refined by
building in the positions of deeply incised palaeochannels
containing maghemite-bearing sediments, which are well
expressed in the aeromagnetic imagery. Additional controls can
be added such as depth estimates calculated by deconvolution of
aeromagnetic data or thickness data for non-magnetic sediments
derived from electromagnetic surveys.
The basic type of exploration go map can be further refined
to show the depth of weathering of the in situ regolith, again
using available information from drilling combined with data
from the detailed regolith-landform map. This can be useful for
indicating areas of likely greater chemical leaching, related to
preserved deep weathering profiles, and establishing the position
of the present surface in variably truncated profiles. Other
derivative maps can be constructed which show such features as
the provenance and direction of source of transported regolith.
These can be constructed by incorporating information about the
present and palaeolandscapes (from regolith-landform mapping
and aeromagnetic imagery) and data from radiometric imagery
combined with bedrock geology. This information can help
understand the geochemistry of transported regolith and interpret
displaced geochemical anomalies.
ACKNOWLEDGEMENTS
The work reported here was supported by the Australian
Governments Cooperative Research Centre Program within
CRC LEME and by the New South Wales Department of Mineral
Resources. A team of more than eighteen LEME researchers,
students and technical staff working on the Girilambone and
Cobar Projects, have contributed to the findings described in this
article. We also thank the various land owners and mining
companies in the Girilambone-Cobar region for providing access
to their properties. Peter Hancock read and constructively
commented on an early draft of the article.
PACRIM 2004
 NEW WAYS TO EXPLORE THROUGH THE REGOLITH IN WESTERN NEW SOUTH WALES

FIG 4 - An example from part of the Hermidale 1:100 000 sheet area illustrating the key stages in the construction of a derivative regolith
thickness map. Stage 1 shows the landform units (from detailed regolith-landform map). Stage 2 shows simplification to in situ and
transported units. Stage 3 shows interpolated regolith thickness based on available drillhole information. Stage 4 is the final derivative
map incorporating the previous attributes and refinements from interpretation of detailed regolith-landform units and aeromagnetic data
(for magnetic palaeochannels). This map shows areas of in situ regolith, transported regolith less than 5 m thick and transported regolith
greater than 5 m thick. It can be used to advise appropriate exploration methods for different parts of the area.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 237

K G McQUEEN and A McRAE
REFERENCES
Brooks, R R, 1998. Phytochemistry of hyperaccumulators, in Plants that
hyperaccumulate Heavy Metals: their role in Phytoremediation,
microbiology, archaeology, mineral exploration and phytomining
(Ed: R R Brooks), pp 15-54 (CAB International: New York).
Cairns, C J, McQueen, K G and Leah, P A, 2001. Mineralogical controls
on element dispersion in regolith over two mineralised shear zones
near the Peak, Cobar, New South Wales, Journal of Geochemical
Exploration, 72:1-21.
Cohen, D R, Shen, X C, Dunlop, A C and Rutherford, N F, 1998. A
comparison of selective extraction soil geochemistry and
biogeochemistry in the Cobar Area, New South Wales, Journal of
Geochemical Exploration, 61:173-189.
Cohen, D R, Dunlop, A C and Shen X C, 2000. Contrasting geochemical
and mineralogical patterns in residual and transported regolith in the
Cobar region, NSW, in Central West Symposium Cobar 2000 (Eds:
K G McQueen and C L Stegman) pp 8-12 (Cooperative Research
Centre for Landscape Evolution and Mineral Exploration: Perth).
Dronseika, E and Evers, A, 2000. Aspects of EDTA partial digest
extractions in variable pH terrains the vulnerability of unbuffered
digests, Explore (Newsletter of the Association of Applied
Geochemists), 109:10-14.
Duk-Rodkin, A, Chan, R A and McQueen, K G, 2004. Drainage
evolution of the Cobar region, in Proceedings Exploration Field
Workshop Cobar Region 2004 (Eds: K G McQueen and K M Scott),
pp 37-42 (Cooperative Research Centre for Landscape Environments
and Mineral Exploration: Perth).
Fidler, R, McGeogh, M and Hannan, K, 2004. An unexpected aspect of
EDTA partial digest extractions, Explore (Newsletter of the
Association of Applied Geochemists), 124:1-4.
Gilkes, R J, 1998. Biology and the regolith: an overview, in The State of
the Regolith, Proceedings of the Second Australian Conference on
Landscape Evolution and Mineral Exploration (Ed: R A Eggleton),
pp 110-125 (Geological Society of Australia Special Publication No
20).
Hill, S M and Hill, L J, 2003. Some important plant characteristics and
assay overviews for biogeochemical surveys in western New South
Wales, in Advances in Regolith, (Ed: I C Roach), pp 187-192
(Cooperative Research Centre for Landscape Environments and
Mineral Exploration: Perth).
Huang, X, 1998. Biogeochemical techniques for environmental
monitoring and mineral exploration: A case study of the Temora gold
mine, MSc thesis (unpublished), University of Canberra, Canberra.
Leverett, P, McKinnon, A R and Williams, P A, 2004. A supergene
exploration model for Cobar style deposits, in Proceedings
Exploration Field Workshop Cobar Region 2004 (Eds: K G
McQueen and K M Scott), pp 46-50 (Cooperative Research Centre
for Landscape Environments and Mineral Exploration: Perth).
238 Adelaide, SA, 19 - 22 September 2004
McGowran, B and Li, Q, 1998. Cainozoic climatic change and its
implications for understanding the Australian regolith, in The State of
the Regolith, Geological Society of Australia Special Publication 20
(Ed: R A Eggleton), pp 86-103 (Geological Society of Australia:
Sydney).
McQueen, K G and Munro, D C, 2003. Weathering-controlled
fractionation of ore and pathfinder elements at Cobar, NSW, in
Advances in Regolith (Ed: I C Roach), pp 296-300 (Cooperative
Research Centre for Landscape Environments and Mineral
Exploration: Perth).
McQueen, K G, Hill, S M and Foster, K A, 1999. The nature and
distribution of regolith carbonate accumulations in southeastern
Australia and their potential as a sampling medium in geochemical
exploration, Journal of Geochemical Exploration, 67:67-82.
McQueen, K G, Pillans, B J and Smith, M L, 2002. Constraining the
weathering history of the Cobar region, Western NSW, in Abstracts
of the 16th Australian Geological Convention, Adelaide, Geoscience
2002: Expanding Horizons (Ed: V P Preiss), p 426 (Geological
Society of Australia: Sydney).
OSullivan, P B, Gibson, D L, Kohn, B P, Pillans, B J and Pain, C F,
2000. Long-term landscape evolution of the Northparkes region of
the Lachlan Fold Belt, Australia: Constraints from fission track and
palaeomagnetic data, The Journal of Geology, 108:1-16.
Scott, K M, Rabone, G and Chaffee, M A, 1991. Weathering and its
effects upon geochemical dispersion at the polymetallic Wagga Tank
deposit, NSW, Australia, Journal of Geochemical Exploration,
40:413-426.
Skirka, M, David, V and Mackenzie, I, 2004. The discovery and regolith
expression of the Hera Au-Cu-Zn-Pb-Ag deposit, in Proceedings
Exploration Field Workshop Cobar Region 2004 (Eds: K G
McQueen and K M Scott), pp 78-82 (Cooperative Research Centre
for Landscape Environments and Mineral Exploration: Perth).
Stockton, I R, 2004. The strategy for geochemical exploration around the
CSA mine, in Proceedings Exploration Field Workshop Cobar
Region 2004 (Eds: K G McQueen and K M Scott), pp 83-93
(Cooperative Research Centre for Landscape Environments and
Mineral Exploration: Perth).
Tate, S E, Greene, R S B, Scott, K M and McQueen, K G, 2004. The
aeolian input to the Girilambone regolith Implications for
Exploration, in Proceedings Exploration Field Workshop Cobar
Region 2004 (Eds: K G McQueen and K M Scott), pp 94-100
(Cooperative Research Centre for Landscape Environments and
Mineral Exploration: Perth).
PACRIM 2004
Prospectivity Mapping Using GIS With Publicly Available Earth
Science Data A New Targeting Tool Being Successfully Used
for Exploration in New Zealand
G A Partington1 and M J Sale2
ABSTRACT
The Epithermal and Mesothermal Gold Prospectivity Modelling Projects
recently completed by Crown Minerals provided explorers in New
Zealand with a new compilation of historical exploration data combined
with new geological information from the GNS QMap 1:250 000 scale
mapping project. These data were used to produce predictive mineral
potential maps for gold mineralisation in New Zealand. The aim of these
projects, which was to stimulate mineral exploration and investment in
exploration, has been successful with new companies acquiring new
tenements and committing significant exploration expenditure to
exploring in New Zealand in the coming years.
The projects were undertaken at a national scale and consequently not
all exploration data were compiled into the prospectivity models. HPD
New Zealand Ltd recognised the value of the prospectivity modelling
work and committed exploration funds to continue the modelling process
started by the projects. They recognised the need to compile the
remaining data and run the models again to allow detailed exploration
targeting.
Detailed data compilations including digitising historic exploration
stream sediment sample, rock chip sample, soil sample and drilling data
has been completed. New models have been completed in Otago for
mesothermal gold mineralisation and in the Coromandel for epithermal
gold. The new models have been compared with the original regional
scale models and used to target prospect scale exploration. This work has
allowed exploration models for epithermal and mesothermal
mineralisation in New Zealand to be refined. More importantly this work
has identified significant areas with potential to host gold mineralisation
with little or no systematic geochemical data including soil sampling or
drilling. Exploration work programs have been designed to acquire these
missing data and exploration funds have now been committed to test the
areas highlighted by the prospectivity modelling.
In summary, the Epithermal and Mesothermal Gold Prospectivity
Modelling Projects commissioned by Crown Minerals have successfully
attracted new investment and ideas to the exploration scene in New
Zealand. The projects cost an estimated $NZ 250 000 and could in the
next two years, just through exploration expenditure, attract more than
$NZ 10 M in investment. If a mine is discovered the return on investment
will be considerably greater.
INTRODUCTION
The completion of the Epithermal and Mesothermal Gold
Prospectivity Modelling Projects by Crown Minerals (the New
Zealand Department of Mines) provided explorers in New
Zealand with a new compilation of historical exploration data
(Partington and Smillie, 2002; Rattenbury and Partington, 2003),
which has been combined with new geological information from
the GNS (the New Zealand Geological Survey) QMap 1:250 000
scale mapping project (Nathan 1994; Nathan 1998). These data
were used to produce predictive mineral potential maps for gold
mineralisation in New Zealand (Partington and Smillie, 2002;
Partington et al, 2002; Rattenbury and Partington, 2003). The
aim of these projects was to stimulate mineral exploration and
investment in exploration in New Zealand, which has been
successful, with new companies acquiring tenement positions
1. MAusIMM, Kenex Knowledge Systems, PO Box 41136, Eastbourne
Wellington, New Zealand. E-mail: g.partington@kenex.co.nz
2. MAusIMM, HPD New Zealand Ltd, 2nd Floor, 15 Upper Grosvenor
Street, London W1K 7PJ, UK.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
and committing significant exploration expenditure to exploring
in New Zealand.
Both the Mesothermal and Epithermal Gold Projects were
compiled at a national scale and consequently not all historic
exploration data were included in the regional prospectivity
models. HPD New Zealand Ltd (HPD NZ), who has a significant
exploration permit holding in New Zealand, recognised the value
of the prospectivity modelling techniques and committed
exploration funds to continue the modelling process started by
the government projects. All remaining historic exploration data
were compiled and new prospect scale geological and
geochemical data collected in the field. These data were
integrated with the data from Crown Minerals in a Geographic
Information System (GIS) and detailed prospectivity models
created for both epithermal gold in the Coromandel and
mesothermal gold in Otago. These models were used the for
prospect scale exploration targeting and the development of
ongoing exploration work programs and budgets. This work has
now been completed, allowing HPD NZ to move rapidly from
the reconnaissance stage to prospect scale exploration, so
reducing costs and risk at the earliest stage of exploration.
This paper presents a summary of the results of the follow-up
work carried out on the HPD NZ Otago permit areas and
compares the new detailed Otago prospectivity model with the
original regional Otago prospectivity model over the same area
(Partington and Smillie, 2002). This has allowed the exploration
model for mesothermal gold mineralisation in Otago to be
refined. More importantly, new exploration targets have been
developed using the mineral potential maps and prospective areas
that require follow-up geochemical sampling and drilling
identified. The value and use of mineral potential maps for
exploration targeting and exploration management are also
discussed.
MESOTHERMAL GOLD MINERALISATION IN
OTAGO
Mesothermal gold deposits in New Zealand are restricted to the
South Island of New Zealand (Figure 1) and occur in greywacke
rocks of Palaeozoic age located in the Greymouth and Nelson
districts (eg Reefton, including the Globe-Progress deposit,
which is greater than two million ounces of gold), and also in
Mesozoic age schists of the Marlborough and Otago districts (eg
the Macraes Flat deposit, which is greater than five million
ounces of gold) and in schists and greywackes of the Southern
Alps (Christie and Braithwaite, 1999; Christie, 2002). A
summary of the characteristics of the two styles of mesothermal
gold mineralisation in New Zealand is given in Table 1.
The schist belts in the South Island of New Zealand are one of
the high priority exploration targets for exploration companies in
New Zealand, and comprise psammitic, pelitic and interlayered
psammitic-pelitic schist, with local metavolcanic and quartzite
units (Mortimer, 1993). The rocks have been assigned to three
tectonostratigraphic units: Caples Terrane, Torlesse Terrane and
Aspiring Lithologic Association (Mortimer 1993; Mortimer
2000), which were metamorphosed in the Mesozoic as a result of
collision between the Torlesse and Caples Terranes. The
239
G A PARTINGTON and M J SALE

metamorphic grade is mostly greenschist facies (chlorite zone

New Zealand mesothermal and garnet-biotite-albite zone), but rises to amphibolite facies
gold deposits near the Alpine Fault (Craw 1998; Mortimer 2000; Mortimer
2001; Foster and Lister 2003). The abundance of mesoscopic
folds with different styles is generally taken as evidence for syn-
Northwest
and post-metamorphic recumbent and nappe folding (MacKenzie
Nelson and Craw, 2001). More brittle features are associated with
Golden Blocks regional uplift, including low-angle thrust zones and steeper
Marlborough north west and north east trending normal faults (Mortimer,
KEY
Wakamarina 1993). Forster and Lister (2003) concluded that central belt of
Paleozoic
rocks Westland
Lyell
schist in the Torlesse Terrane, adjacent to the Caples-Torlesse
Mesozoic Reefton Goldfield boundary was exhumed as the result of the operation of
schist large-scale extensional ductile shear zones. These shear zones
Historical Langdons
workings
now mark the lateral extent of the higher grade parts of the schist
Mt Greenland
Gold
and the structural arch in which the Otago Schist is now exposed.
mine The location and development of these structures appears to have
ult
played an important role in the formation of mesothermal gold
Fa
deposits in Otago.
ine
Alp Gold mining on the South Island of New Zealand was
historically from small quartz lode deposits, but recent
exploration and mining in Otago has identified the importance of
Shotover
Bendigo
lower grade disseminated regional shear zone deposits (eg
Fiordland
Carrick Macraes Flat Macraes Flat, Carrick and Bullendale). Mesozoic schist quartz
lode deposits typically dip steeply and are discordant to the
Otago Nenthorn
Barewood foliation in the host schist. The Otago lodes tend to strike north
west, parallel to the antiformal axis of the schist belt (Christie,
Preservation
2002). Regional shear zone (Macraes type) deposits occur as
Inlet gold-bearing quartz veins and alteration zones within large
regional-scale shear zones. The Macraes Flat deposit on the
Hyde-Macraes Shear Zone is the type example, although it was
formerly mined as a quartz lode deposit. The Hyde-Macraes
Shear Zone consists of a 30 km long shallow dipping shear zone
0 200 km that is semi-concordant to the foliation in the host schist. The
shear zone is up to 160 m thick, and consists of brecciated and
FIG 1 - Location of mesothermal gold mineralisation in sheared graphitic, pelitic schist separated from the host
New Zealand (Christie, 2002). psammitic schist by well-defined upper and lower bounding

TABLE 1
Characteristic features of New Zealand Palaeozoic and Mesozoic mesothermal gold deposits (Christie, 2002).
Palaeozoic deposits Mesozoic deposits
Geographic occurrence West Coast, Nelson and Fjiordland Marlborough, Southern Alps and Otago
Mineralisation style Quartz lode and disseminated Au in shear zones Quartz lode and disseminated Au in shear zones
Operating mines Globe-Progress in development Macraes Flat
Maximum known resource in one deposit greater than 2 M oz Au greater than 5 M oz Au
Lode dimensions L D W 1070 m 1000 m 14 m 1800 m 150 m 21 m
Host rock Greywacke Schist
Age of host rocks Cambrian-Ordovician Permian-Triassic
Age of metamorphism Silurian-Devonian Jurassic to Cretaceous
Age of mineralisation Silurian-Devonian Early Cretaceous
Geologic terranes Buller Haast, Caples, Torlesse
Metamorphism Lower greenschist facies, weakly cleaved Greenschist facies, strongly foliated
Structural controls Faults, shears and folds. Lodes parallel strike of Faults and shears. Lodes parallel axis of schist
fold axes. Density of quartz veining belt
Hydrothermal alteration Sericite, carbonate, arsenopyrite and pyrite Sericite, arsenopyrite and pyrite. (Graphite at
Macraes)
Lithological controls Sandstone/mudstone, and bedding Psammitic/pelitic/interlayered schist, and
bedding
Potential source influences Nearby granites Greenschists and metacherts
Main metallic minerals Gold, arsenopyrite, pyrite Gold, arsenopyrite, pyrite
Minor metallic minerals Local stibnite Local stibnite and scheelite
Main non-metallic minerals Quartz and carbonate Quartz and carbonate
Minor non-metallic minerals Feldspar and chlorite Graphite at Macraes
Geochemical signature Au, As, Sb Au, As, Sb, W

240 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 PROSPECTIVITY MAPPING USING GIS WITH PUBLICLY AVAILABLE EARTH SCIENCE DATA
thrusts. The gold is generally fine grained and closely associated
with quartz, pyrite and arsenopyrite. Quartz lode K-Ar and Rb-Sr
age determinations from Bendigo, Oturehua, Bendigo and
Glenorchy suggest mineralisation was formed in the Early
Cretaceous (ca. 100 - 145 Ma), after regional metamorphism, at a
late stage in the uplift and cooling history of the schist. Skippers,
Macetown and Carrick are inferred to be Miocene in age and
Macraes Flat mineralisation has been dated at about 140 - 150
Ma by Ar-Ar, and K-Ar and Rb-Sr methods (Christie, 2002).
Most researchers working on the genesis of the Mesozoic
schist gold deposits consider that the ore fluid for the quartz vein
and regional shear zone deposits was metamorphic, generated by
dehydration reactions within the sedimentary pile at the
greenschist to amphibolite facies transition (Craw and Norris,
1991). In contrast, some workers, based on micro-analytical
studies of fluid inclusions, suggest that the ore fluids were a
mixture of magmatic and meteoric rather than metamorphic
fluids (De Ronde et al, 2000). Graphite in the host rock sequence
may have played an important chemical role in the deposition of
gold.
The Mesothermal Gold Prospectivity Project completed by
GNS and Crown Minerals has helped define a regional
exploration model for Mesozoic Gold mineralisation in Otago by
identifying the most important geological spatial correlations
with gold mineralisation as discussed below. Table 1 highlights
the most important exploration parameters at a regional
exploration scale that were used as a basis for the Otago
prospectivity modelling.
SPATIAL DATA MODELLING METHODOLOGIES
Most exploration targeting is currently performed by searching
prospect information from mineral occurrence databases. While
this type of analysis has been effective in the past, many areas
have now been well explored and this type of approach will not
find new or buried mineral deposits. Effective targeting can only
be done if all data are compiled and integrated in a way that
matches the mineralisation model being sought. The data
available to assess the potential of an area for mineralisation
come from various modern day exploration campaigns, research
organisations and government surveys. These data are not only
diverse but voluminous, including regional geology,
geochemistry, remote sensing and geophysical data, and make
the task of interpretation difficult. It is critical for exploration
targeting that effective analysis of the available datasets is carried
out with respect to each other and that only the relevant factors to
the exploration model being used are extracted and combined
into a single mineral potential map by using spatial data
modelling techniques.
Spatial data modelling is a rapidly developing technique that is
increasingly being used in geology (Bonham-Carter et al, 1988;
Bonham-Carter, 1997; Agterberg et al, 1993; Partington, 1999,
Raines, 1999; Partington, 2000a; Partington 2000b; Partington et
al, 2002, Rattenbury and Partington, 2003; Tangestani and
Moore, 2003), other spatially based sciences such as
Archaeology (Mensing et al, 2000) and by government
organisations such as Crown Minerals; the New Zealand
Department of Mines (Partington et al, 2001; Partington et al,
2002; Partington and Smillie, 2003; Rattenbury and Partington,
2003), United States Geological Survey (Boleneus et al, 2001;
Mihalasky, 2001) and the Canadian Geological Survey for
resource assessment (Bonham-Carter et al, 1988). Government
agencies use these techniques to make land-use decisions in
relation to competing land-uses. These techniques are also
increasingly being used by leading exploration and mining
companies, but not to the same level as government agencies. It
is not clear why the industry lags behind as these techniques are
ideally suited for synthesising large spatial data sets for
reconnaissance exploration so allowing more focussed targeting
and risk reduction at an early stage of exploration (eg Henley,
1997).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
A variety of new tools are available for use with computer
aided geographic data management systems or Geographic
Information Systems (GIS) for evaluating the distribution of
spatial data in a statistical framework (Atterberg et al, 1993;
Bonham-Carter, 1997; Looney, 1997; Kemp et al, 2001;
Tangestani and Moore, 2003). The creation of derivative spatial
data maps that can be used by the modelling software from base
geological, geochemical and geophysical data is the key to
creating successful mineral potential maps (Partington et al,
2001; Partington and Smillie, 2002; Rattenbury and Partington,
2003). These derivative datasets must be reclassified in a way
that matches the mineralisation model being used; in this case
Mesothermal or Orogenic Gold (Cox and Singer, 1986; Kerrich
and Cassidy, 1994; Groves et al, 1998; Partington and Williams,
2000). For example, the creation of geochemical anomaly maps
from point data values (eg Figure 2 and Figure 3) or digital
terrane models from elevation point or contour data. GIS
techniques such as buffering, density grid interpolation, grid
extrapolation, grid interpolation, and the use of expert-assigned
attributes of genetic significance were also used to create the
derivative themes (eg Figure 4 and Figure 5). These themes were
then used to calculate spatial correlation statistics between the
data themes and a training data set selected from historic areas of
gold production. A more detailed description of some of the
derivative data themes, their spatial correlation results and
implications of these to exploration models used to explore for
gold in Otago is given below and by Partington and Smillie
(2002).
The simplest type of predictive spatial analysis is where maps,
with the chosen input variable(s) represented by a series of
integer values, are combined together using arithmetic operators.
For example differing lithologies can be reclassified into numeric
values or geochemical data can be interpolated into a raster grid.
This type of analysis takes no account of the relative importance
of the variables being used and is based on expert opinion. Fuzzy
Logic techniques address the problem of the relative importance
of data being used, but this technique still relies on expert
opinion to derive weights that rank the relative importance of the
variable for the map combination. An example of the use of this
technique in mineral exploration is given by Tangestani and
Moore (2003). Weights of Evidence, in contrast uses statistical
analysis of the map layers being used with a training data to
make less subjective decisions on how the map layers in any
model are combined. Bonham-Carter (1997) gives a good
summary of the maths and algorithms used in Weights of
Evidence and Partington (2000), Partington and Smillie (2002)
and Rattenbury and Partington (2003) give examples of how this
technique can be applied to geological datasets. Weights of
Evidence is a Bayesian statistical approach that allows the
analysis and combination of data to predict the occurrence of
events. It is based on the presence or absence of a characteristic
or pattern and the occurrence of an event. The technique was
initially developed as a diagnostic tool in medicine. In spatial
analysis, it has been used extensively in the mineral and mining
fields. An estimate of the (prior) probability of the occurrence of
granite gold mineralisation can be calculated from the total
number of known deposits distributed over the region being
targeted divided by the area of that region. A unit area is chosen
that represents the potential areal extent of known mineralisation
and is used as a grid for the spatial calculations. A probability or
statistical value of importance can then be calculated for all
geological features that are part of the exploration model. This
probability is based on the prior probability and the presence or
absence of the geological feature in question. The odds of
occurrence (logits) are then used to combine the various
statistically valid geological layers that represent the
mineralisation model to produce a probability map. This map
then defines the probability of finding the mineral at a point on a
grid covering the area being explored.
241
G A PARTINGTON and M J SALE

FIG 2 - Geochemical anomaly map for rock chip gold in Otago. Blue is <0.1 ppm Au. Yellow 0.1 - 1 ppm Au and red is >1 ppm Au.
Note the lack of data over the Macraes deposit due to data being held in closed-file.

FIG 3 - Gold in stream sediment samples averaged into stream catchment areas in Otago with blue less than 3 ppb Au,
light blue 3 - 17 ppb Au, yellow 17 - 33 ppb Au and red >33 ppb Au.

242 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 PROSPECTIVITY MAPPING USING GIS WITH PUBLICLY AVAILABLE EARTH SCIENCE DATA

FIG 4 - Faults that were formed pre to syn mineralisation, buffered at 100 m intervals in Otago. The buffers were used to test the spatial
relationships between various orientations of faults, fault movement sense and subtle variations in fault orientation with gold mineralisation.

FIG 5 - Quartz vein density maps of mineralised veins in Otago. These maps were used to test the spatial relationship of the density of
quartz veining in an area to economic mineralisation. The areas of a high density of quartz veining should represent areas of high fluid flow
in the crust during the mineralising event.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 243

G A PARTINGTON and M J SALE
SPATIAL DATA CORRELATIONS IMPLICATIONS
FOR CURRENT EXPLORATION MODELS
Data compiled prior to the modelling stage for both the regional
Otago prospectivity model and detailed Otago prospectivity
model were reclassified in accordance with the mineralisation
models described by Christie (2002), and as discussed above and
in Partington and (Smillie, 2002). These data were used to
calculate spatial relationships between the exploration model
data layers and the data layer containing the spatial data to be
modelled; in this case historic occurrences of hard rock gold
mineralisation with significant production or historic drill hole
intersections of gold mineralisation. These mineral deposit
locations for hard rock gold mineralisation were extracted from a
national mineral deposit database held by GNS. Following data
validation, the prospect database was then reviewed, checking
data and excluding all prospects that were classified as alluvial or
unrelated younger gold mineralisation. A training data set was
then subset for the regional Otago prospectivity model from this
database by selecting those prospects with historic production
recorded in the database (Partington and Smillie, 2002). This
training data-set accounts for 60 per cent of the total number of
hard rock mesothermal gold deposits in the database and gave
182 training sites for the regional Otago prospectivity model.
These training data were further subset according to production
greater than 10 000 ounces, rechecked in the field with a GPS
and locations modified for the detailed Otago prospectivity
model. The 182 training sites for the regional Otago
prospectivity model and 22 training data sites selected for the
detailed Otago prospectivity model are shown on Figure 6.
FIG 6 - Location of mines used as training data as black crosses for the regional Otago model. Mines with significant production are shown
as larger crosses, which were used to choose the training data for the detailed Otago model. Exploration permit outlines are shown as black
lines.
244 Adelaide, SA, 19 - 22 September 2004
The key derivative exploration data that were tested for
geological spatial association with known mineralisation in both
studies were the same and included:
host rock and terrain preference of known deposits;
proximity to macroscopic fold hinges;
proximity to concentrations of mesoscopic folds;
lithofacies (eg sandstone:mudstone occurrences, greenschist
horizons);
correlation with density of quartz veins;
the relative structural depth of Mesozoic deposits, particularly
in relation to the textural zones, isograds and the foliation
thickness and mica grain size work of Mortimer (2001);
metamorphic grade discontinuities as shown in Mortimer
(2000);
relationship to mesoscopically folded versus platy schist;
relationship to the dip of foliation; and
geochemistry of host rocks.
As a first step in the spatial correlation calculation, a 200 by
200 metre grid was generated over the regional Otago area and a
40 by 40 metre grid over the detailed Otago area. The size of the
grids were chosen as the minimum scale that the models should
be used at. The spatial correlations were calculated assuming the
known deposits have a 0.7 km2 sphere of influence, which was
assumed to be the smallest footprint for economic mineralisation
in the Otago area.
PACRIM 2004
 PROSPECTIVITY MAPPING USING GIS WITH PUBLICLY AVAILABLE EARTH SCIENCE DATA

As described by Bonham-Carter (1997), the spatial correlation
(prior probability) of a feature can be calculated by using the
relationship of the area covered by the data variable being tested
and the number of training data points (Weights of Evidence
technique). This produces a W+ result for when the feature is
present and a W- result when the feature is absent. A contrast
value C is then calculated from the difference (Table 2). The
standard deviations of W (Ws and Cs) are calculated, from which
the Studentised value of the contrast (StudC) can then be
calculated (the ratio of the standard deviation of the contrast Cs
to the contrast C). StudC gives an informal test of the hypothesis
that C = 0 and as long as the ratio is relatively large, implying the
contrast is large compared with the standard deviation, then the
contrast is more likely to be real (Table 2). Ideally a StudC value
larger than (-)1.5 can be considered as a positive or negative
correlation (Bonham-Carter, 1997). This ratio is best used as a
relative indicator of spatial correlation, rather than an absolute
sense. In this study a strong correlation is inferred from StudC
values >3.0, moderate correlations inferred from StudC values
between 3.0 - 1.5, weak correlations inferred from StudC values
between 1.5 - 0.5 and poor correlations inferred from StudC
values <0.5.
A more detailed discussion of the methodologies used to
develop the various data themes for the spatial correlation
analysis are given in Partington (2000) and Partington and
Smillie (2002). Table 2 summarises the results of the analysis,
provides a summary of the methodology used to create the map
theme and compares the spatial correlations results for each
model.
The most important exploration parameters derived from the
spatial analysis that should be considered when exploring for
gold mineralisation in Otago are given in Table 3. The
exploration parameters place the genetic model for mesothermal
gold mineralisation in Otago into an exploration context. These
results can therefore be used to develop realistic models for
targeting during more detailed scale exploration. Interestingly,
the main predictors of mineralisation in Otago are geological and
can be derived from geological maps or collected in more detail
in the field during prospect scale mapping. With the advent of
more sophisticated geophysical techniques good field based
geological mapping has become less fashionable. However, the
spatial correlation data from both Otago prospectivity models
demonstrate the value contained in geological maps when
assessing an areas potential for hosting mineralisation.

TABLE 2
Summary of derivative data maps created and analysed for spatial correlations with known gold mineralisation in Otago.

Spatial variable Measure Technique Variable ID Reg Reg Pros Pros

C StudC C StudC
Foliation style Foliation style as a proxy for Interpretation of field mapping of rock S2 Foliation (2) 2.2 11.3 2.80 3.77
metamorphism structure.
Fold style Fold style as a proxy for deformation Interpretation of field mapping of rock F2 (2) 2 11.4 2.00 4.60
structure.
Foliation Mapped metamorphic and structural Zones interpreted from petrographic "-5.5*(4)" 3.07 6.0034
Thickness grade, association with analysis.
metamorphism
Distance From Crustal scale structural control 1 Buffered faults 20 000 m at 200 m 3800 (1) 1.2 7 1.11 2.59
Crustal Faults Order faults chosen only. Young faults intervals around faults.
excluded.
2nd Order Faults Structural control at a local level. 2nd Buffered faults 20 000 m at 200 m 2000 (1) 0.2 0.8 2.89 6.73
Order Faults Chosen. Young faults intervals around faults.
excluded.
Fault Orientation Structural control by preferred fault Used Linanal USGS tool to subdivide NNW Trend (1) 0.6 2.7 1.77 2.29
orientation. Young faults excluded. faults into statistically valid subsets.
Shear Zones Correlation with ductile stuctures Buffered shear zones 20 000 m at 3000 (1) 2.8 6.5
subdivided from faults database. 200 m intervals, using SDM buffer
Young faults excluded. tool.
Fault Intensity Structural control at a regional scale Density modelled in Arcmap, using High Density (2) 1.6 9.6 3.65 8.38
as determined by intensity of faulting. density modelling tool in spatial
Young faults excluded. analyst.
Fault Mineralisation controlled by Used Linanal USGS tool to subdivide NW-NE 0.6 1.7 2.21 2.80
Intersections intersecting faults. Young faults faults into NE and NW trending faults intersections (1)
excluded. then analysed these for intersections
along the faults every 100 m.
NNW Fault Jogs Variation in average azimuth, which Used Linanal USGS tool to subdivide Jog (1) 3.43 4.85
should define fault jogs. Young faults faults into NW trending faults then
excluded. analysed these for deviations in trend
along the faults every 100 m.
Fold - Fault Mineralisation controlled by Used intersection lines tool to identify 800 3.3 9.2 4.13 6.56
Intersections intersection of folds and faults. fault-fold intersections then buffered
Young faults excluded. 100 100 m to test correlation.
Folds Structural control at a local level. Buffered folds 10 000 m at 200 m 1300 0.7 1.6 2.32 5.23
intervals around folds within 2500
metres of known workings using ESRI
buffer tool.
NNW Folds Structural control at a local level. Buffered folds 10 000 m at 200 m 1300 2.77 6.49
intervals around folds within 2500
metres of known workings using ESRI
buffer tool.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 245

G A PARTINGTON and M J SALE

TABLE 2 (continued)
Summary of derivative data maps created and analysed for spatial correlations with known gold mineralisation in Otago.

Rock type and Rock composition. Qmap geology. Pelitic Schist 2.8 16.1 1.99 4.35
stratigraphy
combined
Rock type and Rock composition. Qmap geology. Schist 2.5 16.3 0.89 2.01
stratigraphy
combined
Depth Depth of burial Interpretation of field mapping of rock Mid Crust 3.2 6.3
structure
Metamorphic Mapped metamorphic and structural Zones interpreted from field mapping. IV 0.6 4 0.67 1.5602
Textural Grade grade, association with
metamorphism
Qtz Vein Density Density of quartz veining representing Create point theme from linear quartz High (2) 4.3 27 6.07 13.79
possible centres of fluid flow. vein theme. Then create density map
from points.
All Quartz Veins Relationship of mapped quartz veins Buffered out from mapped veins. 300 (1) 6.66 10.67
to economic mineralisation.
Rock As Geochemical pathfinder for gold, Calculate anomalous values using 90 > 2500 ppm As (2) 3.6 14.1 2.8761 4.4244
which has a larger geochemical percentile to cut the data and then the
signature within alteration halos 90 percentile for anomalous threshold.
around gold deposits. Use nearest neighbourhood gridding
tool with 400 m search to create grid.
Rock Au Rock chip samples for gold Calculate anomalous values using 90 >0.6 ppm Au (2) 3.9 14.6 4.076 3.8583
percentile to cut the data and then the
90 percentile for anomalous threshold.
Use nearest neighbourhood gridding
tool with 400 m search to create grid.
SS As Geochemical pathfinder for gold, Use probability graphs to estimate SS As> 20 ppm (2) 3.1 15.2 3.3556 3.2312
which has a larger geochemical anomalous populations. Use point
signature within alteration halos averaging tool to calculate max and
around gold deposits, especially with average values for relevant catchment
dispersion down rivers and streams. areas. Develop grid using anomalous
threshold values.
SS Au BLEG and ordinary sieved stream Use probability graphs to estimate 2>20 x Anomalous 1.2 4.7
sediment samples analysed for gold. anomalous populations for each (2)
Both BLEG and Sieved stream collection methodology. Normalise the
sediment samples have similar data using threshold values. Use point
statistical distributions. averaging tool to calculate max and
average values for relevant catchment
areas. Develop grid using anomalous
threshold values.
Soil As Geochemical pathfinder for gold, Use probability graphs to estimate >40 ppm As (2) 2.5251 5.071
which has a larger geochemical anomalous populations for each
signature within alteration halos collection methodology. Grid soil data
around gold deposits. using the Local Polynomial
Interpolation algorithm in Arcmap.
Use anomalous thresholds to develop
anomaly grid.
Soil Au Rock chip samples for gold Use probability graphs to estimate >=35 ppb Au (2) 2.9735 5.9346
anomalous populations for each
collection methodology. Grid soil data
using the Local Polynomial
Interpolation algorithm in Arcmap.
Use anomalous thresholds to develop
anomaly grid.

PROSPECTIVITY MODELLING RESULTS
Once the prior probabilities for each data layer are calculated, it
is then possible to calculate a post probability model by
combining the various data layers. These are calculated by
weighting the values of each cell in the data layer according to
their prior probabilities and then adding the weighted values of
each data layer together (Bonham-Carter, 1997). A model for the
prospectivity of gold mineralisation was constructed using the
grid themes listed in Table 4 for the regional Otago prospectivity
model (Partington and Smillie, 2002) and Table 5 for the regional
Otago prospectivity model. The model was developed using
ARC SDM software through spatial analyst in ArcView. The
themes were all converted into binary grid themes where possible
to speed up the processing time. The number of variables used
was also kept to a minimum to reduce the potential for
conditional dependence between the selected themes as
discussed by (Bonham-Carter, 1997). The model consists of a
grid response theme containing the intersection of all of the input
themes in a single integer theme. Each row of the attribute table
contains a unique row of input theme values, the number of
training points, area in unit cells, sum of weights, posterior logit,

246 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 PROSPECTIVITY MAPPING USING GIS WITH PUBLICLY AVAILABLE EARTH SCIENCE DATA

TABLE 3
Key geological and geochemical criteria for exploration for Mesozoic mesothermal gold mineralisation in order of greatest spatial
correlation.

Exploration Data Comment

Presence of quartz veins Simple and effective exploration tool to be collected during geological mapping.
Fold-fault intersections Requires good geological mapping that identifies those structures that were pre to syn mineralisation.
Au and As geochemistry of rocks, stream Mesozoic mineral systems geochemically complex. Therefore multi-element geochemical techniques
sediments and soils appropriate.
Textural grade and foliation thickness Suggests a metamorphic control at a broad scale (Mortimer, 2001).
NW trending faults Well developed structural control that can be mapped in the field.
Second order faults Similar control to other styles of mesothermal gold mineralisation.
Local variations in fault orientation Requires more detailed geological mapping in the field.
Shear zones What if the difference between the brittle and ductile structures?
Stratigraphy Good geological mapping critical.
Rock type Strong host rock control, geochemical or mechanical?
Structural intensity Suggests a structural control at a regional scale. Detailed structural mapping and analysis in GIS critical.
Regional fault intensity There appears to be a regional structural control along deep-seated faults.
Fold style Relates to degree and style of deformation. Possible depth relationship. More detailed mapping.
Foliation style Suggests a relationship to metamorphism. Studies on metamorphism important.

TABLE 4
Grid themes used in the regional Otago prospectivity model based on the mineral deposit model described by Christie (2002).

Grid theme Description Variables C StudC

Qvdenrc Quartz vein density 5 6.95 16.74
Ssas As stream sediment anomalies 2 2.80 7.86
Rkau Au rock chip anomalies 2 3.43 6.44
Qstrain Deformation intensity 3 2.50 13.32
Rkas As rock chip samples 2 2.58 8.68
Fltauden Density of faulting related to mineralisation 2 3.80 22.68
Texmet Textural zones 4 3.41 3.39
Fltmzne NE trending faults 2 3.62 22.67
Fltmznw NW trending faults 2 5.13 34.02
Ssau Au stream sediment anomalies 2 1.58 4.65

TABLE 5
Grid themes used in the detailed Otago prospectivity model based on the mineral deposit model described by Christie (2002).

Grid theme Description Variables C StudC

Qtzdenrc Quartz vein density 2 6.0654 13.7933
Ssasrc As catchment anomaly map 2 3.3556 3.2312
Rkaurc Au rock chip anomaly map 2 4.0760 3.8583
Fltdenrc Fault density map 2 3.6493 8.3801
Rkasrc As rock chip anomaly map 2 2.8761 4.4244
Schist Schist metamorphic subdivision map 5 6.1185 0.6116
Fltazrc Fault trend variation map 4 5.7362 0.5733
Ssaurc Au catchment anomaly map 2 7.1486 0.7145

posterior probability, and the measures of uncertainty. The
variances of the weights and variance due to missing data are
summed to give the total variance of the posterior probability.
The response theme can be mapped by any of the fields in the
attribute table. Various measures to test the conditional
independence assumption were also calculated. Conditional
dependence as described by Bonham-Carter (1997) is a
significant problem in most geological models, especially
between geochemical themes, which if they correlate will have
similar spatial patterns. Hence the posterior probabilities, which
will be overestimated, should be thought of as relative
favourabilities rather than true probability values. The
normalised probability attribute gives a much better measure of
probability.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 247

G A PARTINGTON and M J SALE

The results of the modelling are shown in Figure 7 and Figure
8. The models represent a typical exploration program starting
with regional data acquisition, especially geological mapping,
moving to district scale geological mapping and regional
geochemical data acquisition such as stream sediment sampling.
The regional Otago prospectivity model identifies all the historic
areas of gold mineralisation, but because of the scale of the
model it over-estimates prospectivity around the areas most
likely to host gold mineralisation. This type of analysis is
appropriate for project acquisition and permit or tenement
acquisition where the area acquired tends to be larger than the
target area. The detailed prospectivity model reduces the
potential prospective areas significantly, although the original
map pattern is similar. A statistical correlation between both
maps gives a Spearman Rank Correlation of 0.766. This suggests
the exploration model and data used for the regional Otago
prospectivity model are an adequate predictor of mineralisation
at a regional scale, but the acquisition of district scale data,
mainly geochemistry has significantly improved the mineral
potential map.
Any follow-up exploration programs should be designed to
further reduce the search area and hence any prospectivity map
should provide a more focussed map pattern. As is the case when
more detailed geochemical data are added to the model (cf
Figure 7 and Figure 8). Statistical map correlation appears to be
a powerful technique for testing the effectiveness of any data
acquisition program.
The benefits for HPD NZ are that several of the prospective
areas are now at a stage where new detailed data are required to
make a difference to the prospectivity. The modelling has helped
define the most important datasets. In this case, many of the
prospective areas are lacking detailed geological and structural
mapping and soil sampling, and the most important dataset not
used in either study as a predictor of mineralisation, drilling. The
most obvious conclusion that comes from this analysis is that the
Otago area has a large number of areas that could host gold
mineralisation that have not been tested with detailed
geochemical sampling including drilling. The probability values
derived from the model also allow a ranking of each prospect
area, which allows efficient exploration programs to be
developed that have the best chance for success.
CONCLUSIONS
The regional Otago prospectivity model developed by Crown
Minerals successfully reduced the initial search area at a regional
scale, allowing HPD NZ to effectively prioritise its exploration
effort, and acquire additional prospective areas. The model
focuses on the areas with similar combinations of geological and
geochemical variables that have recorded gold production in the
past. The detailed Otago prospectivity modelling has further
refined these search areas allowing HPD NZ Ltd to add value to
their properties in the most cost efficient way.
More case studies need to be developed to test the analysis and
interpretation of mineral potential maps as they, along with
spatial correlation analysis have the potential to change how
exploration project generation is done and allow better and more
objective exploration management decisions to be made. This
must lead inevitably to reduced costs and increased chances of
discovery.

FIG 7 - Regional Otago prospectivity map showing areas above prior probability in blue, greater than 0.1 post probability in yellow and
greater than 0.7 post probability in red. Permit outlines are in black. This map was used to help HPD NZ prioritise its regional permits in
Otago for exploration.

248 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 PROSPECTIVITY MAPPING USING GIS WITH PUBLICLY AVAILABLE EARTH SCIENCE DATA

FIG 8 - Detailed Otago prospectivity map showing areas above prior probability in blue, greater than 0.1 post probability in yellow and
greater than 0.7 post probability in red. This map was initially was developed using the same map themes as in Figure 7. The difference
between Figure 7 and Figure 8 is the scale of modelling and use of more detailed data, especially geology.

In summary, Epithermal and Mesothermal Gold Prospectivity
modelling work has successfully attracted new investment and
ideas to the exploration scene in New Zealand. The project had
an estimated cost of $NZ 250 000 and will in the next two years,
just through grass roots exploration, attract more than $NZ 10 M
in exploration investment. If a mine is discovered the return on
investment will be considerably greater.
ACKNOWLEDGEMENTS
HPD New Zealand Ltd is thanked for allowing the results from
their recent exploration in New Zealand to be used in this paper.
Crown Minerals are acknowledged for their foresight to commit
to new technologies and ways of marketing New Zealand to the
exploration industry. Gary Raines and Graham Bonham-Carter
are thanked for discussions and the anonymous AusIMM
reviewer for helpful suggestions for improving the document.
Finally, thanks goes to all field geologists who have worked in
New Zealand, whether mapping for the government or collecting
data for exploration companies. This type of work cannot be
done without that basic work and knowledge.
REFERENCES
Atterberg, F P, Bonham-Carter, G F, Cheng, Q and Wright, D F, 1993.
Weighs of evidence modelling and weighted logistic regression for
mineral potential mapping, in Computers in Geology, 25 Years of
Progress (Eds: J C Davis and U C Herzfeld), pp 13-32 (Oxford
University Press: Oxford).
Boleneus, D E, Raines, G L, Causey, J D, Bookstrom, A A, Frost, T P
and Hyndman, P C, 2001. Assessment method for epithermal gold
deposits in northeast Washington State using weights-of-evidence
GIS modelling, USGS Open-File Report, 01-501.
Bonham-Carter, G F, 1997. Geographic Information Systems for
Geoscientists: Modelling with GIS, 398 p (Pergamon: London).
Bonham-Carter, G F, Agterberg, F P and Wright, D F, 1988. Integration
of geological data-sets for gold exploration in Nova Scotia, American
Society for Photogrammetry and Remote Sensing, 54:171-183.
Christie, A B, 2002. Mesothermal gold deposits in New Zealand: mineral
deposit models, Institute of Geological and Nuclear Sciences client
report, in Mesothermal Gold in New Zealand GIS Data Package and
Prospectivity Modelling, 33 p (Crown Minerals, Ministry of
Commerce: New Zealand).
Christie, A B and Brathwaite, R L, 1999. The mineral potential of New
Zealand, Institute of Geological and Nuclear Sciences Science
Report, 99/4.
Cox, D P and Singer, D A, 1986. Mineral deposit models, United States
Geological Survey Bulletin.
Craw, D, 1998. Structural boundaries and biotite and garnet isograds in
the Otago and Alpine Schists, New Zealand, Journal of Metamorphic
Geology, 16:395402.
Craw, D and Norris, R J, 1991. Metamorphic Au-W veins and regional
tectonics: Mineralisation throughout the uplift history of the Haast
Schist New Zealand, New Zealand Journal of Geology and
Geophysics, 34:373-383.
De Ronde, C E J, Faure, K, Bray, C J and Whitford, D J, 2000. Round
Hill shear zone gold deposit, Macraes Flat, Otago, New Zealand:
Evidence of a magmatic ore fluid, Economic Geology, 95:1025-1048.
Forster, M A and Lister, G S, 2003. Cretaceous metamorphic core
complexes in the Otago Schist, New Zealand, Australian Journal of
Earth Sciences, 50:181198.
Groves, D I, Goldfarb, R J, Gebre-Mariam, S G, Hagemann, S G and
Robert, F, 1998. Orogenic gold deposits a proposed classification
in the context of their crustal distribution and their relationship to
other gold deposit types, Ore Geology Reviews, 13:7-27.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 249

G A PARTINGTON and M J SALE
Henley, R W, 1997. Risky business: The essential blending of financial
and scientific skills in the modern resource sector, in Proceedings
1997 New Zealand Minerals and Mining Conference, pp 29-33
(Crown Minerals, Ministry of Commerce).
Kemp, L D, Bonham-Carter, G F, Raines, G L and Looney, C G, 2001.
Arc-SDM: ArcView extension for spatial data modelling using
weights of evidence, logistic regression, fuzzy logic and neural
network analysis, http://ntserv.gis.nrcan.gc.ca/sdm/
Kerrich, R and Cassidy, K F, 1994. Temporal relationships of lode gold
mineralisation to accretion, magmatism, metamorphism and
deformation Archaean to present: A review, Ore Geology Reviews,
9:263-310.
Looney, C, 1997. Pattern Recognition Using Neural Networks (Oxford
University Press: NY).
MacKenzie, D J and Craw, D, 2001. A three-dimensional section through
the Otago Schist, New Zealand, in Proceedings 34th Annual
Conference 2001, pp 31-41 (The Australasian Institute of Mining and
Metallurgy: New Zealand).
Mensing, S A, Elston, R G, Raines, G L, Tausch, R J and Nowak, C L,
2000. A GIS model to predict the location of fossil packrat
(Neotama) middens in central Nevada, Western North American
Naturalist, 60(2):111-120.
Mihalasky, M J, 2001. Mineral potential modelling of gold and silver
mineralisation in the Nevada Great Basin: A GIS-based analysis
using weights of evidence, USGS Open-File Report, 01-291.
Mortimer, N, 1993. Geology of the Otago Schist and adjacent rocks,
Institute of Geological and Nuclear Sciences Geological Map, 7:
scale 1:500 000.
Mortimer, N, 2000. Metamorphic discontinuities in orogenic belts:
examples of the garnet-biotite-albite zone in the Otago schist, New
Zealand, International Journal of Earth Sciences, 89:295-306.
Mortimer, N, 2001. Foliation thickness and mica grain size: two new
ways to subdivide the Otago Schist, New Zealand, in Proceedings
34th Annual Conference 2001, pp 43-49 (The Australasian Institute
of Mining and Metallurgy: New Zealand).
Nathan, S, 1994. Geological mapping, GIS and mineral exploration into
the 21st century: the QMAP programme and allied database, in
Proceedings 28th Annual Conference 1994, pp 1-8 (The Australasian
Institute of Mining and Metallurgy: New Zealand).
Nathan, S, 1998. QMAP, national geological mapping programme, and
the implications for mineral exploration, New Zealand Mining,
23:28-30.
250 Adelaide, SA, 19 - 22 September 2004
Partington, G A, 1999. Database management and GIS: practical
considerations and use in the exploration industry, in Fourth National
Forum on Information Management and GIS in the Geosciences, pp
177-187 (AGSO Record 1999/24).
Partington, G A, 2000a. Drummond Basin prospectivity mapping,
Australian Institute of Geoscientists Bulletin, 29:1-10.
Partington, G A, 2000b. Mineral exploration in the Drummond Basin,
North Queensland, using spatial analysis in a GIS, in SIRC 2000
The 12th Annual Colloquium of the Spatial Information Research
Centre, pp 141-148 (University of Otago: Dunedin, New Zealand).
Partington, G A, Christie, A B and Cox, S C, 2001. Mineral resources
assessment for the West Coast of New Zealand using spatial analysis
in a GIS: A new exploration management and land-use management
tool, in Proceedings 34th Annual Conference 2001, pp 141-16 (The
Australasian Institute of Mining and Metallurgy: New Zealand).
Partington, G A, Christie, A B, Cox, S C, Rattenbury, M, Smillie, R and
Stigley, P, 2002. Prospectivity modelling for mesothermal gold in
New Zealand using spatial analysis in GIS, in Proceedings Annual
Conference, pp 123-128 (The Australasian Institute of Mining and
Metallurgy: New Zealand).
Partington, G A and Smillie, R W, 2002. A national-scale GIS and
prospectivity models of mesothermal gold mineralisation in New
Zealand, in Mesothermal Gold in New Zealand GIS Data Package
and Prospectivity Modelling, 72 p (Crown Minerals, Ministry of
Commerce: New Zealand).
Partington, G A and Williams, P J, 2000. Proterozoic lode gold and
(iron)-copper-gold deposits: A comparison of Australian and global
examples, in Gold 2000 (Eds: P Brown and S Hagemann) (Society of
Economic Geologists Reviews in Economic Geology 13).
Raines, G L, 1999, Evaluation of weights of evidence to predict
epithermal deposits in the Great Basin of the Western United States,
Natural Resources Research, 8(4):257-276.
Rattenbury, M and Partington, G A, 2003. Prospectivity models and GIS
data for the exploration of epithermal gold mineralisation in New
Zealand, in Epithermal Gold in New Zealand GIS Data Package and
Prospectivity Modelling, 68 p (Crown Minerals, Ministry of
Commerce: New Zealand).
Tangestani, M H and Moore, F, 2003. Mapping porphyry copper potential
with a fuzzy model, Northern Shahr-e-Babak, Iran, Australian
Journal of Earth Sciences, 50:311-317.
PACRIM 2004
Recent Developments in Electrical Exploration
T J Ritchie1
ABSTRACT
The MIMDAS geophysical data acquisition system developed and tested
by the former MIM Exploration Pty Ltd (MIMEX) is now commercially
available worldwide through Geophysical Resources and Services Pty Ltd
(GRS). The system implements several new features in its design that
give it significant advantages over conventional systems and allows for
new approaches to electrical exploration. In particular, the achievable
signal to noise levels permit exploration to depths once thought beyond
the range of electrical exploration. The flexible multi-channel
configuration lends itself to truly 3D surveys in the style of modern
seismic techniques. Each of these aspects is illustrated with a number of
case histories from both Australia and South America.
INTRODUCTION
There should be two elements to any electrical geophysical
survey: a properly conducted survey and a thorough job of
interpretation that does justice to that data. As targets get deeper
and more difficult regions are investigated, it has become
increasingly difficult to measure and record data of the required
quality.
Over the history of geophysical exploration, numerous data
acquisition systems have been developed, each taking advantage
of the best and most recent technical developments of the time.
At the start of the 1990s the state of the art for such systems was:
16-Bit analogue to digital (A/D) converters.
A relatively small number (eight) of receiver channels.
(Usually) Separate A/Ds, all housed in a single central unit.
A limited amount of storage capacity meaning that not all
data acquired could be saved. Rather only a minor amount or
the most significant results were saved and usually this was
not data but a value derived from the data.
A basic graphical interface limiting operator involvement and
intervention.
At that time, MIM Exploration Pty Ltd (MIMEX) embarked
on a project to design and build a system that reflected the best
current technology and thinking. The resulting hardware was
called MIMDAS (MIM Distributed Acquisition System).
MIMDAS retained all the best features of its predecessors and
included a lot more that previously could not be implemented.
The result was a system that had more in common with the
modern seismic reflection systems than with the existing
electrical geophysical equipment.
MIMDAS DESIGN
The key elements to the MIMDAS design were:
1. Improved resolution. This meant using two A/D converters,
one 19-Bit and the other 24-Bit. The latter is used for low
frequency acquisition (up to 4800 Hz) and the former for
speeds from there to 48 000 Hz. The increase in quantisation
is from approximately 64 000 to 16 000 000 parts.
2. Linear, un-aliased data across the entire bandwidth.
3. Multiple gain stages.
4. Recording and storage of all data acquired.
1. Geophysical Resources and Services Pty Ltd, 1 Nolina Court,
Indooroopilly Qld 4068. E-mail: ritchie@gil.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
5. Monitoring of the input current excitation to the same
accuracy as all other channels.
6. All channels synchronised to within 1.5 microseconds.
There is also a choice of standards to synchronise against
which allows the use of GPS timing when required.
7. A large number of channels. Conventional systems at the
time commonly used eight channels. MIMDAS is capable
of measuring data on 200 channels.
8. Given a large channel capacity, the acquisition had to be
done in a distributed manner. That is, to avoid a multiplicity
of cabling leading to a common point, the digitisation had
to take place as close as practical to the measurement point.
9. The system was to be suitable for all types of electrical
exploration. This includes but is not limited to induced
polarisation (IP), time and frequency domain
electromagnetics (EM), controlled source (CSAMT) and
natural source magnetoetellurics (MT).
10. The entire acquisition and processing operation is software
controlled. This allows full operator control and easy
intervention.
DEVELOPMENT AND TESTING
The system was developed and tested in a relatively short period
between initial discussions in 1995, first field tests in
Weatherford, Fort Worth, in 1996 and extended field trials in Mt
Isa later that year. Although some modifications were then made
to both hardware and software, none of the design features above
were compromised.
BENEFITS
The benefits of the above design are both improved data quality
and increased utility. The improvements in data quality allow for
exploration in areas that were previously regarded as off limits to
electrical exploration. These include vast amounts of Australia
where the cover is very thick and or conductive and large parts of
Chile affected by caliche. In the former case the reason is that
even if large input signals can be generated, they are usually
attenuated to such an extent that they dominated by noise at even
moderate separations. In the Chilean case, the nature of the cover
precludes the generation of even moderate signals. In some areas
this problem is further exacerbated by large telluric activity.
A number of specific areas of improvement will now be
discussed. Even though the system has been used for all the
techniques mentioned above, all the following examples are
taken from IP surveys.
EM coupling removal
Although there have been many attempts to remove EM coupling
by mathematical means, at best none has been shown to work
satisfactorily all the time. At worst they can fail completely. Wait
and Gruszka (1986) among others have written of the problems
with approaches and claims of this type. Several of these
algorithms have been tested in the MIMDAS processing streams
but the most successful so far has been to record over an
extended time period and let the faster decaying EM signal die
out. Figure 1 shows an IP decay derived from a MIMDAS
survey. The data has been stacked and corrected for variations in
the input waveform. The initial decay (up to ~0.3 seconds) is
251
T J RITCHIE
negative due to the influence of EM coupling. After about two
seconds, a very clean IP decay is evident. This would be
completely missed by any conventional system recording a
standard Newmont slice or similar time range.
Telluric cancellation/high contact impedance
Telluric noise is persistent, naturally occurring noise that can
dominate IP signals especially at low frequencies. As with EM
coupling there have been attempts over the years to reduce this
kind of noise. The approach adopted with MIMDAS has been to
establish a remote site sufficiently far (~30 - 50 km) from the
survey area that no part of the impressed IP signal can be detected.
Using a pair of orthogonal horizontal magnetometers and GPS
timing, a relationship is established between the measured
magnetic fields and the natural or telluric electric field at the
survey site. When IP surveying is being done, the telluric field can
be accurately predicted and hence removed from the signal. A
number of methods have been used successfully to transmit the
telluric correction to the survey area. Currently both a wireless
local area network (LAN) and a satellite link are available.
FIG 1 - Typical MIMDAS IP decay. Note that red portion is actually
negative. The blue curve is the fitted Cole-Cole decay.
FIG 3 - IP decay before (blue) and after Telluric cancellation (red).
252 Adelaide, SA, 19 - 22 September 2004
As stated this can provide great benefits at low frequency in
many areas, but perhaps the most spectacular improvements are
to be found in Northern Chile where the injected IP signals are
very weak because the caliche layer limits input currents and the
telluric activity is usually quite high. The benefits of telluric
cancellation are demonstrated by the data in Figures 2 - 5. Figure
2 displays the raw signal and the calculated telluric noise.
Examples of improved individual decays are given in Figure 3.
The data is for a dipole at N = 7 and a transmit current of only
0.25 Amps.
Figures 4 and 5 present the resistivity and chargeability
pseudosections collected near Calama before and after
correction. The improvement in data quality is obvious. Data
which was previously reliable only to N-levels of 2 - 3, after
telluric cancellation becomes useful down to N = 7 or 8.
Typically, good quality IP data can be collected out to N = 6 with
input currents less than 0.2 Amps. Operating this way there has
been no need to go to the expense of mechanical diggers or
drilling current electrodes. The end result is that IP surveys can
now be run in Northern Chile in areas once regarded as
impossible for IP surveys.
FIG 2 - Measured signal (red) with calculated natural field (green).
PACRIM 2004
 RECENT DEVELOPMENTS IN ELECTRICAL EXPLORATION

FIG 4 - Non-Telluric corrected data, Northern Chile.

FIG 5 - Telluric corrected data, Northern Chile.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 253

T J RITCHIE

2D Chargeability Inversion

BOX

540 m

0 .5 %
Cu
Significant Results :
c lip ? 300m Cu / Au from 316m
200m

FIG 6 - Kalkaroo chargeability inversion and drilling results.

L
P
02
02

L
P
02

Drill Hole LP02

At 296m intersected mafic intrusive
containing 1-5% Pyrrhotite + Pyrite

FIG 7 - Resistivity (above) and chargeability inversions from La Pampa, Argentina.

Spectral parameter estimation
Estimating spectral parameters in IP surveys has long been
suggested (Pelton et al, 1978) One of the main problems however,
has been the inability of systems to measure good quality
broad-band data especially at low frequencies. As described
above, MIMDAS using telluric cancellation can acquire data to
arbitrarily low frequencies and so overcomes this problem. It is
apparent that the data in Figure 1 is suitable for this sort of
analysis. In fact the data has been fitted to the Cole-Cole decay as
shown. Further examples may be found in Rowston et al (2003).
Increased depth of penetration
Although there are no definitive statements about the effective
depth of investigation of electrical methods, it is generally
assumed to be insufficient in areas of deeply weathered, low
resistivity terrains. At Kalkaroo in North East South Australia a
MIMDAS IP/MT survey was used to define a chargeable target
beneath over 100 m of conductive material. Figure 6 shows the
derived chargeability section and drilling results. A chargeable
target was defined beginning at 300 m depth. This was
subsequently verified by drilling as shown. The drill hole
terminates at 540 m depth.
A second example (Figure 7) comes from La Pampa, in central
Argentina. Here a drill hole has been successfully positioned to
intersect sulfides beneath more than 200 m of very conductive
(2.5 ohm-m) cover.
These examples demonstrate that using MIMDAS the depth of
investigation in even deeply weathered terrains is at least as deep
as we are currently willing to drill. While no examples are given
here it should also be kept in mind that using other methods
(MT) the above depths can be extended considerably.
Near mine environments
MIMDAS underwent its initial extended trials near MIMs Mt
Isa mine in NW Queensland. Almost immediately it became
apparent that there was a large amount of high amplitude
electrical noise in that environment. Since one of the
implementation milestones was to be able to explore around this
and other mines, a data processing method called Selective
Stacking was implemented almost immediately. Since then, this
process has been refined and others like remote referencing for
magnetotellurics (MT) (Gamble et al, 1978) added. The result is
a system that has worked in and around several working mines.

254 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RECENT DEVELOPMENTS IN ELECTRICAL EXPLORATION

MIMDAS surveys are still in fact 2D (ie receivers and

transmitters distributed in two directions) but we now measure
components of the electric normal to the profile direction. This
contributes to better defining discrete 3D targets in the
inversions. A number of surveys have now been performed using
the configuration shown in Figure 10. Using such an array,
resistivity and IP data are read simultaneously at all receiver
locations, for each current injection site. This results in a large
amount of redundant data which in turn results to better
reconciliation of drilling with prediction from the modelled data.
In surveys with compact dimensions it can also produce
operational efficiencies and cost savings.

FIG 8 - MIMDAS system operating within the Bajo de la Alumbrera

mine.

One of these was the Bajo de la Alumbrera mine in Argentina.

For this survey IP and MT were collected at the edge of the pit.
The location within the mine workings can be seen in Figure 8.
Some of the data collected is provided as modelled sections in
Figure 9. Although mine noise was high, the data is of high
quality and the results verified by drilling.

3D surveys
The recent advances in 3D-inversion software and computer
power have meant that it is now possible in many cases to use
standard office computing platforms for interpretation. However
the ability to collect electrical geophysical data in a true 3D
mode was not available using conventional ground geophysical FIG 10 - Typical MIMDAS 3D configuration.
systems. Data was collected in a series of standard 2D lines and
then inverted in 3D. It has recently been shown (Webb et al, An example of the data reconciliation can be seen in Figure
2003) that the use of true 3D rather than this psuedo-3D data 11. Here drilling results from a Cu-Au porphyry target in Eastern
with such inversion schemes provides substantial improvements Australia are displayed over the modelled sections. It is clear
in the detail of the inversion results. from this example that the 3D method has produced good results.

FIG 9 - Modelled resistivity and chargeability data Alumbrera mine.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 255

T J RITCHIE
FIG 11 - Modelled chargeability using 3D data collection and processing.
CONCLUSION
MIMDAS is now a mature combination of geophysical hardware
and software that over the past eight years has proven itself in
many varied regions. The inclusion of a robust method for
telluric cancellation of IP data has led to large improvements in
data quality and the ability to operate far more effectively in
extremely difficult conditions.
The technology and its application are still under revision
under GRS and new and improved ways of attacking traditional
problems can be expected. The appearance recently of systems
that have implemented some of the MIMDAS design features
silently endorses the approach taken. But while those systems
have realised some improvements it is the totality of the
MIMDAS design that produces the large benefits.
MIMDAS has made a significant advance in the two step
process of conducting electrical surveys and users can at last
confidently expect data that is at least as good as our ability to
interpret it.
256 Adelaide, SA, 19 - 22 September 2004
ACKNOWLEDGEMENTS
The author would like Xstrata Copper for permission to publish
the results from South America. Thanks are due to David Burt
for his assistance and the anonymous referee whose editing
improved the readability of the text.
REFERENCES
Gamble, T D, Goubou, W M and Clarke, J, 1978. Magnetotellurics with a
remote magnetic reference, Geophysics, 44:53-68.
Pelton, W H, Ward, S H, Hallof, P G, Sill, W R and Nelson, P H, 1978.
Mineral discrimination and removal of inductive coupling with
multifrequency induced polarization, Geophysics, 43:588-609.
Rowston, P, Busuttil, S and McNeill, G, 2003. Cole-Cole inversion of
Telluric cancelled IP data: ASEG Adelaide 2003.
Wait, J R and Gruszka, T P, 1986. On electromagnetic coupling removal
from induced polarization surveys, Geoexploration, 24:21-27.
Webb, D, Rowston, P and McNeill, G, 2003. A comparison of 2D and 3D
IP from Copper Hill NSW: ASEG Adelaide 2003.
PACRIM 2004
Weathering of Base Metal Deposits and Implications for
Exploration in the Lachlan Fold Belt
K M Scott1
ABSTRACT
The weathering of different types of base metal deposits in the semi-arid
environment of the Lachlan Fold Belt of western New South Wales
affects the geochemistry of outcropping ironstones (gossans) and derived
soils. Weathering of the pyrite- and pyrrhotite-rich mineralisation at the
Elura Zn-Pb-Ag deposit leads to highly acidic conditions and strong
depletion of Cd and Zn throughout the profile. There are local
attainments of high concentrations of Ag, As, Ba, Cu, Hg, Pb, Sb and Sn
in the profile and As, Ba, Pb and Sb are still highly anomalous in surficial
material (ironstone and soil). However, at Parkers Hill, in the Mineral Hill
Field, the primary Zn-Pb mineralisation is pyrite-poor and so weathering
is not as acidic as at Elura. Thus, although Ag, As, Cd, Cu, Sb and Zn
may be concentrated at various levels in the Parkers Hill weathering
profile, Cu and Zn (as well as As, Pb and Sb) are retained in surficial
ironstones in essentially the same abundances as in primary
mineralisation. Thus, variations in the suite of geochemical indicators
present in ironstones and residual soils are likely to reflect differences in
the original mineralogy and the degree of acidity produced during the
weathering of base metal mineralisation. Hence, an understanding of
weathering processes is necessary to fully interpret exploration
geochemical data.
INTRODUCTION
How weathering of base metal deposits occurs in the semi-arid
environment of the Lachlan Fold Belt of western New South
Wales determines the geochemistry of their outcrop and derived
soils and hence interpretation of exploration geochemical results.
Thus, comparison of the weathering at the pyrite-and
pyrrhotite-rich Elura mineralisation (43 km NNW of Cobar) and
pyrite-poor Parkers Hill mineralisation in the Mineral Hill Field
(65 km N of Condobolin, see Figure 1) is instructive.
SAMPLES AND METHODS
Samples from drill core, drives, pits and the surface at Elura and
Parkers Hill were crushed to <75 m with a Mn-steel mill and
chemically analysed, generally by Inductively Coupled Plasma
Atomic Emission Spectrometry (ICP-AES) and Inductively
Coupled Plasma Mass Spectrometry (ICP-MS) after a
HF/HClO4/HNO3/HCl digest. Neutron activation analysis (NAA)
was used to determine As, Au, Ba, Br, Cd, Co, Cr, Hg, Sb and W
at Parkers Hill and X-ray fluorescence spectrometry (XRF) was
used to determine major element contents at Elura. It should be
noted that Ti determined by ICP is probably low due to
incomplete dissolution of resistate Ti phases, like rutile, during
the acid dissolution process.
Guided by the chemical analyses, selected samples were
analysed by X-ray diffraction, using graphite-crystal
monochromated, Cu radiation to determine the major and minor
mineralogical phases present. The trace and minor element
contents of primary and secondary mineral phases from Elura
have been determined by electron microprobe.
ELURA
The Elura Zn-Pb-Ag orebody occurs in an area of low relief as
seven discrete subvertically plunging elliptical pods in a
1. CRC for Landscape Environments and Mineral Exploration, C/-
CSIRO Exploration and Mining, PO Box 136, North Ryde NSW
1670. E-mail: keith.scott@csiro.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
New
New
SouthSouth
Wales
Elura Wales
Broken Hill Cobar
Mineral Hill
Condobolin
Sydney
Lachlan
Fold
Belt
Victoria
100 km
FIG 1 - Location of studied deposits in the Lachlan Fold Belt, NSW.
sequence of Devonian siltstones and sandstones (eg Webster and
Lutherborrow, 1998). It was discovered in 1974, with mining
starting in 1982, when only the Main Lode mineralisation was
known. The total resource is about 40 Mt (Leevers, 2000) with a
current Indicated and Measured Resource of 15 Mt @ 8.7 per
cent Zn, 5.3 per cent Pb and 65 g/t Ag (Reed, 2004). The Main
Lode, which accounts for about 60 per cent of the total resource,
almost reaches the surface whereas the tops of the NNW trending
pods of the Northern Zone occur at about 450 m depth (Reed,
2004). Massive pyrite, pyrrhotite, sphalerite and galena
mineralisation of the Main Lode forms a vertical pipe-like body
and is completely weathered to a depth of 80 m, with partial
weathering extending to 100 m and effects of weathering still
present in cavities at depths >800 m (Reed, 2004). The orebody
(or a vein extending from it) probably originally cropped out.
However this has been slightly eroded and is now covered by a
thin red colluvial soil, rendering the mineralisation blind.
The Elura gossan profile displays an upward progression from
primary to supergene sulfides through the supergene oxidate zone
(sulfates, carbonates and arsenates) to the Fe-oxide-dominated
near-surface material (ferruginous oxidate zone, see Figure 2). The
supergene sulfides are about 3 m thick and are dominated by the
occurrence of pyrite, marcasite and galena and the absence of
sphalerite. Anglesite and cerussite are also present in this zone but
are not well developed (Figure 3). The interface with the overlying
oxidate zones is marked by a 15 cm band of blue-black sooty
chalcocite, with some digenite and enargite. At the base of the
15 m supergene oxidate zone, native silver and cassiterite occur
before giving way to beudantite-, mimetite- and hidalgoite-rich
assemblages (Figure 3). Fifty metres of hematite- and
goethite-rich material (ferruginous oxidate zone) then extends to
the surface.
Weathering results in the progressive depletion of S, Cd and Zn,
so that surface concentrations are several orders of magnitude less
than in primary ore (Table 1). In the supergene sulfide zone, the
basal galena-rich sub-zone is enriched in Ag, Cu, Pb and Tl
257
K M SCOTT

South North

? ? 0 50m

Some leaching
Supergene of sulfide
zone FERRUGINOUS OXIDATE ZONE
Water table
? Ironstone ?
Base of complete gossan Gossan SUPERGENE OXIDATE ZONE
?

?
weathering
Barite, Beudantite,
Base of partial Mimetite, Hematite
? ? Au
weathering
Ag SUPERGENE COPPER SULFIDES
Pyritic

Massive sulfide ore

Silic e ou
SUPERGENE SULFIDE ZONE
Siliceous

Fresh
siltstones
Pyrrhotitic Pyritic Pyrite, Galena, Chalcopyrite,

s
Marcasite, Anglesite, Beudantite

FIG 2 - Section through the upper portion of the Elura orebody, showing principal weathering zones (Taylor et al, 1984).

This complex sequence is considered to have formed as a

two-stage process (Scott, 2000). A period of initial oxidation
Cu sulfides
Beudantite
Hidalgoite

Sphalerite
Native Ag

Cerussite
Anglesite
Goethite/
Hematite

resulted from the interaction of oxygenated water with the

Mimetite

Galena
Quartz
Barite

Pyrite

sulfides prior to the onset of aridity during the Middle Miocene

(cf Leah, 1996). During the more arid period to the present, the
watertable fell and a Cl-overprint modified some of the earlier
formed minerals to more Cl-rich ones (eg some beudantite
10m altered to mimetite and Pb oxychlorides formed). An effect of
this complex weathering is the strong depletion of Zn and the
local attainment of high concentrations of Ag, As, Ba, Cu, Hg,
Hidalgoite Pb, Sb and Sn in the profile.
Supergene Oxidate Zone

PARKERS HILL
Rich secondary Ag-Pb-Zn mineralisation has been mined in the
Mineral Hill Field since its discovery in 1908. From that time
until the present, 2700 t Pb, 10 890 kg Ag and 73 kg Au have
been produced from 14 000 t ore (McClatchie, 1971). Such
5m
secondary ore extends to depths of 100 m in some sections
Mimetite before giving way to primary galena and sphalerite associated
with dolomite/siderite and chlorite in volcanic host rocks.
Weathering of the sulfides at Parkers Hill, in the Mineral Hill
Beudantite Field, shows a progression from carbonate-associated
sphalerite-galena rich mineralisation into about 15 m of partially
Native Ag weathered sulfides with anglesite (where the primary carbonates
have been destroyed) and capped by a thin tetrahedrite-rich
interval before passing into a 30 - 40 m thick zone dominated by
Supergene cerussite (where chlorite has been weathered to kaolinite and Fe
Sulfides
0m oxides and Pb-bearing alunite-jarosite minerals are developed
from the weathered sulfides) and an upper (20 m thick) Fe oxide
Primary
Sulfides
zone containing pyromorphite and plumbogummite. Within the
cerussite zone, a bindheimite-rich, Pb2 Sb2 O6 (O, OH), sub-zone
(45 - 49 m) and a talc-rich interval can also be identified (Figure
4).
FIG 3 - Mineralogical variation in the supergene oxidate zone at
High abundances of Ag, As, Br, Cd, Cu, In, Mo, Pb and Sb in
Elura.
both the tetrahedrite-rich top of the sulfide zone and in the
bindheimite sub-zone (Table 2) suggest that the two zones may
whereas the upper chalcocite-rich sub-zone is enriched in Ag, As, be related, ie the bindheimite-rich material may be derived from
Ba and Sn (as well as Cu) relative to primary ore. The supergene tetrahedrite-rich intervals in the profile (as previously implied by
oxidate zone shows enrichments in As, Ba, Pb, Sb and Sn relative McClatchie, 1971). The presence of Br in both these intervals is
to the primary ore and these elements are preserved in anomalous interesting considering the presence of Br (and presumably other
concentrations in surficial material (Table 1). halides) in gossanous material from the Iodide dump material.

258 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 WEATHERING OF BASE METAL DEPOSITS AND IMPLICATIONS FOR EXPLORATION IN THE LACHLAN FOLD BELT

TABLE 1
Average compositions of zones within the Elura profile (data from Schmidt, 1980; Taylor et al, 1984 and Scott and Taylor, 1989)
(majors wt per cent, minors ppm).

Zone Surface ironstone Ferruginous oxidate Supergene oxidate Supergene copper Supergene galena Pyritic ore
Depth (m) 0-2 2 - 63 63 - 78 78 78 - 80 >80
SiO2 3.8 2.1 46.3 n.d. n.d 4.56
Al2O3 5.5 1.32 0.27 0.44 <0.02 0.50
Fe 50 54 17.0 5.43 29.5 32.0
MgO <0.1 <0.1 0.05 0.06 0.02 0.57
CaO 0.09 0.22 0.04 0.04 <0.01 0.53
K2O 0.12 <0.1 0.07 n.d. <0.01 0.06
TiO2 0.87 <0.1 0.03 0.02 <0.02 0.09
P2O5 0.12 <0.04 0.06 0.04 <0.02 0.03
MnO <0.5 <0.5 <0.01 n.d. 0.02 0.25
CO2 n.d. <0.2 0.1 0.56 0.18 5.15
S 0.61 0.61 1.40 17.7 39.9 37.7

Ag <1 <1 - 100 150 4200 330 160

As 1.3 % 2.2 % 7.37 % 2.67 % 1700 5600
Au n.d. n.d. 6 2 <2 0.7
Ba 2300 400 7000 5800 810 740
Bi <1 <1 n.d. n.d. n.d. 21
Cd <3 <3 <5 25 42 210
Cu 100 900 840 51.5 % 3800 2000
Hg 0.04 1.5 n.d. n.d. n.d. 21
Mo 14 40 50 9 4 4
Pb 4.3 % 5.5 % 16.2 % 2.47 % 28.5 % 5.70 %
Sb 600 1000 6800 140 510 500
Sn 5 1-300 1800 2000 110 100
Tl <1 <1 3 60 500 44
Zn 500 1600 740 290 6500 8.82 %

Note: n.d. = not determined

840N
160E
Block Nine
155E
150E

TMH 186
Air Shaft

Section
TM

Shaft
Fault
H
18
4

Chert

300RL Fe oxide / Plumbogummite / Pyromorphite

No 5 Lode

No4 No3 No2 No1

Talc qtz vein
Ce
ru

? Bindheimite
ss

Upp
ite

er S
ed
Upp
er V
olc
250RL

BOCO Tetrahedrite

Anglesite
Up

Sphalerite/galena
pe
Up

r S lc
pe

TFR
ed
rV
o

FIG 4 - Mineralogical zoning at Parkers Hill.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 259

K M SCOTT

TABLE 2
Average composition of zones in the Parkers Hill profile (ppm, unless otherwise indicated).
Zone Gossan Fe oxide/ Talc Cerussite Bindheimite Tetrahedrite Partially Fresh
Plumbogum weathered
Depth (m) 0-1 0 - 20 20 - 35 35 - 64 45 - 49 64 - 65 65 - 81 81 - 91
Al % 1.80 1.71 1.31 1.60 0.62 1.17 2.46 2.95
Fe % 51.9 14.5 7.33 6.56 1.58 2.30 7.16 8.09
Mn 2800 2900 2400 560 140 100 890 450
Mg % 0.08 1.20 9.18 4.00 1.00 0.97 3.27 3.65
Ca 600 500 150 200 100 200 700 700
Na 170 350 410 310 130 <100 170 300
K% <0.20 <0.20 <0.20 <0.20 <0.20 <0.20 <0.20 1.22
Ti 360 390 330 220 70 100 810 560
P 6700 1200 110 110 38 69 230 150
S 200 350 190 790 1200 9.04 % 1.46 % 1.07 %

Ag 3 28 27 42 780 920 15 19
As 660 77 79 66 1800 7200 58 54
Au ppb <5 6 <5 <5 <5 <5 <5 <5
Ba 140 170 110 310 <100 <100 <100 250
Bi 0.8 0.3 0.2 0.4 0.5 12 0.4 0.5
Br 3 3 2 2 38 85 <1 1
Cd <20 <20 <20 <20 350 760 55 58
Co 37 19 12 20 9 300 36 36
Cr 24 19 8 8 <5 <5 16 13
Cu 2800 2400 4100 3800 3.85 % 12.3 % 1700 3200
Hg nd 1.4 1.7 <1 <1 <1 1.9 1.0
In nd 0.4 0.4 0.7 6.3 15 0.5 2.1
Mo 4.9 4.7 2.3 5.0 210 98 6.1 9.1
Ni 170 44 19 18 15 24 18 16
Pb % 1.25 1.58 2.90 3.00 19.9 7.74 1.80 1.21
Sb 290 240 91 260 1.29 % 5.03 % 160 420
Sr 14 13 2 4 3 2 2 4
Tl nd <0.5 0.5 0.8 3.3 <0.5 <0.5 1.2
W 25 29 9 9 <2 <2 12 14
V 56 31 16 17 8 11 21 18
Zn 1.10 % 4100 3700 4200 1700 2.46 % 2.41 % 1.52 %

The occurrence of secondary Cu minerals in this zone (Figure 4)
is similar to the Cu-enrichment that occurs at the sulfide/oxidate
interface at Elura.
The weathering profile at Parkers Hill with its extensive zone
of cerussite and retention of substantial amounts of normally
mobile elements like Cu and Zn is different to that generally
encountered in the weathering of base metal sulfides in the
western Lachlan Fold Belt, eg Elura. There the original sulfides
are pyrite-rich and very acid conditions are developed during the
weathering so that only a thin zone of cerussite is developed with
anglesite at the top of the supergene sulfide zone and a thick
sequence of Pb-bearing alunite-jarosite minerals and mimetite is
developed in the overlying oxidate zone (Taylor et al, 1984;
Scott, 2000). Under these acid weathering conditions elements
like Ag and Cu may be strongly concentrated in supergene Cu
sulfides immediately below the oxidate zone and then be strongly
leached toward the surface. Zinc is strongly leached as soon as
the host sphalerite starts to weather but Pb generally maintains
similar abundance levels in fresh and surficial material although
the actual Pb minerals present vary up the profile.
The retention of Cu and Zn, as well as Pb, in surficial samples
in similar levels to those in fresh sulfides at Parkers Hill (Table 2)
reflects the much more alkaline weathering conditions caused by
the lack of pyrite at Parkers Hill and the strong buffering action
of talc in the rocks. It also means that the base metal contents in
the surficial gossan are a good indicator of the primary ore
grades. Elements like Ag, Cd and Mo do appear to have suffered
some depletion during weathering but As, Sb and W are enriched
in the ferruginous gossan at Parkers Hill (Table 2).
IMPLICATIONS FOR EXPLORATION
Differences in the original composition of the mineralisation and
its host rocks affect the weathering products and hence the
geochemical signature in surficial rocks and soils. In particular,
the metal:sulfur ratio of the sulfide minerals has a profound
effect upon the pH during oxidation. For a base metal sulfide
MrS, in which the metal:sulfur ratio is r, the following general
equation can be written:
2MrS+(3+2r)H2O+(3+r)O2 2rM2++2SO42-+4H++4rOH-

260 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 WEATHERING OF BASE METAL DEPOSITS AND IMPLICATIONS FOR EXPLORATION IN THE LACHLAN FOLD BELT
For mineralisation dominated by sulfides with r<1, such as
pyrite FeS2 and pyrrhotite Fe7S8, as at Elura, 4H++4rOH- is
positive, reflecting acid conditions. However for pyrite-poor
mineralisation dominated by monosulfides where r = 1, such as
galena PbS, sphalerite ZnS, 4H++4rOH- is zero and pH will not
change if there is no hydrolysis of the metals.
Weathering of the pyrite-rich mineralisation at Elura leads to
highly acidic conditions and strong depletion of Cd and Zn
throughout the profile. There are local attainments of high
concentrations of Ag, As, Ba, Cu, Hg, Pb, Sb and Sn in the
profile and As, Ba, Pb and Sb are still highly anomalous in
surficial material (ironstone and soil). However, at Parkers Hill,
the primary mineralisation is pyrite-poor and, although Ag, As,
Cd, Cu, Sb and Zn may be concentrated at various levels in the
weathering profile, Cu and Zn (as well as As, Pb and Sb) are
retained in surficial ironstones in essentially the same
abundances as in unweathered material. Thus, variations in the
suite of geochemical indicators present in ironstones and residual
soils are likely to reflect differences in the original mineralogy
and the degree of acidity produced during the weathering of base
metal mineralisation. Hence, an understanding of weathering
processes is necessary to fully interpret exploration geochemical
data.
ACKNOWLEDGEMENTS
The support of the geologists at the Elura Mine over an extended
time is acknowledged, particularly, the assistance of the former
Pasminco Exploration geologists, David Hall, Ian Kelso and
Angela Lorrigan. The continued interest and assistance of CBH
Resources, Endeavour Mine (particularly Geoff Reed), is also
acknowledged. Similarly the assistance of Triako Resources Ltd
geologists, especially Ian Mackenzie and Angus Collins, is
appreciated.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
REFERENCES
Leah, P A, 1996. Relict lateritic weathering profiles in the Cobar District,
NSW, in The Cobar Mineral Field A 1996 Perspective, (Eds: W G
Cook, A J H Ford, J J McDermott, P N Standish, C L Stegman and
T M Stegman), pp 157-177 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Leevers, P, 2000. Stratigraphy and controls on mineralisation on the Elura
orebody, in Central West Symposium 2000, Extended Abstracts, (Eds:
K G McQueen and C L Stegman), pp 46-47 (CRC LEME: Perth).
McClatchie, L, 1971. Base metal mineralisation at Mineral Hill, central
western New South Wales. Geological Survey of New South Wales
Memoir 11 (118 pp + maps).
Reed, G, 2004. Exploring the Elura system, in Proceedings Exploration
Field Workshop Cobar Region 2004, (Eds: K G McQueen and K M
Scott), pp 62-65 (CRC LEME: Perth).
Schmidt, B L, 1980. Geology of the Elura Ag-Pb-Zn deposit Cobar
district, NSW. MSc thesis (unpublished), ANU, Canberra, ACT.
Scott, K M, 2000. Elemental mobilities during the weathering of the
Elura Zn-Pb-Ag orebody the influence of mineralogy, in Central
West Symposium 2000, Extended Abstracts, (Eds: K G McQueen and
C L Stegman), pp 94-97 (CRC LEME: Perth).
Scott, K M and Taylor, G F, 1989. Weathering of pyritic and pyrrhotitic
ore on 1 Drill Level and above, Elura Zn-Pb-Ag orebody, NSW. 29 p,
CSIRO Division of Exploration Geoscience Restricted Report 55R.
Taylor, G F, Wilmshurst, J R, Togashi, Y and Andrew, A S, 1984.
Geochemical and mineralogical haloes about the Elura Zn-Pb-Ag
orebody, western New South Wales, Journal of Geochemical
Exploration, 22:265-290.
Webster, A E and Lutherborrow, C, 1998. Elura zinc-lead-silver deposit,
Cobar, in Geology of Australian and Papua New Guinean Mineral
Deposits, (Eds: D A Berkman and D H Mackenzie), pp 587-592 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
261
A Spatially Enabling Relational Geologic Database of Java
Island, Western Sunda Arc of Indonesia A Step to the
Geoscience and Mining Data Model
L D Setijadji1and K Watanabe2
ABSTRACT
A relational database in which its complex contents can be navigated
through a map interface (ie spatially enabling) is relatively a new concept
yet attractive for geoscience fields. This concept is made possible using
the new breed of geographic information system (GIS) that utilises
component object model and relational database management system.
Such sophisticated yet open and highly interoperable platform makes
possible the integration of data management, analysis, and modelling
processes into single workflow. Taking benefit of this technology trend,
several initiatives emerge in geoscience to establish digital data models,
such as the petroleum industry, the North America digital geologic map
data model, and the ESRI geology data model. The mining industry
unfortunately has not followed same paths due to the lack of data model
and standards in its history.
This paper presents an effort on establishing a digital data model for
mining-related geoscience databases. The aim is to promote several
standards by proposing a preliminary data model and demonstrating that
the model is working for the mining business datasets. The case study is
the island of Java, the western part of the Sunda island arc of Indonesia.
Many types of datasets, such as geologic maps, topography, mineral
deposits, structures, geochemical and geophysical data were dealt with
and were transformed relational geologic databases of Java using the new
Geodatabase geographic data model. Intelligent GIS tools were
investigated for geologic applications, ranging from generic to specific
purposes. Subtypes and domain were used to set up constraints, while
topology rules and relationship classes were used to create links among
data layers. The linear referencing system was used to model drillhole
data. The current results show that the generic GIS and relational
database can handle typical tasks related with mining business datasets. A
preliminary data model is finally suggested.
INTRODUCTION
Although the geographic information system (GIS) has been
widely used by mining industry for handling geologic datasets,
such a digital data model is still lacking. A common practice is
that each party develops in-house standards that meet their
immediate needs without much thought on the larger picture.
The authors propose some generic standards and a data model
for building digital geoscience (mining) databases that can serve
many geoscience communities, such as industry, public agencies,
research universities, and personal users. The basic reason is that
geologists deal with the same earth regardless the fields of
interest they are working on. For a better understanding of our
complex earth, geologists should be able to share and utilise all
kinds of data collected from the earth. A common standard of
data management is a crucial part for facilitating the easiness of
data sharing.
PROBLEM DEFINITION
Creating such standards for mineral-related geoscience is difficult
for:
1. Department of Earth Resources Engineering, Graduate School of
Engineering, Kyushu University, 6-10-1 Hakozaki, Higashi-ku,
Fukuoka 812-8581, Japan. E-mail: lucas@mine.kyushu-u.ac.jp
2. Department of Earth Resources Engineering, Graduate School of
Engineering, Kyushu University, 6-10-1 Hakozaki, Higashi-ku,
Fukuoka 812-8581, Japan. E-mail: wat@mine.kyushu-u.ac.jp
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
1. there is no available data model to start with;
2. geologic datasets are too diverse, many with different
native formats;
3. the earth is a complex phenomenon; and
4. geoscientists are divided into professional segments that
hardly to accept a single standard for serving all.
Such data model initiative is expected to be successful if it is
capable to satisfy multi users both for general and specific
applications. The model must be built using a technology
platform that is generic, open and available for public. But at the
same time the selected platform is capable to accommodate the
current technology trends and anticipate the trends in foreseen
future.
METHODS
First, ongoing international projects related with geoscience data
models were evaluated. These included the North America
Digital Geologic Map Data Model (NADM), the Public
Petroleum Data Model (PPDM), and the ESRI Geology Data
Model. The authors also considered the USGS Global GIS
Project (http://webgis.wr.usgs.gov/globalgis/) for this project
provides cheap global geologic datasets. After some common
visions and technology trends used were concluded, the next step
was to choose the proper model and technology platform for this
project. The final stage was to implement using real geologic
datasets and to evaluate the results. The implementation strategy
is to use out-of-the-box package from the selected system and
avoid the developing of customised programming.
On-going data model projects
The NADM project (http://geology.usgs.gov/dm/) started in 1996
and focuses on defining the grammar, vocabulary, and contents to
be used to represent information in a geologic map database
system. The project has released the Digital Geologic Map Data
Model Version 4.3 (Johnson et al, 1999), the Geological Map
Unit Classification Version 6.1, and the NADM Conceptual
Model 1.0 (North American Geologic-map Data Model Steering
Committee, 2003). Meanwhile, the petroleum industry since year
1999 developed standards to integrate databases with spatial
engines such as the ESRI Geodatabase (Batty, Curtis and
Diakuw, 2002). On the GIS vendors side the ESRI Geology Data
Model project started in 2002 with a focus to implement the
NADM model using their GIS software. A first pass of Geology
Data Model 812 was released in April 2003 and was intended as
a starting point for people working on geologic projects (Grise
and Brodaric, 2003). The Geological Survey of New South Wales
(NSW) responded immediately this development by
implementing their conceptual geology data model using the
Geodatabase. The NSW implementation model is believed to be
the first geologic Geodatabase in the world to be fully
implemented by a government agency (Geological Survey of
New South Wales, 2004).
263
L D SETIJADJI and K WATANABE
Technology trends
All on-going projects enforce the integration of an
industry-standard relational database management system
(RDBMS) and a GIS, so that the contents of a complex,
relational database can be navigated through a map interface (ie
spatially enabling). The integration of a GIS and RDBMS will
support topologically integrated feature classes, complex
networks, relationships among feature classes, and other
object-oriented features (Zeiler, 1999; MacDonald, 2001).
In coincidence, the GIS technology recently migrated into a
non-proprietary architecture such as the component object model
(COM) that makes it more open and interoperable with other
systems. A new geographic data model is also developed towards
a more object oriented data model. These new technology trends
are attractive for geoscience fields, which are considered one of
the most difficult application tasks of a GIS. While a GIS is still
a generic software, the open technology makes possible the
integration of a GIS and other specialised systems to build an
integrated geologic analysis system.
Conceptual model of geoscience data model
As the mining industry has no data model to start with, the ESRI
Geology Data Model and NADM are adapted to initiate a data
model for mining industry. The focus is to implement existing
conceptual models into real databases and to extent the contents
with many other mining business datasets.
Any geologic events are conceptually modelled using three
basic terms, ie the Concept, Occurrence, and Description (Grise
and Brodaric, 2003). The Concept represents all geologic
vocabulary such as map unit names, textures and geologic times.
The Occurrence represents geologic events that occur within a
concept, such as a map unit name has many member polygons.
In a GIS, each occurrence is linked optionally to at most one
spatial feature (ie a point, line or polygon) and necessarily to one
Concept. The Description is an abstract class that compiles
different types of attributes. The basic design approach is to have
a set of Concepts with Descriptions, and Occurrences with
Descriptions (Figure 1). The Concept class controls vocabulary
for attributes in the Description tables. One type of description
for an Occurrence is a spatial representation. A GIS uses the
spatial representation to create a map interface to explore
contents of Description tables.
FIG 1 - The basic design of Geology Data Model
(Grise and Brodaric, 2003).
While the basic concept is simple, a complex database can be
built by compiling and creating interwoven network of many
classes of concepts, occurrences, and descriptions. The ESRI
Geology Data Model 812 included models for several geologic
objects such as lithology, stratigraphy age, structure
measurement and drillhole.
Platform and software
This project used the ArcGIS 8.3, the Geodatabase geographic
data model and many intelligent GIS tools from ESRI. During
the implementation process, the authors tried to explore
264 Adelaide, SA, 19 - 22 September 2004
out-of-the-box generic GIS tools that can be applied for mining
applications. The authors utilised the native format of the
Geodatabase, ie MS Access (.MDB) that limits the size of each
database (approximately two GB) and restricts the contents for
vector maps and tables only.
The MS Visio was used for designing the data model using a
Unified Modelling Language (UML) diagram. The UML
diagram was implemented into real databases using the
ArcCatalog module of the ArcGIS. The ArcMap module of the
ArcGIS was used to create map interfaces from spatial features.
Several ArcGIS extensions such as the Geostatistical Analyst,
Spatial Analyst, and 3D Analyst were used to evaluate analytical
performance of the Geodatabase.
Geodatabase fundamentals
The Geodatabase is an object-oriented data model with the
defining purpose to make features in a GIS database smarter by
endowing them with natural behaviour and to allow any sort of
relationship to be defined among features (Zeiler, 1999). We can
characterise features more naturally by defining our own types of
objects and behaviours using topology, domains, validation rules,
subtypes, relationship, and how these objects interact with other
objects. The contents of a Geodatabase can be spatial and
non-spatial objects. The spatial objects contain spatial
information and are implemented as feature classes. A feature
class is a collection of features that have the same geometry type,
attribute fields, spatial reference, and behaviour. A feature class
can be a point, polyline, or polygon feature. A feature dataset is
a collection of feature classes that have the same spatial
reference and a topological relationship. The non-spatial data are
implemented as object classes or stand-alone tables that keep
descriptive information about objects that are related to
geographic features. These tables are linked into their associated
feature class using a relationship class.
CASE STUDY AND DATASETS
The case study is Java island, Indonesia, as a part of the senior
authors doctoral research on the evaluation of Cenozoic
magmatism in Java in relationship with gold and copper
mineralisation. Java is located at the western part of the Sunda
arc (Figure 2) and it holds various types of earth resources (ie
minerals, petroleum and geothermal energy) as well as
geohazards (eg active volcanoes and landslides).
The authors tried to deal with various types of geologic
datasets especially those that serve for the geological mapping,
mineral exploration, mining and geothermal industry. Datasets
were variable from generic such as digital elevation model
(DEM), stream, road and geologic maps into more specific types
such as drillhole, ore deposits, radiometric data, fluid inclusion,
gravity and geochemical survey. Data sources were local
agencies, reports and publication, foreign agencies and websites.
The GIS was also used to generate data, such as producing
topographic vector maps from the raster elevation maps.
RESULTS
Two geoscience data models are simultaneously studied from
Java datasets, ie data models for mining and geothermal. This
effort is a realisation of the idea of supporting different
geoscience applications using single platform. This paper
describes only the results on the study of mining data model,
while the geothermal data model is described in another paper
(Setijadji et al, in press).
Figure 3 shows the general structure and different relational
databases developed from Java datasets as seen using the
ArcCatalog module of ArcGIS. There are four relational
databases created based on the similarity of data types and size
consideration. Those are the JavaGeoscience.MDB,
PACRIM 2004
 A SPATIALLY ENABLING RELATIONAL GEOLOGIC DATABASE OF JAVA ISLAND

FIG 2 - The geologic setting of the study area and locations of major metal deposits. The geologic province map is taken from the USGS
Global GIS data, while the backdrop raster elevation map is the USGS GTOPO30 map that was also used to create vector topographic
maps of the islands and surrounding undersea.

FIG 3 - The structure and contents of relational geoscience databases developed from Java datasets.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 265

L D SETIJADJI and K WATANABE

JavaGeothermal.MDB, JavaExplorationProject.MDB and

JavaMetalDeposit.MDB. As raster maps cannot be modelled
using native Geodatabase, they are currently managed within
folders of operating system data management. Following are
descriptions of the results of the data model study, ranging from
generic to more complex modelling.

Modelling generic database

Building a generic relational database is easy as long as we have
the original data in common digital formats. The regional
geoscience datasets of Java were used to build a generic
relational geoscience database of Java, ie the
JavaGeoscience.MDB (Figure 3). This database was created
from many individual geologic datasets of Java and surrounding
areas. Original datasets were of different GIS formats, such as
shapefiles and coverages, and of text files and tables. All were
transformed into single, relational container. Spatial geologic
FIG 4 - Three dimensional visualisation of the Java trench of
datasets were grouped as feature classes and feature datasets (as
western Sunda arc from the digital geoscience database.
point, polyline, or polygon), while tables were managed as
stand-alone tables directly under a Geodatabase. This result
shows that the Geodatabase is capable to compile many types of
geologic datasets (ie spatial and non spatial) into single Modelling geologic relationship
container. While complex modelling is yet to apply, this
capability has already given a significant benefit from the data A small dataset from the Cineam gold mine in Tasikmalaya
management point of view. district, West Java (Widi and Matsueda, 1998) was selected to
study modelling geologic relationship using Geodatabase. The
Some good practice within the Geodatabase, such as the use
Cineam deposit is a cluster of small epithermal
of a real world spatial reference and appropriate
gold-silver-telluride deposits, located in the eastern part of the
three-dimensional spatial information, directly support the
southern West Java Mountain Range (Figure 5). Ores are hosted
standardisation effort. Therefore geologic data managed within
the Geodatabase can be easily integrated and visualised, even by volcanic rocks of lower members of the Jampang Formation
three dimensionally. Figure 4 shows the 3D visualisation of the of Oligo-Miocene age that were mapped as three different units.
terrain topography and subduction slab of the western Sunda These units were intruded by diorite, dacite and hornblende
Arc. This demonstrates that the contents of relational database andesite of middle Miocene. A major orogenic event occurred in
are truly georeferenced and they comply the three-dimensional the middle Miocene that an uplifting resulted an anticline and
nature of earth. was followed by intrusions and hydrothermal activities

FIG 5 - Geologic map of the Cineam epithermal gold, Tasikmalaya, West Java (drawn from Widi and Matsueda, 1998).

266 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A SPATIALLY ENABLING RELATIONAL GEOLOGIC DATABASE OF JAVA ISLAND

producing gold mineralisation. A strike-slip fault (Citambal

fault) occurred after the mineralisation. Information related with
ore is alteration minerals, gangue and ore minerals, K-Ar dating
and fluid inclusions.
This description was translated into Geodatabase as several
feature classes, tables and relationship classes. Those were
compiled inside a feature dataset named Cineam within the
JavaMetalDeposit.MDB (Figure 3). First, all concept names
present (map unit names, formation name, district name, ore and
mining names, and structure names) were listed to build a local
concept table and to set up domains and subtypes for
implementing validation rules on database contents. Another
concept table was created for standard geologic vocabulary
adapted from the NADM (for the map unit classification names)
and other sources (for geologic time scale, rock unit rank,
composition, structure, texture, fabric and relationships). Four
feature classes were then created: GoldQuartzVein, Lithology,
Mine and Structure. The feature classes were connected to multi
attribute tables through relationship classes. As the Geodatabase
is an object-oriented data model, relationships were regulated on
each polygon so that two polygons of the same map unit may
have different attribute information.
Figure 6 shows an example of data exploration result on this
small but relatively complex database. A query on a gold-quartz
FIG 6 - Interwoven attribute information associated with a
vein ore body (ie occurrence ID 16) has returned bunch of
gold-quartz vein occurrence as exposed using a map interface
information, such as its radiometric age, gangue and ore minerals
query on the relational database of mineral deposit of Cineam,
composition, and the host rocks. In turn, each attribute may have
West Java.
also links to its own attributes such as the rock unit with its
name, composition, age and relationships with other rock units or
structures. were carried out within the Southern Mountains of Eastern Java
(JICA-MMAJ, 2002; JICA-JOGMEC, 2004). Datasets were
received from JOGMEC mainly in the ArcView 3.x GIS
Modelling geochemical survey shapefiles and MS Excel spreadsheet formats. Datasets were then
A Geodatabase was created for managing mineral exploration organised into the Geodatabase as several feature datasets based
data in Java and was named JavaExplorationProject.MDB on the types and prospect area names.
(Figure 3). The main data source so far was from the In relationship with stream sediment samples, a data model
collaborative mineral exploration by the Metal Mining Agency study was done to build a relationship between a stream sediment
of Japan (MMAJ, now the Japan Oil, Gas and Metals National sampling point and its geochemical assays and the geologic
Corporation or JOGMEC), the Directorate of Mineral Resources information within its drainage basin area. Authors modelled this
Inventory of Indonesia (DMRI) and the Japan International by creating a relationship class to link sampling point and
Cooperation Agency (JICA) from year 2001 to 2003. During this geochemical assay attribute and another link between sampling
period, regional and semi-detailed geochemical surveys, point and drainage polygon. Figure 7 shows the map interface of
geological survey, soil sampling, geophysical survey, and drilling the stream sediment data. A new polygon map was created by

FIG 7 - The map interface of the regional stream geochemistry data in East Java.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 267

L D SETIJADJI and K WATANABE
intersecting the drainage polygon map on the geologic map. This
resulted in a new polygon map consisting of drainage polygon
and geologic names. For each geologic unit has area information,
advanced stream geochemical analysis involving the
consideration of the basement rocks is now made possible.
Figure 8 shows the result of database modelling for stream
sediment data, in which a data query using a map on a selected
sampling point (H 171 E) resulted a bunch of information on its
geochemical assay (SS Assay 498, description shown on the right
column), field geologic observation (SS Geology 498), drainage
polygon (ID no 236) and bedrock geology of the drainage basin
(Drainage Geology IDs 1915, 1649 and 1426).
Modelling drillhole data
The linear referencing system (LRS), a GIS tool traditionally
applied for modelling transportation and hydrography, was used
to model drillhole data. Using the LRS, a borehole line is
modelled as a line feature with x, y, z and m (measure length)
dimensions. Each line path has an interval measure dimension.
The path is referred the linear reference (or simply route) system
and events that occur along hole paths are referred the route
events (ESRI, 2001, 2003). All events along the paths can be
determined dynamically as long as the path ID and the measure
distance along the path (m) are provided.
This concept was implemented on the JOGMEC-DMRI-JICA
project data from the Prambon District, East Java. In this area,
several mineralisation outcrops of base metals quartz veins were
discovered and four exploration holes were drilled
FIG 8 - Relational attribute information on assay results and catchment area bedrock geology associated with a stream sediment sample as
exposed by a query through its map interface.
268 Adelaide, SA, 19 - 22 September 2004
(JICA-JOGMEC, 2004). Datasets associated with the drillholes
include geochemical assays, log descriptions, XRD analysis, thin
section, polish section and fluid inclusion analysis. Route
features of the four holes were first created and the x, y, z and m
dimensions at the two end points were set up. All attribute
information were managed as stand-alone tables within the
database, each holds attribute columns of drill ID and measure
depth (m) from which samples were taken.
Using a map interface of drillhole routes, authors successfully
explored drill-related data using route events geoprocessing tool
(Figure 9). After displaying data, customised symbols and
queries can be applied on attributes of interest, such as metal
contents, etc.
Data integration, analysis and 3D visualisation
Still from the Prambon prospect area, Figure 10 shows the
visualisation of all exploration datasets (ie topography, mineral
occurrence, soil geochemistry and drillhole) from the area three
dimensionally. This work contained some analytical process
regarding with the soil geochemistry data. Contours of Au, Ag
and As anomalies in soil were derived from a remote soil
sampling assay table, accessed through a link (relationship class)
within the relational database. The contouring was done using
the ArcGIS Geostatistical Analysis and Spatial Analyst. It is
concluded that analytical process can be performed well in a
relational database environment. The integration and
visualisation of analytical results with other geoscience datasets
was also done well.
PACRIM 2004
 A SPATIALLY ENABLING RELATIONAL GEOLOGIC DATABASE OF JAVA ISLAND

FIG 9 - Displaying drillhole attribute data using linear referencing system.

FIG 10 - The 3D visualisation of the Prambon mineral prospect. A transparent display is applied on the DEM map to show the drillhole paths
under the ground.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 269

L D SETIJADJI and K WATANABE
DISCUSSION AND CONCLUSIONS
Several cases described above address some achievements and
practical benefits on studying data model for managing
geoscience datasets using a combination of generic GIS and
relational database. The benefits of being spatially enabling for
geoscience (mining) applications are briefly summarised as
follows:
1. Serving multi application purposes. Using the same spatial
database engine, different kinds of databases for different
purposes can be created. Database structures can be simple
or complex and the attribute behaviours can be generic or
customised to model specific geologic phenomenon.
2. Data sharing easiness. The practice to apply real world
coordinate systems and common standards to serve many
different geologic datasets facilitate the easiness of data
sharing and integration across geoscience fields and
enterprises.
3. Scalable database. Databases can be small to (potentially)
unlimited size. In the case of Java, individual databases can
be integrated seamlessly into a single, bigger database of
Java.
4. Highly visual and acknowledge the 3D nature of earth. A
spatial feature of database is the key to a spatially enabling
database through the creation of a map interface. For spatial
features are fully georeferenced and many of them contain z
values, their contents can be presented and interrogated in a
3D view. This brings a new horizon to geoscientists to
visualise geologic data.
5. Key to an integrated geologic analysis system. Many
perceptions in the past on the uniqueness of geologic
problems compared to typical geographic problems (eg
Raper, 1989) have driven many developments of highly
sophisticated and proprietary systems for geologic
FIG 11 - The Unified Modelling Language (UML) diagram for the digital data model of geoscience (mining). This diagram is modification and
extension version of the ESRI Geology Data Model 812 (Grise and Brodaric, 2003).
270 Adelaide, SA, 19 - 22 September 2004
characterisation. Each system usually deals with very specific
area (eg solid 3D geology characterisation) without much
consideration on other geologic analytical areas. The
development of data model may facilitate such integration by
enforcing data input standards for geologic analytical process.
Some GIS is now open and highly interoperable system that
such integration is now also possible through development of
extensions to a GIS. Setijadji, Zolnai and Grise (2003)
evaluated some integration strategy and demonstrated on a
modelling process of ore from drillhole data.
After working with different datasets from Java, the authors
come to the preliminary data model for digital geoscience
(mining) database (Figure 11). This is a modified model from the
ESRI Geology Data Model 812. The striking feature is that all
description object classes are compiled under one description
abstract class. In this case, all description object classes inherit
attributes of the description abstract class, such as the description
ID. Object classes have been extended to other datasets, such as
stream, soil and rock geochemistry, fluid inclusion, ore deposit,
and gravity. The attribute field names within each object class are
not shown for the space limitation reason and that it is needed to
work on more data to establish standard attribute fields for each
object class. The authors propose a differentiation between the
local concept class (Concept) and the universal concept
(GeologicVocabulary).
Finally, the authors realise that many mining companies have
been in more advanced modelling and geologic characterisation
stages than those described in this paper. But the main concern
for authors is to convince geoscientists on current and future
benefits to work with an open and interoperable system.
Regardless the selection on the platform systems, the geoscience
enterprises should move forward the more open systems. At least
the generic standards will be accepted at the data input level (ie
data model) that facilitates the easiness of data sharing among
geoscientists.
PACRIM 2004
 A SPATIALLY ENABLING RELATIONAL GEOLOGIC DATABASE OF JAVA ISLAND
RECOMMENDATIONS FOR FURTHER WORK
Defining geologic vocabulary for geologic concepts is the key to
the realisation of a global geoscience data model. The NADM
project brings benefits on this model by providing standard
geologic vocabulary on the map unit names. It is expected that
the NADM team (and other international projects) will continue
working on this matter beyond the map unit vocabulary. Dealing
with more data types and different projects will improve the
existing model. Special attention should be given on modelling
geophysical datasets that are often managed in very unique,
specialised formats.
ACKNOWLEDGEMENTS
The senior author doctoral research is currently funded by the
JICA through the AUN/SEED-Net project. But he started this
data model project in spring 2003 in Redlands, CA, by support
from the ITC-ESRI International Internship Scholarship. The
Environmental Systems Research Institute Inc is thanked for
their software support. The JOGMEC, Directorate of Mineral
Resources Inventory of Indonesia and the Geological Research
and Development Centre of Indonesia are gratefully thanked for
supplying most of datasets. Purchasing some datasets and
software were partially funded by the Kyushu University alumni
association grant and the JICA AUN/SEED-Net collaborative
research project involving Gadjah Mada University, Indonesia
and Kyushu University, Japan. The early manuscript of this paper
benefitted from an anonymous reviewer.
REFERENCES
Batty, I, Curtis, T and Diakuw, M, 2002.
http://www.ppdm.org/products/model/ppdm36 PPDM Association
PPDM version 3.6: spatially enabling 2002 reference guide,
November.
ESRI, 2001. Linear referencing and dynamic segmentation in ArcGIS
8.1, ESRI Technical Paper May.
ESRI, 2003. Linear referencing in ArcGIS: practical considerations for the
development of an enterprisewide GIS, ESRI Technical Paper April.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Geological Survey of New South Wales, 2004.
http://www.minerals.nsw.gov.au/geosurvey/exp_nsw/geol_gdb.htm
Geology Data Model and Geodatabase Implementation, June.
Grise, S and Brodaric, B, 2003. http://support.esri.com/datamodels,
ArcGIS Geology Data Model, April.
JICA-JOGMEC, 2004. Report on the mineral exploration in the east Java
area, the Republic of Indonesia Consolidated report, the Japan
International Cooperation Agency and the Japan Oil, Gas and Metals
National Corporation, March.
JICA-MMAJ, 2002. Report on the mineral exploration in the east Java
area, the Republic of Indonesia Phase I, the Japan International
Cooperation Agency and the Metal Mining Agency of Japan, March.
Johnson, B R, Brodaric, B, Raines, G L, Hastings, J T and Wahl, R, 1999.
http://geology.usgs.gov/dm/ Digital Geologic Map Data Model
Version 4.3, 69 p, September.
MacDonald, A, 2001. Building a Geodatabase, 480 p (ESRI Press:
Redlands).
North American Geologic-map Data Model Steering Committee, 2003.
http://geology.usgs.gov/dm/ NADM Conceptual Model 1.0, A
conceptual model for geologic map information, 55 p, November.
Raper, J F, 1989. The three-dimensional geoscientific mapping and
modeling system: a conceptual design, in Three Dimensional
Applications in Geographic Information Systems (Ed: J F Raper), pp
11-19 (Taylor and Francis Ltd: London).
Setijadji, L D, Watanabe, K, Wahyuningsih, R and Wintolo, D, (in press).
Towards the digital data model for geothermal databases: technology
trends, fundamental concepts, case study of Java island and
preliminary data mode, paper submitted to the Proceedings World
Geothermal Congress 2005, Antalya, Turkey, 24-29 April 2005.
Setijadji, L D, Zolnai, A and Grise S, 2003. ArcGIS geoscience data
model for spatially enabling mining databases, paper presented to the
Indonesian Mining Conference and Exhibition, Jakarta, 1-3 October.
Widi, B N and Matsueda, H, 1998. Epithermal gold-silver-tellurides
deposit of Cineam, Tasikmalaya district, West Java, Indonesia,
Special Publication the Directorate of Mineral Resources,
Directorate General of Geology and Mineral Resources Indonesia No
96.
Zeiler, M, 1999. Modeling our World: the ESRI Guide to Geodatabase
Design, 198 p (ESRI Press: Redlands).
271
The Controls on Element Distribution in the Regolith at the Portia
and North Portia Prospects, Curnamona, South Australia
K P Tan1, P de Caritat2, K M Scott3, R A Eggleton1 and D Kirste1

ABSTRACT
The Portia Au and North Portia Cu-Au-Mo prospects are located on a
magnetic structure called the Benagerie Ridge Magnetic Complex,
comprising varying pyrite-bearing Proterozoic albitite, phyllite and
carbonaceous phyllite. The prospects are covered by up to 75 m of
Cainozoic sediments, which comprise from top down, surface soils,
Quaternary sediments, the Miocene Namba Formation and the
(Late-Oligocene?) Portia Unit. The Portia Unit is dominantly clay-rich
with local sand-rich base and it fills palaeo-topographic lows and a
palaeo-valley that incised into the weathered basement rock (saprolite).
This sedimentary unit hosts the majority of the Au grains found at Portia,
with additional Au grains also present in the underlying saprolite. The
North Portia prospect comprises saprolite hosted Cu-Au-Mo, with
elevated concentrations of As-Pb-V present in the northern part of the
prospect. These two groups of elements are delineated using factor
analysis, and are regarded as the pathfinder elements for the North Portia
mineralisation. Secondary Cu minerals including covellite and native Cu
are present in the lower saprolite, and are stable under reducing
conditions in the saturated zone. In the upper saprolite (unsaturated), Cu,
Au, As, Mo and Pb are associated with Fe-oxides, mainly goethite and
hematite. Saprolite leached of Fe generally contains low pathfinder
elements concentrations.
The Au grains found in the Portia Unit and saprolite contain mineral
inclusions such as arsenopyrite, galena and altaite (PbTe), and the Ag
contents range from one to ten per cent, ie the Au grains are electrum.
These mineral assemblages and chemistry suggest that the Au grains are
primary. Taking the lithologic hosts into consideration implies that the Au
grains present in the saprolite are primary in situ Au, whereas those
hosted in the sedimentary Portia Unit are detrital Au (or deep leads).
Partial dissolution and mobilisation of Au is suggested to have taken
place with subsequent precipitation of Au-rich rims on pre-existing Au
grains, and as small diagenetic Au particles (~1 - 2 m).

INTRODUCTION
The Portia Au and North Portia Cu-Au prospects are situated
approximately 50 km north of Olary at 3125S and 14025E, on
the Curnamona (SH/54-14) 1:250 000 sheet (Callen, 1990)
(Figure 1). Pasminco Exploration targeted a buried structure with
a distinct airborne magnetic signature, ie the Benagerie Ridge
Magnetic Complex (BRMC) (Figure 2) in the Curnamona
Province in 1995. Air-core drilling traverses across the strike of FIG 1 - The Portia and North Portia prospects are located on the
the magnetic eastern limb delineated the Portia prospect, which Benagerie Ridge Magnetic Complex (BRMC) (also see Figure 2),
is situated in a local zone of low magnetic signature. Au grains in the southern part of the Curnamona Province (Source: Adapted
were discovered at the unconformity between the Tertiary from de Caritat et al, in press).
sediments and the saprolite of the Proterozoic bedrock, with a
mineral resource of 270 000 t ore at 7 g/t Au inferred (Skidmore
and Fielding, 1999). Follow-up air-core drilling towards the
OBJECTIVES
north and along the strike of magnetic pyrite-bearing albitite When the Portia Au prospect was first discovered, the nature of
units in 1997 led to the discovery of a saprolite- and the Au grains present in the clay-rich material was uncertain, ie
bedrock-hosted Cu-Au-Mo mineralisation at the North Portia supergene or hypogene. In addition, the element dispersion and
prospect. enrichment in the regolith were not well established in the
Curnamona region covered by Cainozoic sediments. This project
was thus undertaken to establish the following aims to aid
mineral exploration in this region.
1. CRC LEME, C/O- Australian National University, Department of
This paper aims to:
Earth and Marine Sciences, Canberra ACT 0200.
determine the nature of the Au grains at the Portia Prospect;
2. CRC LEME, C/O- Geoscience Australia, GPO Box 378, Canberra
ACT 2601.
and

3. CRC LEME, C/O- CSIRO Exploration and Mining, PO Box 136,

establish the useful pathfinder elements and the factors
North Ryde NSW 1670. controlling their distribution in the Proterozoic saprolite and
Cainozoic sediments.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 273

K P TAN et al
FIG 2 - Schematic diagram of the total magnetic intensity of the
Benagerie Ridge Magnetic Complex (BRMC). Shaded polygons
are >1 10-4 SI. The Portia prospect is situated in a local area
with low magnetic intensity. The thickness of sedimentary cover
increases towards the north, ie average 6 m at Manning Granite
to > 50 m at the prospects. The black dot at North Portia denotes
the location of drillhole BEN388 (see Figures 6a and 6b).
Pathfinder elements include all elements associated with
mineralisation, or elements unique to certain marker lithology or
alteration assemblage, that can give vectors to mineralisation.
Useful pathfinder elements in the weathering environment are
those that remain in the saprolite and sediments in relatively high
abundance (ie above detection limits of most chemical analysis
techniques such as XRF).
SCOPE AND METHODS
This paper describes the regolith, bedrock and mineralisation at
the Portia and North Portia prospects. Subsequently, the element
distribution throughout the regolith units is illustrated, followed
by an investigation into the spatial distribution of pathfinder
elements in the regolith profile. The associations amongst the
pathfinder elements are also delineated using factor analysis. The
discussion focuses on the palaeo-environment and the factors
controlling the mobility and preservation of the pathfinder
elements in the regolith units.
Determination of the mineral composition of regolith materials
and bedrock was carried out by X-ray diffraction, using a
Siemens D501 diffractometer (CuK, operating at 40 kV and
274 Adelaide, SA, 19 - 22 September 2004
40 mA) with a graphite post-sample monochromator (Collimated
by 1 divergence slits). Sample preparation and analysis follow
the method of Moore and Reynolds (1989).
Eighty-six samples (including both tertiary sediments and
saprolite materials) were pulverised using a tungsten carbide mill
and the elements were analysed using a PW2400 wavelength
dispersive X-ray fluorescence (XRF) spectrometer. For major
element determination (SiO2, TiO2, Al2O3, Fe2O3, MnO, MgO,
CaO, Na2O, K2O, P2O5, and S), samples were prepared as fused
discs following the method of Norrish and Hutton (1964).
Thirty-five trace elements were determined on pressed powder
pellets using a SPECTRO X-Lab energy dispersive XRF
spectrometer. The pellets were also measured on a PW1400
wavelength dispersive XRF spectrometer for Sc, V and Cr, using
methods described in Chappell (1991) and Norrish and Chappell
(1977).
The gold specimens were mounted onto aluminium studs,
carbon coated and the morphology analysed using scanning
electron microscopy (Cambridge JEOL 6400) with an energy
dispersive system (SEM-EDX). Elements such as silica, iron,
titanium and tungsten were calibrated using pure standards. Pure
gold and silver standards were used but since the gold grains
analysed are Au-Ag alloys, the internal correcting factors on the
SEM have to be applied to determine the gold and silver
proportions. High-resolution electron microscope (HREM) has
been utilised to determine the morphology of the small
(approximately 1 m) gold particles. The small size of the gold
particles prevented the use of electron microprobe whereas the
larger Au grains have had their chemistry (Au, Ag, Cu, Te and
Hg) analysed using an electron microprobe (CAMECA
CAMEBAX wavelength dispersive system).
THE REGOLITH UNITS
Varying thickness of Quaternary and Tertiary sediments covers the
BRMC, from 6 m at Manning Granite in the south to
approximately 50 to 75 m at the North Portia and Portia prospects
respectively, where the regolith (Figure 3) includes, from the top
down: surface soils, Quaternary sediments, Tertiary Namba
Formation, Tertiary Portia Unit, saprolite and Proterozoic bedrock.
FIG 3 - Schematic cross-section diagram showing the regolith
units at Portia and North Portia prospects. Refer to text for unit
descriptions.
The soils developed in degraded dunes and present-day
floodplain deposits, consisting of 1 - 2 m of light brown fine
sands and silty loam, with soft powdery regolith carbonate and
groundwater gypsum in the subsoil.
The Quaternary sediments comprise 3 - 8 m of moderately
sorted, unconsolidated fine- to medium-grained olive brown sand
and sandy mud, interpreted as deposited under fluvial
environments.
PACRIM 2004
 THE CONTROLS ON ELEMENT DISTRIBUTION IN THE REGOLITH AT THE PORTIA AND NORTH PORTIA PROSPECTS
The Tertiary Namba Formation (50 m thick) has been
informally divided into three sub units, ie upper, middle and
lower, in accord with the stratigraphic guide by Hedberg (1976).
The upper-Namba Formation (NFU) comprises fine to coarse
sand, and grey and pink sandy muds interpreted as deposited
under fluvial environment. The middle-Namba Formation (NFM)
consists predominantly of dark-grey clay and the lower-Namba
Formation (NFL) comprises dominantly of olive-grey clay. For
the following discussion, the NFM and NFL have been grouped
as the lower-Namba Formation, interpreted as low-energy
lacustrine and wetlands deposits.
The Tertiary Portia Unit consists dominantly of light grey clay
with abundant kaolinite and variable amounts of mica (5 to 20
per cent) and minor quartz (<5 per cent), with up to 20 per cent
quartz sand present at the base of unit. A basal sand bed (1 - 2 m
thick) may be locally present, and is most commonly present in
thicker part (>10 m) of the unit. The Portia Unit is transported,
sourced from the saprolite, and filled local depressions and a
palaeo-valley across the Portia prospect (Figure 4, also refer to
discussion).
FIG 4 - 3D model of the surface topography (L1), the base of the
Namba Formation (L2), and the sediment-saprolite unconformity
surface (L3). The Portia Unit fills local depressions and the
palaeo-valley (L3), which extends from southeast of Portia and
continues towards the north, flanking the outcropping ridge that
hosts the North Portia mineralisation. Where the Portia Unit is
absent, the Namba Formation directly overlies on the saprolite (L2).
TABLE 1
Summary of the chemistry of Au grains at the Portia prospect using electron microprobe.
Regolith host Morphology Au (%) Ag (%)
Portia unit Protrusion rim 95.3 - 99.6 0.4 - 4.7
Portia unit Core 90.5 - 94.0 5.2 - 8.6
Saprolite Core 89.6 - 97.5 1.0 - 10.0
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
The saprolite, up to 90 m thick, developed from the Proterozoic
bedrock and comprises dominantly of quartz and kaolinite, with
variable amounts of goethite, hematite and mica. Smectite is
locally present in ferruginous saprolite, whereas pyritic saprolite is
derived from partial weathering of pyrite-bearing albitite.
THE PROTEROZOIC BEDROCK
The Proterozoic protolith comprises evaporite, saline mud and
organic-rich sediments. During paragenesis and metasomatism,
sodic alteration obliterated the rock matrix to form quartz-albitite
and variably altered carbonaceous phyllite (Teale, 1999). The
lithology observed from six diamond core holes at North Portia
has been classified into nine units by Teale (1999) and mainly
consists of massive to banded albitite, pyritic albitite,
calc-silicate unit and carbonaceous phyllite.
A range of minerals is present in the Proterozoic bedrock,
which includes silicates (quartz, albite, K-feldspar,
biotite-phlogopite, chlorite, tremolite, scapolite and tourmaline),
carbonates (calcite, ankerite, dolomite and siderite), phosphates
(apatite, monazite, xenotime and other minor REE-phases),
sulfides and sulfates (chalcopyrite, enargite, galena, sphalerite,
molybdenite, arsenopyrite, pyrite and barite), oxides (rutile,
anatase, haematite and magnetite) and native elements
(argentiferous gold). Fluorite is also locally present (Teale,
1999).
MINERALISATION
The mineralising events at North Portia are complex, and include
early molybdenite and chalcopyrite replacement and infill,
cross-cutting chalcopyrite-pyrite-fluorite-gold, to late stage
molybdenite-chalcopyrite and emplacement of Ba-Pb-Zn-Mo-F
in fault structures (Teale, 1999). The refraction of veins and
faults through beds of different competency also caused
brecciation and allowed the formation of Cu-Au mineralisation.
In general, the Cu-Au mineralisation is associated with the
pyritic quartz-albitite. Coarse-grained Au is present in some
sulfide-poor veins in carbonaceous quartz-muscovite-K-feldspar
phyllites. Two populations of Au grains were found. The first
population has mean compositions of 9.5 per cent Ag and the
second averages 24 per cent Ag (Teale, 1999).
At Portia, irregularly shaped Au grains (Figure 5a) are present
in the basal sand-rich part of the Portia Unit. The Ag contents
ranges from five to ten per cent (Table 1) and are similar to those
of the first population present at the North Portia prospect. Some
contain arsenopyrite inclusions (Figure 5b) and appear to form a
placer deposit. Some parts of the grains have Au-enriched growth
rims (Figure 5c, Table 1), which together with numerous pits,
grooves (Figure 5d) and voids (Figure 5e) on their surfaces,
indicate etching or dissolution. Minor amounts of small
(1 - 2 m) (Figure 5f) possibly diagenetic Au particles also occur
in the Portia Unit.
In the saprolite at Portia, irregular Au grains (larger grains
have dimensions of 800 1400 m across) contain up to ten per
cent Ag and a few have galena and altaite (PbTe) inclusions. The
Ag contents are similar to some of those in the unweathered
bedrock at North Portia, ie the first population classified by Teale
(1999). These are interpreted as in situ primary grains, although
some have secondary Au-enriched rims.
Cu (%) Te (%) Hg (%)
-- -- --
0.03 - 0.08 <0.04 0.19 - 1.32
<0.03 - 0.13 <0.04 - 0.08 <0.11 - 1.58
275
K P TAN et al

FIG 5a - SEM photomicrograph (secondary electron) of

an irregular Au grains from the Portia Unit displaying
subrounded edges.

FIG 5b - Back-scattered image of a dissected Au grain revealed the
presence of euhedral arsenopyrite (FeAsS) fully enclosed within the
Au grains. The presence of such primary mineral suggests that the
enclosing Au grain was not formed in a supergene environment.
Specimen obtained from the Portia Unit.
FIG 5d and 5e - Magnification of Au grains (secondary electron)
showing dissolution grooves and voids, indicating that the Au
grains have been subjected to chemical weathering. Specimen
obtained from the Portia Unit.

FIG 5c - Back-scattered image of a dissection of Au grain revealed

the presence of a secondary Au-enriched rim (up to 99.6 per cent FIG 5f - SEM micrograph (secondary electron) of an irregular,
Au), compared to the massive, primary, Ag-bearing core (4.5 per small Au particle surrounded by platy kaolinite in the Portia Unit.
cent Ag), ie Au electrum. Specimen obtained from the Portia Unit. Such Au particle is postulated to be of diagenetic origin.

276 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE CONTROLS ON ELEMENT DISTRIBUTION IN THE REGOLITH AT THE PORTIA AND NORTH PORTIA PROSPECTS

ELEMENT DISTRIBUTION THROUGHOUT THE
REGOLITH UNITS
The distribution of elements throughout the regolith profile
fundamentally reflects the two lithologic units, ie sediments and
saprolite. Within the saprolite, element abundance varies with
respect to saprolite, ferruginous saprolite and pyritic saprolite.
Geochemical results from bore BEN388 (447 900 mE,
6 522 600 mN) at North Portia, drilled into the mineralisation,
are used for illustration (Figures 6a and 6b).
Na2O are controlled by the phyllosilicate minerals, K is
associated with mica whereas Mg and Na are associated with
smectite. These elements show relative uniform abundance
throughout the sediments. Fe occurs as Fe-oxide mottles and the
abundance ranges from two to five per cent FeO. Abundant Mn,
up to 5000 ppm, is present in the upper-Namba Formation as
dark grey or black mottles. Elsewhere throughout the sediments,
Mn concentrations are below 200 ppm, whereas Zn and Co
averages about 80 and 20 ppm, and show affinity towards the
clay-rich sediments.

Geochemistry of the sediments Geochemistry of the saprolite

CaO has the highest abundance (three per cent) in the top 2 m The degree of weathering and the formation of secondary host
associated with the presence of calcite, dolomite and gypsum in minerals determine the element concentrations in the saprolite.
the subsoil (Figure 6a). CaO abundance elsewhere in the In the pyritic saprolite, Fe is hosted in pyrite and the other trace
sediments is less than 0.5 per cent. In contrast, K2O, MgO and elements are hosted in sulfides or native elements, eg As in

FIG 6a - Geochemical results from drillhole BEN388 at North Portia show that abundant Mn is only present in the Upper Namba Formation.

FIG 6b - Geochemical results from drillhole BEN388 at North Portia show the association between Fe and Au-Mo-As-V in the
ferruginous saprolite.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 277

K P TAN et al

arsenopyrite, Au as argentiferous gold, Cu as chalcopyrite and hosted in pyrite. The factor groups Cu-Au-Mo and As-Pb-V
native copper, and Co in pyrite. Ca, K, Mg and Na occur in the reflect the saprolite hosted mineralisation at North Portia;
primary silicate minerals such as biotite, tremolite and feldspars. whereas factor group Au represents the sediment hosted Au at
Dissolution of these minerals during weathering releases and Portia. Hence, these elements are regarded as pathfinder
removes the elements from the saprolite, unless the elements can elements.
be partitioned into secondary hosts. For example, in the
ferruginous saprolite, up to 20 per cent FeO is present and is SPATIAL DISTRIBUTION OF PATHFINDER
positively associated with Au, Mo, As and V (Figure 6b). These
ELEMENTS
elements are interpreted to be hosted in Fe-oxides. Elements that
are not significantly hosted by Fe-oxides become depleted, eg Establishing the abundances of pathfinder elements in the
Co. In the 6 m of saprolite where Fe is low (<1 per cent FeO), regolith profile across the prospects provides insights into the
Au, Mo, Cu, As and Pb are also low in abundances. geometry of the dispersion halo and the potential sampling
medium for mineral exploration. The pathfinder element groups
ELEMENT ASSOCIATION delineated using factor analyses are selected, ie Au-Cu-Mo,
As-Pb-V and Au. However, due to the significant differences in
Factor analysis was utilised to determine the the element the abundances of the elements (ie in one or more order of
distribution in the saprolite at Portia and North Portia. magnitude), indices are used to normalise element abundances to
Geochemical results obtained from air-core drill holes throughout the same order of magnitude. Three indices were formulated,
the Portia and North Portia prospects have been utilised (n >5000, which include the North Portia mineralisation index, the
courtesy of Pasminco Exploration). The first four factors at North carbonaceous phyllite mineralisation index and the Portia
Portia account for 68 per cent of the total variance (Table 2a) mineralisation index (Table 3) (Tan, 2001).
whereas the first five factors at Portia accounts for 74 per cent The North Portia mineralisation index (Xmin) ranges from 0.5
(Table 2b). The elements within the same factor group are to four and values from two to four represent the anomalous
associated spatially, and to some extent, chemically in the regolith population depicting mineralisation (Figure 7a). The geometry of
profile. There are similarities between the factors at Portia and the bedrock mineralisation exhibits a southeast-northwest trend
North Portia (eg Co-Ni), but significant differences are also across the prospect associated mainly with the banded pyritic
present (eg Au versus Cu-Au-Mo) reflecting the differences in albitite and to a lesser extent, the pyritic carbonaceous phyllite in
mineralisation styles between the two prospects. the northern half of the prospect. Expression of the dispersion
The six major elements Ca, Mn, Fe, Mg, K and Na are halo into the saprolite is generally restricted to the southern half
delineated as a factor group (Factor 1, North Portia) dominated of the prospect, where ferruginous saprolite is present, ie
by the minerals that make up the Quaternary and Tertiary elements hosted in Fe-oxides. The Portia Unit is generally absent
sediments and to a lesser extent the saprolite. The Factor group or thin (1 to 3 m) on the subsurface ridge that hosts the
Co and Ni, delineated at both Portia and North Portia, represents mineralisation. Dispersion of the elements from the saprolite into
elevated abundance in the pyritic saprolite as both elements are the Tertiary sediments is limited to the base of lower-Namba
Formation and does not extend into the upper-Namba Formation.
TABLE 2A The carbonaceous phyllite mineralisation index (Xphyllite)
ranges from one to four, and values 2.8 to four represent the
Factor analysis on geochemical results from North Portia
drillholes showing the element groups with significant factor
anomalous population hosted in the pyritic carbonaceous phyllite
loadings extracted from the first four factors. and preserved in the saprolite at the northern end of the diagram
(Figure 7b). In contrast, other non-pyritic carbonaceous phyllite
Prospect Factor 1 Factor 2 Factor 3 Factor 4 towards the eastern edge of North Portia does not host
North Portia Ca, Mn, Fe, Co, Ni As, Pb, V Cu, Au, Mo
anomalous As, Pb or V concentrations. At background
n = 3325 Mg, K, Na concentrations, V is more abundant (100 - 200 ppm) whereas As
and Pb predominate at high Xphyllite values (>3); the maximum
Spatial Sediments Pyritic Saprolite Saprolite 600 ppm V is far less than the >1000 ppm of As and Pb. This
distribution saprolite
index is not exclusively associated with carbonaceous phyllite, as
Variance 32 % 19 % 9% 8% anomalous Xphyllite may overlap with anomalous Xmin when Cu
mineralisation occurs in the banded albitite. In this case the
Xphyllite is caused by elevated As values (arsenopyrite) associated
TABLE 2B with chalcopyrite (and rare enargite).
Factor analysis of geochemical results from Portia drillholes At Portia prospect, data obtained from 20 closely spaced drill
showing the element groups with significant factor loadings
extracted from the first five factors.
holes (100 m by 100 m across) were used to produce a 3D
diagram. Mapping a smaller area allows detailed depiction of the
Prospect Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Au distribution. Portia index XPortia consists of a single element
(ie Au). On the diagram (Figure 7c), the light-grey clay unit
Portia As, Pb, V, K, M Co, Ni Fe Au actually denotes the Portia Unit, and high Au concentration
n = 1794 Mo
(>3 ppm) are found predominantly at the base of the Portia Unit
Spatial Saprolite Sediments Pyritic Sediments Portia close to the sediment-saprolite unconformity. Au mineralisation
distribution saprolite Unit also extends into the saprolite.

TABLE 3
Table showing the indices and their formulas used in producing the 3D images (Tan, 2001).
Index name Index formula Remarks
North Portia mineralisation (Xmin) log (10Au+0.2Cu+Mo) Cu Mineralisation at North Portia with varying amounts of Au and Mo.
Carbonaceous phyllite mineralisation (Xphyllite) log (As+Pb+V) As, Pb and V found to be in anomalous concentrations in the pyritic
carbonaceous phyllite.
Portia index (XPortia) Au Au at Portia prospect occurs at the Tertiary-Proterozoic unconformity and
in the saprolite.

278 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE CONTROLS ON ELEMENT DISTRIBUTION IN THE REGOLITH AT THE PORTIA AND NORTH PORTIA PROSPECTS
FIG 7a - Mineralisation index (Xmin) distribution at the North Portia
prospect (see text).
FIG 7b - Carbonaceous phyllite index (Xphyllite) distribution at the
North Portia prospect, mainly associated with the carbonaceous
phyllite bedrock (see text).
DISCUSSION
At North Portia, the pathfinder elements (Cu-Au-Mo-As-Pb) are
concentrated in the saprolite. These elements are either
associated with Fe-oxides in the upper saprolite, or as secondary
sulfides and native elements (eg native Cu, Figure 8a), together
with pyrite (Figure 8b), in the saturated and reducing lower
saprolite. Thus, factors affecting the precipitation and
mobilisation of Fe during weathering will dictate whether the
pathfinder elements will be preserved or removed from the
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
FIG 7c - Mineralisation index at Portia shows the concentration of
Au at the base of the Portia Unit and the saprolite (see text).
saprolite. The main factor controlling the stability of sulfides and
oxides is the oxidation-reduction conditions, which on a
macroscopic scale, are mainly affected by the presence of
groundwater, ie groundwater level and its fluctuations. Thus, the
elevation of the groundwater table, or zone of saturation, during
the Tertiary, prior to and after the deposition of the Miocene
Namba Formation, and the current groundwater level,
superimpose a level of control on the pre-existing distribution of
oxides and sulfides in the saprolite.
Although weathering is a continuous process and the
distribution of elements in the regolith units is the net result of
several factors, it is possible to postulate the dominant
weathering conditions from the palaeo-topography of the
bedrock and the sedimentation chronology.
Prior to the deposition of the Namba Formation, the
Proterozoic bedrock cropped out and oxidising weathering
condition prevailed above the palaeo-watertable. Weathering of
pyritic albitite unit (fine-grained quartz and albitite dominant)
would have formed thicker saprolite compared to the non-pyritic
phyllite (quartz and K-feldspar dominant). Preferential erosion of
the thicker saprolite may have partly contributed to differences in
local relief and influenced the geometry of the palaeo-valley and
local depressions. Part of the saprolite has been removed and
deposited in the palaeo-valley and depressions forming the Portia
Unit. The base of the Portia Unit in the palaeo-valley is
commonly sand-rich, suggesting deposition under fluvial
conditions. The rest of the Portia Unit is clay-rich with minor
angular floating quartz sand (fine). The Portia Unit is not
restricted to the palaeo-valley, as a veneer (2 - 3 m thick) of clay
was deposited on the northern part of the North Portia subsurface
ridge. Together with the shallow and broad geometry of parts of
the palaeo-valley (up to 1 km wide), this suggests that the
clay-rich sediments are mainly flood-plain deposits, and have
deposited under low energy environment. In areas proximal to
the palaeo-valley, primary Au grains that were preserved in the
saprolite may have been removed together with the saprolite,
transported, and deposited in the palaeo-valley forming the
placer Au deposit. The Portia prospect is situated in a depression
along the palaeo-valley, which may have acted as a sump for the
detrital Au, resulting in a localised area with elevated Au
abundance.
279
K P TAN et al
FIG 8a - Phase diagram of a Cu-Cl-S-C-O-H system, a [Cu+] =
10-7, taking into account the presence of saline groundwater. The
diagram shows the stability of various Cu phases, including native
Cu, with respect to Eh. The triangle represents the circum-neutral
pH and weakly reducing Eh of the groundwater at the Portia
prospect. The other three symbols represent groundwater taken
from water bores in the surrounding region.
FIG 8b - Phase diagram of a Fe-Cl-S-C-O-H system, a [Fe2+] =
10-6, showing the stability of pyrite under reducing conditions in
the saturated zone. In comparison, at circum-neutral pH, goethite
is only stable under oxidising conditions. The triangle represents
the circum-neutral pH and weakly reducing Eh of the groundwater
at the Portia prospect. The other three symbols represent
groundwater taken from water bores in the surrounding region.
The presence of etch pits, grooves and voids on the surfaces of
Au grains suggests that dissolution and chemical mobilisation of
Au have taken place. However, the formation of secondary Au
rims (up to 99.6 per cent Au) and small Au particles (1 - 2 m),
interpreted to be secondary Au, suggest that chemical
precipitation has also occurred. Since mobilisation of Au require
strongly oxidising conditions (Figure 8c), the preservation of
detrital Au in the Portia Unit and primary Au in the saprolite
280 Adelaide, SA, 19 - 22 September 2004
FIG 8c - The phase diagram shows the Au-Cl-S-C-O-H system,
a [Au+] = 10-11 shows that mobility of Au by Cl- complexing ligands
require a strongly oxidising condition. The triangle represents
the circum-neutral pH and weakly reducing Eh of the groundwater
at the Portia prospect. The other three symbols represent
groundwater taken from water bores in the surrounding region.
suggests that the palaeo-environment has been dominated by
weakly oxidising or reducing conditions. These conditions may
have resulted from a probable high standing water level during
the deposition of the clay-rich Portia Unit. Following a hiatus,
the reducing conditions prevailed again during the deposition of
the olive grey to dark grey clay of the Miocene Namba
Formation in a wetland/lacustrine environment.
CONCLUSION
The Portia Unit is dominantly clay-rich with local sand-rich base
and fills palaeo-topographic lows and a palaeo-valley that incised
into the saprolite. At the Portia prospect, the Au grains hosted
within this unit, have a morphology and Ag content that suggest
a detrital origin. The Au grains have subsequently undergone
various degree of dissolution. Consecutively, secondary Au-rich
rims and small Au particles also precipitated. In comparison, Au
grains present in the saprolite with inclusion of primary sulfides
(arsenopyrite and galena) and telluride (altaite PbTe) are in situ
primary Au.
Factor analysis has delineated two groups of pathfinder
elements at North Portia; these are Cu-Au-Mo and As-Pb-V. The
first group is dominantly associated with the banded pyritic
albitite whereas the second group is associated mainly with
pyritic carbonaceous phyllite. Cu enrichment in the lower
saprolite is associated with secondary Cu such as covellite and
native Cu, both stable under the reducing condition in the
saturated zone. Fe-oxides are the hosts for the pathfinder
elements in the upper saprolite in the unsaturated zone. Thus,
both the Portia Unit and the ferruginous saprolite are good
sampling medium for delineating proximal mineralisation.
ACKNOWLEDGEMENT
This work was funded by the Commonwealth Governments
Cooperative Research Centres (CRC) Program. The authors
would like to thank the staff of Pasminco Exploration,
particularly Colin Skidmore, Mike Hudson, Ian Kelso, Steve
Newberry, Bruce Godsmark, Mark Rendall and the technical
support staff for their assistance.
PACRIM 2004
 THE CONTROLS ON ELEMENT DISTRIBUTION IN THE REGOLITH AT THE PORTIA AND NORTH PORTIA PROSPECTS
REFERENCES
Callen, R A, 1990. Curnamona. 1:250 000 Geological Series, explanatory
notes, 56 pp (Geological Survey of South Australia, Adelaide).
Chappell, B W, 1991. Trace element analysis of rocks by X-ray
spectrometry, in Advances in X-Ray Analysis (Ed: C S Barrett), pp
263 (Plenum Press: New York).
de Caritat, P, Kirste, D, Carr, G and McCulloch, M, in press.
Groundwater in the Broken Hill region, Australia: recognising
interaction with bedrock and mineralisation using S, Sr and Pb
isotopes, Applied Geochemistry.
Hedberg, H, 1976. International Stratigraphic Guide. A Guide to
Stratigraphic Classification, Terminology, and Procedure, 200 p
(John Wiley and Sons: New York).
Moore, D M and Reynolds, R C, 1989. X-ray Diffraction and the
Identification and Analysis of Clay Minerals, 332 p (Oxford
University Press).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Norrish, K and Chappell, B W, 1977. X-Ray Fluorescence Spectrometry,
Physical Methods in Determinative Mineralogy, pp 201-272
(Academic Press, London).
Norrish, K and Hutton, J T, 1964. An accurate X-ray spectrographic
method for the analysis of a wide range of geological samples,
Geochimica et Cosmochimica Acta, 33:431.
Skidmore, C P and Fielding, D C, 1999. Resource potential Portia
prospect, Unpublished Pasminco Report, 18 p.
Tan, K P, 2001. Regolith geology and geochemistry of Portia and North
Portia prospects, Benagerie Ridge Magnetic Complex, Curnamona,
SA, PhD thesis (unpublished), 328 p, Australian National University,
Canberra.
Teale, G S, 1999. Mineralization on the Benagerie Ridge. Stratigraphic
subdivision, sediment provenance, mineragraphic and petrological
observations, mineral chemistry, mineralization and vein paragenesis,
U-Pb age dating, gold and silver studies, comparision with other
Proterozoic provinces, Vol 1, Report prepared for Pasminco
Exploration, 120 p.
281
Compositional Variations of White Mica in the Footwall
Hydrothermal Alteration System of the Hellyer Zn-Pb Deposit,
Tasmania
K Yang1, J F Huntington1, K M Scott1 and P Mason1
ABSTRACT
Short-wave infrared (SWIR) spectral reflectance of the hydrothermally
altered volcanic rocks in the footwall of the Hellyer Zn-Pb massive
sulfide deposit was measured with a portable PIMATM infrared
spectrometer. The Al-OH absorption band was used to derive information
on the cation composition of white mica. The wavelength range of the
Al-OH band from 2194 nm to 2222 nm reflects variation in the number of
octahedral Al (Alvi) in white mica from 3.9 to 3.2 (based on four
octahedral cations per formula). This Alvi range indicates a substantial
compositional variation, covering most of the compositional region
between muscovite (Alvi = 4.0) and phengite (Alvi = 3.0). Moreover, the
spectral reflectance data show that the compositional variations of white
mica are spatially related to the hydrothermal alteration zoning, such that
the white mica of more phengitic composition tends to form in the main
upflow fluid channels in the centre of the footwall alteration system. The
results suggest that the octahedral Al content of white mica, which is
spectrally definable by the wavelength of the Al-OH band using a field
instrument such as the PIMATM, can be used to map fossil hydrothermal
systems and potentially as vectors to mineralisation.
INTRODUCTION
White mica represents a group of di-octahedral sheet silicates,
commonly formed in low-grade regional metamorphic and
hydrothermally altered rocks. This group comprises several
solid-solution series, and the common species are those with
compositions mixed between three end-members, muscovite,
paragonite and celadonite (Eugster and Yoder, 1955; Velde,
1965). The compositional differences between these three
end-members result from cation substitutions in one or more of
the three structural sites: octahedral (six-fold coordinated),
tetrahedral (four-fold coordinated), and interlayer (12-fold
coordinated) ones. The Tschermak substitution, Siiv[Mg, Fe]vi =
AlivAlvi, represents coupled octahedral and tetrahedral cation
substitutions, and is one of the most common causes for
compositional variation of white mica (Guidotti and Sassi, 1976
and 1998). This substitution results in white mica
compositionally transitioning from muscovite to phengite. As
white mica varies in composition as a function of mainly
temperature, pressure and bulk chemical composition (eg Velde,
1967; Massonne and Schreyer, 1987), the composition of white
mica can be used to estimate the physical and chemical
conditions of hydrothermal systems.
Infrared spectroscopy is a powerful tool for characterising the
structure and composition of sheet silicates (Farmer and Russell,
1964). The field-based infrared spectroscopic technique using a
portable PIMATM spectrometer is particularly useful for deriving
detailed mineralogical data in the field, thanks to its rapidity
(30 to 50 seconds for each spectrum acquisition) without need
for sample preparation (suitable for hand specimen, drillcore and
powder) (eg Yang, Huntington and Scott, 1998; Yang et al,
2000). In this paper, spectral reflectance in the wavelength region
from 1300 nm to 2500 nm of drillcore samples of altered
volcanic rocks in the footwall of the Hellyer massive sulfide
deposit was measured with a PIMATM spectrometer. The work
aimed to spectrally determine the cation composition of white
mica in the hydrothermally altered volcanic rocks, and thus to
1. CSIRO Exploration and Mining, PO Box 136, North Ryde, NSW
1670, Australia
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
assess the potential of using the composition of white mica to
map often tectonically disturbed fossil hydrothermal systems and
as vectors to orebodies.
GEOLOGICAL SETTING
The following introduction on the geology and mineralisation at
Hellyer is summarised from McArthur and Dronseika (1990),
Corbett (1992), Gemmell and Large (1992) and Waters and
Wallace (1992).
The Cambrian Mount Read Volcanics consist of felsic,
intermediate and minor mafic volcanic rocks, which form a
200 km long and 20 km wide belt extending from Elliott Bay in
the south to Hellyer in the north. This volcanic belt is bound to
the east by Precambrian metamorphic rocks and to the west by
Middle to Late Cambrian sedimentary rocks.
In the Hellyer area, the Mount Read Volcanics consist of a
lower sequence of feldspar-porphyritic lava and volcaniclastic
rocks (the Central Volcanic Complex) and an upper
volcanic-sedimentary sequence of volcaniclastic rocks,
graywacke, siltstone and shale (the Mount Charter Group). A
suite of intermediate to mafic lavas and volcaniclastic rocks,
known as the Que-Hellyer Volcanics, lie toward the base of the
Mount Charter Group at the northern end of the Mount Read
Volcanics. The Que-Hellyer Volcanics crop out in a belt of 20 km
long and 5 km wide, with a thickness of about 1 km. These
volcanic and volcaniclastic rocks are divided into four units,
which, from the base upwards, are:
1. lower tuff and lavas;
2. lower andesite and basalt (AL);
3. mixed felsic to mafic lavas, epiclastic breccia, tuff, and
minor sedimentary rocks (HVS); and
4. upper basalt and andesite (BL).
The Hellyer massive sulfide deposit is hosted in the unit three.
The Zn-Pb massive sulfide orebody at Hellyer is
mound-shaped. A downward tapering, pipe-like hydrothermal
alteration system is well developed beneath the orebody in the
footwall rocks. Alteration minerals are zoned in the footwall
alteration pipe. A typical zoning pattern is characterised by a
siliceous core at the centre of the pipe, surrounded by an inner
chlorite-sericite zone, and a peripheral sericite-quartz zone. All
three zones are fully developed in the northern part of the
footwall alteration system, whereas in the central and southern
parts the inner chlorite-sericite zone is missing. The central and
inner alteration zones are characterised by intense quartz and
sulfide veins (stringer zone), whereas veining in the peripheral
sericite-quartz zone is weaker (stringer envelope zone). The
footwall alteration system is interpreted as representing the main
fluid pathways (ie feeders). There are three feeder systems, the
northern, central and southern, recognised underneath the
massive sulfide orebody.
Hydrothermal alteration is less intense in the hangingwall
rocks. Cr-bearing white mica (locally termed fuchsite) and
carbonate (mainly calcite) are the main alteration minerals
formed in the hangingwall basalt immediately overlying the
massive sulfide orebody. Minor amounts of chlorite, barite and
283
K YANG et al

4000E
12000N
GN TN
TB

DOL URS
TB

A URS

QRS
Hellyer
Portal
QRS
BL
HELLYER
DOL DEPOSIT
A
QRS (Projected)
10000N URS
BL BL

DA QRS

HA

BL DA
AL
AL 0 300 600m

A A
DOL
Hellyer
QRS
QRS URS BL
BL BL QRS
ACG AL RHY
URS

ACG AL

TERTIARY STZ BL
TB --- Basalt QRS
DEVONIAN (?)
DOL --- Dolerite
CAMBRIAN AL HVS
URS --- Upper Rhyolitic Sequence
QRS --- Que River Shale
BL --- Hellyer Basalt
HVS --- Hangingwall Volcaniclastic Sequence
STZ --- Stringer Zone
HA --- Highly Altered
DA --- Dacite
RHY --- Rhyolite 0 300 600m
AL --- Andersite & Epiclastic Debris Flows
ACG --- Animal Creek Greywacke

FIG 1 - Geology of the Hellyer region (Gemmell and Large, 1992).

pyrite are also present, and albite enrichment occurs at the
margin of the hangingwall alteration system. The alteration effect
in the hangingwall extends up to 200 m above the orebody and
terminates at the upper contact of the basalt with the overlying
black shale unit.
A major N-S trending wrench fault, the Jack Fault, cuts
acutely across the centre of the deposit. Post-mineralisation
sinistral movement along the fault resulted in the relative
displacement of 130 m north and 30 m up for the eastern block.
METHODS AND SAMPLES
SWIR spectral reflectance of samples (split drillcore and
powders) from selected diamond drillholes was measured with
an ISPL PIMA-II portable infrared spectrometer. The PIMA-II
spectrometer measures hemispherical reflectance from 1.3 m to
2.5 m in 601 channels, with a 2-nm interval between adjacent
channels and 6 - 10 nm spectral resolution. For each powder
sample (representing a two-metre length of core), one spectrum
was collected, whereas for drillcore two to four spectra,
depending on the degree of mineralogical heterogeneity as could
be visually recognised, were obtained at each depth.
Locations of the drillholes investigated in this study are shown
in Figure 2. The drillholes were selected to represent both
proximal and distal parts of the footwall alteration system. Of
these drillholes, HL840 and HL850 are deep holes obliquely
drilling through several alteration zones extending well below the
orebody. HL841B was drilled outside the Hellyer ore system to
test for northern extensions of the massive sulfide mineralisation.
This hole starts in the hangingwall basalt and ends deep in the
altered footwall andesite.
SWIR spectra were processed by using the software package
X-Spectra developed by CSIRO Exploration and Mining. A
baseline (hull) was fitted to each reflectance spectrum, and at
each wavelength the reflectance was divided by the
corresponding value on the baseline (hull quotient) to remove
the background effect (Clark, 1983) and the hull quotient spectra
resulted. The wavelength of the Al-OH band of white mica was
measured on the hull quotient spectra. The Al-OH band
wavelength increases as the Alvi decreases (Post and Noble,
1993; Duke, 1994; Scott and Yang, 1997), and therefore is a
useful parameter for spectrally quantifying the substitution of
octahedral Al (Alvi) by other cations such as Fe and Mg.

284 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 COMPOSITIONAL VARIATIONS OF WHITE MICA IN THE FOOTWALL HYDROTHERMAL ALTERATION SYSTEM

5500 mE 6000 mE RESULTS AND INTERPRETATION

HL841B
For altered footwall volcanic rocks proximal to the massive
sulfide orebody (in drillholes on cross-section 10 324 mN and in
the < 200 m depth part of 10 100 mN, Figure 3), the wavelength
11500 mN
of the Al-OH band of white mica ranges from 2202 to 2222 nm.
This 20 nm wavelength change indicates a significant change in
white mica composition from typical muscovite to phengite. The
change, however, is random, without obvious trends. This
suggests mixed muscovite and phengite in the altered footwall
rocks near the orebody.
Drillholes HL840, HL841B and HL850 are located on the
edge or outside of the orebody, and provide examples of
11000 mN compositional variation of white mica on the margin of the
HL850 mineralised hydrothermal system. The data (Figure 4) show the
HL840 Al-OH wavelength variations of white mica across the footwall
alteration system well below the orebody. In HL840, the less
altered andesite on the margin of the alteration pipe has shorter
dy
Orebo

Al-OH wavelength (muscovitic mica) and the wavelength

increases (more phengitic composition) toward the centre of the
system. Similarly, in HL841B white mica in the hangingwall
cted

basalt and in the andesite on the margin of the footwall alteration

Proje

10500 mN
pipe has shorter Al-OH wavelengths than that in the altered
andesites in the centre of the pipe. In HLB850, the Al-OH
HL579, 584, wavelength reaches the maximum value near the centre of the
591 & 599
stringer zone, and decreases toward the weakly altered andesite
and the margin of the stringer zone. All three drillholes show a
N
trend of increased Al-OH band wavelength in the mineralised
HL694, 697 parts of the footwall.
701 & 705
10000 mN The Al-OH wavelength data are compared with the alteration
index in Figure 5. The alteration index, 100 (K2O + MgO)/
FIG 2 - Locations of the studied drillholes. The solid lines are the (Na2O + K2O + CaO + MgO) (Ishikawa et al, 1976) calculated
projection of drillholes. from whole-rock chemistry, has already proved to be a useful

FIG 3 - Al-OH wavelengths for drillholes on cross-section 10100 mN. Refer to Figure 2 for the geology and locations the four holes.
Geology simplified from Aberfoyle Resources.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 285

K YANG et al
FIG 4 - Al-OH wavelengths and ore metal contents for drillholes HL840, HL841B and HL850. Pb and Zn abundance data and geology
(simplified) from Aberfoyle Resources.
parameter for assessing the degree of hydrothermal alteration at
Hellyer (Gemmell and Large, 1992). In HL840, the data show
that for samples with a high degree of alteration (ie the alteration
index > 80) the Al-OH band wavelength of white mica is > 2210
nm, whereas for those with a lower index value (30 to 70), the
Al-OH band wavelength is 2200 nm to 2210 nm. Similarly, in
HL841B the intensely altered andesitic rocks have longer Al-OH
band wavelengths than those weakly altered. Based on the two
drillholes, a division may be placed at 2210 nm to separate
intensely from weakly altered parts of the footwall feeder
system. Altered andesites with the Al-OH wavelength > 2210 nm
are characterised by an alteration index from 30 to 95, whereas
those with the Al-OH wavelength < 2210 nm are characterised
mainly by an alteration index from 25 to 60. This suggests that
phengitic mica (with longer Al-OH wavelength) tends to form in
the more intensely altered part, whereas the muscovitic mica
(with shorter Al-OH wavelength) is developed in the less altered
part of the system.
Hangingwall basalt was sampled in the upper parts of three
drillholes on the cross-section 10 100 mN (Figure 3). The
spectral data indicate that white mica in the hangingwall basalt is
more Al-rich, ie with shorter Al-OH band wavelength, than that
near the orebody does.
Also shown on the cross-section 10 100 mN (Figure 3), the
basaltic rocks from the hangingwall have shorter Al-OH
wavelengths than the footwall andesites. The basaltic and
andesitic rocks from the footwall, however, are undistinguishable
in Al-OH band wavelength. It appears that host lithology has
little influence on the composition of white mica when
hydrothermal alteration is intensified.
286 Adelaide, SA, 19 - 22 September 2004
DISCUSSION AND CONCLUSIONS
The observed variation in Al-OH band wavelength of white mica
in the footwall hydrothermal system at Hellyer are contiguous
from 2196 nm to 2222 nm. According to the data of Scott and
Yang (1997), this wavelength range corresponds to the Alvi
values approximately from 3.9 to 3.2 (based on four octahedral
cations per formula). This Alvi range covers most of the
compositional variation from muscovite (Alvi = 4.0) to phengite
(Alvi = 3.0). This is a substantial compositional range for K-rich
white micas.
Generally, white mica in the altered hangingwall basalts
overlying the sulfide orebody has shorter Al-OH wavelengths
than that in the footwall rocks. With the alteration in the
hangingwall basalt representing the hydrothermal activity
immediately following the deposition of massive sulfides in the
orebody, the derived Al-OH band data suggest that white mica of
more muscovitic composition was formed in the overlying
basalts in the waning stage of the hydrothermal process.
In the altered footwall, white micas with relatively longer
Al-OH band wavelengths tend to occur in or close to the central
part of the alteration pipe. This trend becomes more obvious with
increasing depth below the massive sulfide orebody. With the
Al-OH wavelength 2210 nm as the boundary, the white micas at
Hellyer can be classified as muscovitic ( Al-OH 2196 to 2210 nm)
and phengitic ( Al-OH > 2210 nm) and their spatial distributions
are summarised in Figure 6. Within the centre and the inner part
of the footwall feeder system, both phengitic and muscovitic
micas are present and they do not appear to differentiate from
each other with respect to the vertical distance from the orebody.
PACRIM 2004
 COMPOSITIONAL VARIATIONS OF WHITE MICA IN THE FOOTWALL HYDROTHERMAL ALTERATION SYSTEM

100 basalt

andesite
80
Alteration index

HL306 phengitic +

mu
mus
muscovitic

sco
60 Andesitic

cov

viti
Basaltic

itic

c
40
massive sulfides

20 siliceous core
2190 2200 2210 2220
stringer zone
Al-OH (nm)
stringer
100 envelope zone
HL840 muscovite-phengite
boundary
Alteration Index

80

FIG 6 - A summary of the distribution of white micas at Hellyer.

60
On the other hand, the composition of host-rocks is not likely
40 to have been a major controlling factor for the observed
compositional variations of white mica. As already shown, the
presence of phengitic mica preferentially in the footwall rather
20 than the hangingwall rocks suggests that the more Mg- and
2190 2200 2210 2220 Fe-rich basaltic rocks in the hangingwall did not favor the
Al-OH (nm) formation of Mg- or Fe-rich white mica during the alteration.
Furthermore, within the footwall, no distinction in the Al-OH
100 wavelength of white mica has been observed between the
HL841B
andesitic and the basaltic rocks.
Alteration Index

80 Enrichment of phengitic white mica marks the main upflow

60
40
20
2190 2200 2210
Al-OH (nm)
FIG 5 - Comparison of Al-OH wavelength and alteration index
100*(K2O + MgO)/(Na2O + K2O + CaO + MgO). The solid
diamonds and open squares are andesitic and basaltic volcanics,
respectively. Basaltic samples in HL306 and HL840 are from the
footwall, whereas those in HL841B are mostly from the
hangingwall. The alteration index is calculated from whole-rock
chemical data of Aberfoyle Resources (unpublished).
Laterally across the feeder system, however, the centre and inner
part are characterised by both micas, whereas the margins of the
feeder system are dominated by muscovitic mica.
Fluid chemistry and fluid/rock ratio may have played a
significant role in controlling the white mica composition at
Hellyer. Judged by the degree of veining, which decreases from
the centre to the margin of the footwall hydrothermal system, the
fluid/rock ratio must have decreased accordingly. Therefore, the
centre of the system would have been controlled mainly by the
fluids, and this control would have become weaker toward the
margin of the system. Based on the above equilibrium, elevated
Mg2+, Fe2+, and/or Si4+ relative to Al3+ in the fluids would favour
the formation of phengitic mica. However, there are no data yet
to show how the fluid composition changed spatially in the
alteration system, though Zaw et al (1996) found that the ore
fluids at Hellyer as a whole were more saline and K-rich, but
Mg-poor, than sea water.
fluid channels of the hydrothermal system. Away from the main
channels white mica becomes muscovitic. The recognised
association of phengitic white mica with mineralisation and/or
intense hydrothermal alteration indicates a direct link between
the composition of white mica and the hydrothermal processes.
The results therefore indicate that in exploration the Al-OH band
wavelength, and thus the Alvi, of white mica may be used to map
hydrothermal systems and as vectors to orebodies.
2220
ACKNOWLEDGEMENTS
This work was part of the AMIRA Project 435, Mineral Mapping
with Field Spectroscopy for Exploration. We were greatly
indebted to the following sponsors for their supports: Aberfoyle
Resources, Australian Geological Survey Organisation, Cyprus
Amax Australian Corp., Goldfields of South Africa, Great
Central Mines Ltd, Homestake Gold of Australia Ltd, MIM
Exploration Pty Ltd, Normandy Exploration Ltd, Department of
Primary Industries and Resources of South Australia, PNC
Exploration Ltd, Rio Tinto Exploration Pty Ltd, Stockdale
Prospecting Ltd, Sumitomo Metal Mining Oceania Pty Ltd, and
Western Mining Corp. In particular, thanks are due to Aberfoyle
Resources for allowing access to drillhole samples and
unpublished data, and to AMIRA International for facilitating the
work. This manuscript has been improved by the critical review
and constructive suggestions made by Graham Carr and Melissa
Quigley.
REFERENCES
Clark, R N, 1983. Spectral properties of mixtures of montmorillonite and
dark carbon grains: implications for remote sensing minerals
containing chemically and physically adsorbed water, Journal of
Geophysical Research, 88:10635-10644.
Corbett, K D, 1992. Stratigrafic-volcanic setting of massive sulfide
deposits in the Cambrian Mount Read Volcanics, Tasmania,
Economic Geology, 87:564-586.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 287

K YANG et al
Duke, E F, 1994. Near infrared spectra of muscovite, Tschermak
substitution, and metamorphic reaction progress: implications for
remote sensing, Geology, 22:621-624.
Eugster, H P and Yoder, H S, 1955. The joint muscovite paragonite,
Carnegie Institute of Washington Yearbook, 54:124-126.
Farmer, V C and Russell, J D, 1964. The infrared spectra of layer
silicates, Spectrochimica Acta, 20:1149-1173.
Gemmell, J B and Large, R R, 1992. Stringer system and alteration zones
underlying the Hellyer volcanic-hosted massive sulfide deposit,
Tasmania, Australia, Economic Geology, 87:620-649.
Guidotti, C V and Sassi, F P, 1976. Muscovite as a petrogenetic indicator
mineral in pelitic schists, Neues Jahrbuch fur Mineralogie,
Abhandlungen, 127:97-142.
Guidotti, C V and Sassi, F P, 1998. Petrogenetic significance of Na-K
white mica mineralogy: recent advances for metamorphic rocks,
European Journal of Mineralogy, 10:815-854.
Ishikawa, Y, Sawaguchi, T, Iwaya, S and Horiuchi, M, 1976. Delineation
of prospecting targets for Kuroko deposits based on models of
volcanism of underlying dacite and alteration halos: Mining Geology,
26:105-117 (in Japanese with English abstract).
McArthur, G J and Dronseika, E V, 1990. Que River and Hellyer
zinc-lead-silver deposits, in Geology and Mineral Deposits of
Australia and Papua New Guinea (Ed: F E Hughes), pp 1229-1239
(The Australasian Institute of Mining and Metallurgy: Melbourne).
Massonne, H J and Schreyer, W, 1987. Phengite geobarometry based on
the limiting assemblage with K-feldspar, phlogopite, and quartz,
Contributions to Mineralogy and Petrology, 96:212-224.
288 Adelaide, SA, 19 - 22 September 2004
Post, J L and Noble, P N, 1993, The near infrared combination band
frequencies of dioctahedral smectites, micas, and illites, Clays and
Clay Minerals, 41:639-644.
Scott, K M and Yang, K, 1997. Spectral reflectance studies of white
micas, CSIRO Exploration and Mining Report 439R, 34 p.
Velde, B, 1965. Phengitic micas: synthesis, stability and natural
occurrences, American Journal of Science, 263:886-913.
Velde, B, 1967. Si4+ content of natural phengites, Contributions to
Mineralogy and Petrology, 14:250-258.
Waters, J C and Wallace, D B, 1992. Volcanology and sedimentology of
the host succession to the Hellyer and Que River volcanic-hosted
massive sulfide deposits, northwestern Tasmania, Economic Geology,
87:650-666.
Yang, K, Huntington, J F and Scott, K M, 1998. Spectral characterization
of the hydrothermal alteration at Hishikari, Japan, in Proceedings
Ninth International Symposium on Water-Rock Interaction (Eds: G B
Arehart and J R Hulston), pp 587-590 (AA Balkema: Rotterdam).
Yang, K, Huntington, J F, Browne, P R L, and Ma, C, 2000. An infrared
spectral reflectance study of hydrothermal alteration minerals from
the Te Mihi sector of the Wairakei geothermal system, New Zealand,
Geothermics, 29:377-392.
Zaw, K, Gemmell, J B, Large, R R, Mernagh, T P and Ryan, C G, 1996.
Evolution and source of ore fluids in the stringer system, Hellyer
VHMS deposit, Tasmania, Australia: evidence from fluid inclusion
microthermometry and geochemistry, Ore Geology Reviews,
10:251-278.
PACRIM 2004
Exploration and Mining Geology Case Studies

Reducing Grade Uncertainty in High-Nugget Effect Gold Veins S C Dominy and 291
Application of Geological and Geochemical Proxies G F Johansen

Hydrothermal Alteration at the Bowdens Silver Deposit, NSW J Elliot, T Leach and 303
I Pringle

Results From Induced Polarisation Surveys Over the Beruang P J Elliott 307
Copper-Gold Deposit in Central Kalimantan, Indonesia

A Solution to Grade Estimation in a High Nugget Environment G F Johansen 317

The Bendigo Experience

A Review of the Kapit Orebody, Lihir Island Group, Papua New Guinea R P Kidd and J R Robinson 323

The Tooloom Gold Project, NSW Forgotten But Not Gone! R M D Meares, 333
Rediscovering a 147-Year-Old Gold Field G G Lowder, M J White,
B A Wake and M D Vickers

The Perseverance Gold Deposit The Next Step at Peak S Munro and R Berthelsen 339

Geology, Mining and Grade Control at the Batu Hijau Porphyry A Prihananto 345
Copper-Gold Deposits, Sumbawa, Indonesia

The Bong Mieu Gold Project, Phuoc Son Tam Ky Suture, Central N N Quynh, R H Murfitt, 347
Vietnam T Sirinawin and
W Shywolup

Discovery of the Toguraci Epithermal Au-Ag Deposits, Gosowong T H Richards and 359
Goldfield, Halmahera Island, East Indonesia M D Basuki Dwi Priyono

Discovery Performance of the Western World Gold Industry Over the R C Schodde 367
Period 1985 - 2003

The Hillgrove Gold-Antimony-Tungsten District, NSW, Australia C K Switzer, P M Ashley, 381

B Hooper and B Roach

The Discovery and Early History of the Mt Leyshon Gold Deposit, G S Teale and J E Lynch 385
North Queensland

A Search for Gold in Indonesian New Guinea M R Thirnbeck 391

Nagarze Gold Porphyry, Tibet PRC D S Tyrwhitt 401

Mining in Fiji B Wesson 403

A Three-Dimensional Structural Interpretation of the Olympic Dam Deposit H Widdup, T Fouet, 417
Implications for Mine Planning and Exploration J Hodgkison, T C McCuaig
and J Miller
Reducing Grade Uncertainty in High-Nugget Effect Gold Veins
Application of Geological and Geochemical Proxies
S C Dominy1 and G F Johansen2
ABSTRACT
The erratic and localised occurrence of economic gold grades is a
common feature of high-nugget effect gold veins. The nugget effect is a
consequence of small-scale geologically controlled enrichment, and the
inclusion or exclusion of sparse gold particles because of the sampling
process. Such deposits do not always contain visible or coarse gold, and
can be characterised by fine gold particles. Economic grade
accumulations are generally contained within discrete oreshoots, which
are surrounded by barren to low-grade material. Problems are often
encountered where small sample types (eg drill core or channel samples)
often vastly understate in situ grade, due to their general inability to
intersect rare gold particles. As a result, the definition of grade and its
distribution is often challenging, and leads to substantial resource
uncertainty. Proxies, or indicators for grade, can help the geologist gain a
better understanding of grade distribution and grade potential. These are
based upon geological (eg mineralogical, textural, etc) and/or
geochemical (eg specific elemental ratios, fluid inclusion compositions,
elemental correlations, etc) criteria. The paper discusses the use and
application of such proxies through three case studies.
INTRODUCTION
The challenges of sampling gold-bearing veins, especially those
containing coarse gold particles, with a high-nugget effect are
well-documented (Dominy, 2004; Dominy, Johansen and Annels,
2001; Dominy et al, 2000a, 2000b, 2001, 2003; Garnsey, 1993;
Johansen, 1997, 2004; Pelham, 1991, 1992; Platten and Dominy,
2003; Roberts, Dominy and Nugus, 2003; Royle, 1989;
Wooldridge, 1998). Coarse gold-bearing systems are the most
problematic, with special consideration required to maximise
field sample and assay charge sizes cost-effectively (Dominy et
al, 2000a). It is often difficult to maintain assay accuracy and
precision in such samples, due to low grades, the nuggetty
behaviour of gold grains (eg poor disintegration during
pulverising), and high contrast between the densities of gold and
gangue minerals (leading to segregation problems). In many
cases, the field sample size is simply too small to have any
chance of intersecting gold grains unless it weighs a few hundred
tonnes (eg New Bendigo: Johansen et al, 2003; Johansen, 2004).
In vein-hosted gold mineralisation, diamond drilling generally
provides an effective measure of geological continuity, but only
produces relatively small samples for assay (eg 2 kg of 1 m NQ
half-core). Such samples generally result in the understatement
of grade up to 75 per cent or more (eg New Bendigo; Johansen,
2004). Underground development and bulk sampling can most
reliably assess grade and its distribution. In general, it is unlikely
that anything above an Inferred Mineral Resource can be
estimated from surface drilling alone (Pelham, 1991, 1992;
Dominy et al, 2000a). Underground development and/or
close-spaced drilling is likely to be the most effective way of
defining Indicated and Measured Resources, and Reserves.
1. MAusIMM, Formerly: MCA Senior Lecturer in Mining Geology and
Resource Engineering, Economic Geology Research Unit, School of
Earth Sciences, James Cook University, Townsville Qld 4811.
Currently: Principal Mining Geologist, Snowden Mining Industry
Consultants Pty Ltd, PO Box 77, West Perth WA 6872. E-mail:
sdominy@snowden.com.au
2. MAusIMM, Exploration Manager, Bendigo Mining Ltd, PO Box
2113, Bendigo Vic 3554. E-mail: gjohansen@bendigomining.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Proxies, or indicators for grade, can help the geologist gain a
better understanding of grade distribution and grade potential.
These are based upon geological (eg mineralogical, textural, etc)
and/or geochemical (eg specific elemental ratios, fluid inclusion
compositions, elemental correlations, etc) criteria. This paper
provides an introduction to proxies for gold grade, which can be
used to ameliorate the above grade issues, through a general
discussion and three case studies.
BACKGROUND
What are proxies?
The definition of a proxy is a substitute or deputy to an original
feature or item. In the context of this contribution, proxy is being
used as an alternative to gold assay grade that is indicative of
grade potential.
For example, a drill intersection of 0.2 g/t Au in the middle of
an oreshoot is extremely misleading, as an intersection within a
few metres of the original could return 10s or even 100s g/t.
Clearly the 0.2 g/t value may represent the zone of influence of
that intersection, but is not representative of the overall shoot
grade. In rare cases where the nugget effect is extreme, and gold
particle distribution characterised by very coarse particles, drill
hole assays may rarely return economic grades, despite the
unknown average grade being economic (eg Johansen, 2004). In
such cases, the application of a proxy will permit definition of a
potentially mineralised economic zone during geological
modelling and estimation. Further interpretation and grade
manipulation may be required to estimate global grades for the
orebody (eg Johansen, 2004).
To define the limits of the gold distribution it is necessary to
identify geological and/or geochemical associations that broadly
correlate with gold grade, and that are less susceptible to an
erratic distribution. In essence we are trying to define indicators
for gold by recognising geological and/or geochemical domains
linked with gold grade domains. It would then be possible to
define a range of criteria or specific indices (eg for geochemical
proxies) that correlate with the grade potential of, for example:
1. barren,
2. low-grade,
3. medium-grade, and
4. high-grade.
Geochemical proxies
Rock geochemistry
Litho- and soil-geochemical studies have been used to detect the
presence of gold mineralisation during grass roots and near-mine
exploration for many years. Work has invariably used
pathfinder elements such as As, Sb, Bi, Hg, Ag, Te, Se, etc that
mirror gold mineralisation (Boyle, 1979).
Attempts to use pathfinder elements to discriminate between
individual auriferous and non-auriferous veins have had little
reported success. Work undertaken in the Dolgellau Gold-Belt
(Wales, UK), display a poor correlation between pathfinders
291
S C DOMINY and G F JOHANSEN
(including As, Sb, Bi, Pb, Zn, Cd, etc) and auriferous veins
(Dominy, 1996a, 1996b). The only correlation was found at the
Clogau St. Davids Mine where a Te (eg tellurbismuth and
tetradymite) halo exists around highly localised gold
occurrences. However, this halo was virtually as localised as the
gold itself so the visual and geochemical occurrence of Te was
often associated with visible gold (Dominy, 1996a).
Hydrothermal fluid geochemistry
One of the most exhaustive studies using fluid chemistry as a
proxy is presented in Shepherd et al (1991), who used thermal
decrepitation and mass spectrometry to analyse bulk fluid
inclusion compositions of quartz veins. They examined the role
of fluid inclusion volatiles (CH4, CO2 and N2) as an exploration
guide for gold in the Clogau St Davids and Gwynfynydd mines
in the Dolgellau Gold-Belt (UK). They conclude that, in general,
enhanced contents of both CH4 and N2 relative to CO2 are related
to wallrock reaction and gold deposition. They show that volatile
composition discriminates between gold-rich and gold-poor
veins at Clogau St Davids, which is an important conclusion in
the context of exploration and evaluation. Unfortunately the
discriminator did not perform well at Gwynfynydd.
Platten and Dominy (2003) describe textural features at the
Gwynfynydd mine, and show that the quartz, which carries the
fluid inclusions, has a complex history of deposition even when a
single vein is considered. The deposition of gold is associated
with the deposition of very small amounts of quartz and records
a distinctive event in the vein filling history. The relatively large,
framework-forming quartz crystals form the most attractive
target for fluid inclusions and form the major part of any bulk
sample. The preparation technique used by Shepherd et al (1991)
involves the recovery of gases from 0.5 - 1.0 mm diameter quartz
grains separated from a crushed sample of vein material. Such a
sample is likely to be a mixture of quartz from different stages in
vein filling and hence will provide a mixed sample of gases. This
suggests that any future attempts to develop and use this
discriminator at Gwynfynydd, and any other deposit, would need
to sample quartz from specific textural sites within the vein as
precisely as the laboratory technique permits and/or assess the
relative volumes of framework, interstitial and late quartz in the
sample.
Geological proxies I: textural and mineralogical
indicators
Textural and mineralogical indicators have been investigated as
potential discrimination tools. It is sometimes possible link
economic gold occurrences/oreshoots with a particular quartz
texture. In the Tarnagulla project (Victoria, Australia),
exploration was aided by the recognition of narrow, laminated
veins that were indicative of high-grade oreshoots (Cuffley et al,
1998; Parsons, 2000). In many cases textural studies have led to
detailed classification systems of ore deposit type, but not to the
prediction of gold grades (see Dowling and Morrison, 1989;
Vearncombe, 1993).
The occurrence of certain minerals can also be linked to the
occurrence of gold. In the Clogau St Davids Mine the appearance
of galena and/or telluride minerals is indicative of the presence
of gold (Dominy, 1996a). In the Gwynfynydd Mine the presence
of a light-coloured, iron-poor, sphalerite has been linked with
high-grade gold occurrences (Dominy, 1996b; Dominy, Phelps
and Camm, 1996; Platten and Dominy, 2003). Elsewhere, for
example, in the Charters Towers Goldfield (Queensland,
Australia) the simple content of sulfide (eg pyrite and/or galena)
generally indicates the presence of gold. Thus the occurrence of
a specific quartz texture and/or mineral(s) could provide some
degree of confidence during exploration that a gold-bearing
domain is either present or absent.
292 Adelaide, SA, 19 - 22 September 2004
Thermoluminesence (TL) has been used as a tool to
distinguish between types of quartz on the basis of TL spectra.
Work reported in Charlet and Quinif (1990) on samples from
Limousin (France) enabled them to identify quartz from different
vein sets. More importantly, they were able to discriminate
between barren and auriferous quartz.
Geological proxies II: geological features
The geological features that can assist in the definition of gold
mineralised zones are potentially; percentage quartz veining and
vein types, alteration type and intensity, and known structural
controls/orientations.
Quartz veining and vein type is closely related to texture, with
gold potential being related to the density of veining, and/or the
presence of a particular vein set. A specific alteration type or
intensity may indicate gold mineralisation, for example
calc-silicate alteration. With the development of a detailed
structural model, it may be possible to define which sets of veins
are gold-bearing even if actual sample grades are low.
Studies at the Marvel Loch gold mine (Western Australia) by
Nugus and Dominy (2003), demonstrated that quartz-rich
amphibolite, diopside dominant alteration, and the percentage of
veining, sulfides and quartz could be used to indicate grades
above 3.5 g/t Au. Whilst this was effectively developed as an
in-pit grade control methodology, it shows clearly how
geological and mineralogical features can be applied to define
likely gold grade.
NORTH AMERICAN CASE STUDY
Introduction
This mine, which cannot be identified due to a confidentiality
agreement, currently produces about 50 000 tonnes per annum at
a 17.5 g/t Au run-of-mine grade.
The deposit contains a series of gold veins that are hosted
within metasediments, and cut by metabasalts. The main host
vein, Type A, forms a continuous mineralised fault zone. The
more restricted Type B mineralisation is related to specific zones
of the Type A. In both structures gold is dominantly coarse
(generally 80 - 2500 m). In detail, the two styles of
gold-bearing mineralisation are:
Type A steeply dipping massive to laminated quartz veins
with widths from 0.60 to 2.5 m (Figure 1). The structures contain
less than ten per cent sulfides (eg chalcopyrite, pyrite, galena and
sphalerite), with fine and coarse gold hosted in quartz. Gold
occurrences are structurally controlled, and located within
steeply plunging oreshoots that yield a mean grade of 12 g/t Au.
Type B quartz stockwork zones with widths up to 10 m.
These zones are located principally in the footwall zone of the
Type A mineralisation (within the oreshoot zones) and are
FIG 1 - Gold-quartz reef (Type A) from the North American case
study. This view is of the drive backs (looking up), and shows a
0.7 m wide well laminated quartz vein grading at about 26.9 g/t Au
(Source: S C Dominy).
PACRIM 2004
 REDUCING GRADE UNCERTAINTY IN HIGH-NUGGET EFFECT GOLD VEINS
related to the multiple branching of the vein fracture within
massive basic sills. This style contains up to about ten per cent
sulfides (eg pyrite, galena and sphalerite) and localised
high-grade occurrences of coarse gold grading 1000s g/t Au.
Gold is often associated with sphalerite and rarely with tellurides
(eg tellurobismuthite). Gold distribution is highly erratic with
recovered grades ranging from 10 to 40 g/t Au.
Outside of the well-mineralised zones (eg A and B) the main
vein structure is typified by massive book and ribbon structure
quartz with occasional concentrations of pyrite. Ore minerals
such as chalcopyrite, galena, sphalerite, etc are notably absent as
is gold.
Geochemical analysis of samples
The aim of the study was to identify which elements correlated
with gold, and reflect the grade of a particular quartz vein rather
than depending upon gold assay alone.
Sixty-five vein samples were collected underground
representing the Types A and B mineralisation. The samples
were from a single oreshoot zone. Twenty further samples of
vein material were collected from a barren zone away from the
oreshoot as a control. All samples were collected as channel
samples cut with a diamond saw to ensure the best quality. Each
sample, depending upon vein width, weighed between 5 - 10 kg.
Samples were prepared and assayed at a commercial laboratory.
All samples were pulverised in their entirety to 75 m for
analysis.
Samples were analysed by an Inductively Coupled Plasma
Mass Spectrometer (ICP-MS) for Ag, Sb, Rb, Li, Bi, Mo, Te and
Se. The elements Al, Fe, Mg, Mn, Ca, K, Na, Ba, Cu, Pb, Zn, Sr,
Cr and As were analysed using an Inductively Coupled Plasma
Atomic Emission Spectrometer (ICP-AES). Gold was
determined using an eight-hour accelerated cyanide extraction
(ACE) on two 2 kg subsamples with an ICPAES determination
(detection limit 0.01 g/t Au), and fire assay of residues.
Sample dataset
Inspection of the data showed that a wide range in elemental
concentration was common, eg Zn ranges from 0.5 ppm to 13.2
per cent. The values of some of the elements were anomalous
and after further investigation including re-analysis they were
identified as being valid. These were considered to have sound
geological explanation, reflecting the erratic elemental
distribution within the veins and were thus not removed from the
database.
Detection limits for the elements were standard for the
methods of analysis employed. Where the analysed value for a
sample fell below the detection limit, the detection limit value
was added to the database prior to statistical analysis.
Elemental correlation
The study required that major and trace element values were
correlated with the gold grades. This was undertaken using the
analytical methods and 65 samples reported previously. The
20 barren non-oreshoot control samples were also analysed.
Most of the elemental populations display a strongly positively
skewed distribution, thus the raw data was log transformed,
yielding a near lognormal population more amenable to
statistical analysis.
To facilitate the identification of elemental correlations, linear
regression correlation coefficients (cc) for each element against
each other element were defined. The method yields a score of
0 - 100 per cent where a perfect correlation is at either 100 per
cent. Only scores of 60 per cent or above were used in this
study to indicate a reasonable or better correlation.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Correlations were undertaken using four datasets:
1. Type A mineralisation (50 samples),
2. Type B mineralisation (35 samples),
3. Type A and B mineralisation (85 samples), and
4. barren non-oreshoot mineralisation (20 control samples).
Interpretation of the mineralised-vein (eg A and B) correlation
data reveals notable differences in elemental correlations
between different datasets, especially between gold and other
elements. The Type A mineralisation exhibits a correlation
between gold with Cu and Zn (CC of >65 per cent). In the Type B
mineralisation, gold displays correlation (CC 65 per cent) with
Mn, Ca, K, Ba, Sr, Te, As, Ag, Rb, Rb, Cu and Zn. The dataset
containing both Type A and B mineralisation showed a
correlation between gold with Ag, Zn, Cu, Mg and Rb.
The large number of correlations with the Type B
mineralisation is likely to be in response to the small sample
population influencing the results. The CC values identified in the
combined dataset are thus of the most use due to the larger
number involved. The elemental correlations identified in the
total dataset, where cc >65 per cent, can be accounted for in
terms of the following mineralogical relations:
Gold shows very strong correlation with Ag (87 per cent)
suggesting the formation of a high silver-gold alloy. This is
confirmed by mineralogical studies that yield a general gold
fineness of between 650 and 800. Correlation with Zn (89 per
cent) is explained by the fact that the presence of gold is often
associated with sphalerite. Again to some extent the correlation
between gold and Cu (61 per cent) is explained by the presence
of chalcopyrite within the mineralised areas. The strong
correlation between gold and Mg and Rb is not fully explained,
though may relate to the observed presence of chlorite in the vein
associated with some mineralised zones. The correlation between
gold and typical indicator elements such as As, Sb, Bi and Te are
low (all less than 25 per cent).
Silver displays a strong correlation with Cu, Pb, Zn, Te, Se and
Bi suggesting a base metal sulfide association. This is in
agreement with paragenetic studies showing an association
between galena, chalcopyrite and sphalerite. A strong
relationship between Ag and Pb exists through the presence of
Ag-rich galena in the mineralisation (confirmed by microprobe
analyses as being up to 4.2 wt per cent Ag). The Te correlation
with silver is likely to be related with the occurrence of telluride
minerals within the Type B mineralisation. Surprisingly the
correlation between Te and gold is not high (19 per cent).
Potassium correlates with Mg, Na, Al and Ba indicating
feldspar and mica formation within the vein. This is supported by
the presence of K-feldspar within the vein.
Interpretation of the barren non-oreshoot mineralisation data
reveals no correlation with gold, as all gold assays were below
0.1 g/t Au. Nine out of the 20 samples were below the 0.01 g/t Au
detection limit for the analytical method used. The key
correlations were not related to gold-associated elements as
displayed in the well-mineralised samples above. Elemental values
for Ag, Zn and Cu were low compared to the well-mineralised
samples (eg all less than 100 ppm concentrations).
Classification and ranges of production-based
gold grades
This study aims to identify elemental correlations that reflect the
grade range of vein material, whilst not using actual assayed gold
grade. A classification of grade ranges based on production data
was devised (Table 1).
During development and mining, the vein was channel
sampled (each approximately 5 - 10 kg). A 1 kg subsample was
produced for an ACE assay, and residues were fire assayed.
293
S C DOMINY and G F JOHANSEN

Replacing gold assays with a less erratic variable

A number of elemental ratios were tested to determine an
effective discrimination index. The best index was based on the
ratio of gold-related and gold-unrelated chalcophile and
lithophile elements. Chalcophile elements are those that have a
strong affinity for sulfur (eg Cu, Zn, etc) and lithophile elements
are those that prefer the silicate phase such as Li, K, etc. The
gold proxy index (GPI) is given as (Dominy, 2001):
[Ore - related lithophile element] [Ore - related chalcophile element]
[Unrelated lithophile element] [Unrelated chalcophile element]
Gold-related (>65 per cent correlation with respect to Au) and
unrelated (<60 per cent correlation with respect to Au) elements
can be determined by inspection of correlation data. The major
and trace element data was correlated with gold grades determined
from the same sample by ACE assay. Sixty GPI were defined with
the elements that showed the strongest correlation and
non-correlation with gold. All values that failed to show a
systematic increase with gold grade or an increase in the GPI
value between grade categories of about 100 per cent or less were
rejected.
The following GPI satisfy the criteria stipulated above for all of
the three data sets; [Mg Zn]/[Li Sb]. Other GPI for all samples
include [Mg Zn]/[Na Sb] and [Rb Zn]/[Na Mo] (Table 2).
GPI were also defined that were applicable to only the Type A or
Type B mineralisation, however those that are applicable across
the board are considered more useful (Table 3).
A graphical comparison of the GPI values that increase
substantially with the mineralisation categories is shown in
Figures 2 and 3.
Discussion
This study has produced a series of elemental gold-proxy ratios
that allow the definition of a discrimination index. The GPI
values have been calibrated against four types of gold grade
range; barren, low-, medium- and high-grade. The GPI is actively
being applied to allow better interpretation of surface and
underground drilling and underground development-sampling
programs. The method allows the geologist to gain a feel for gold
grade potential, which would be difficult based simply on core
assays that understate gold grade.
TABLE 3
GPI specific values for Type A and B mineralisation only. Values
calculated from the true mean of the log-transformed data for
each element for each grade type.
Mineralisation type Type A Type B
Grade type/GPI [K x Mg]/[Na x Sb] [Ba x Zn]/[Sr x Na]
Barren 662 695
Low-grade 23 472 3973
Medium-grade 37 664 19 673
High-grade 88 265 37 436
Such a method is never likely to be infallible. This case study
has been trialled now for a few years, and appears to be working
well. Data shown in Table 4 shows that out of 15 diamond drill
holes that are available for direct comparison with bulk sample
results, eleven (73 per cent) displayed a good correlation. The
original assay of the intersections is also shown in Table 4, and
shows how in this case such data consistently understate bulk
sample results.
The technique provides strong support during evaluation.
Preliminary grade modelling uses the average grade as defined in
Table 1, rather than the sample indicated grade. This
methodology has been found as very effective in helping to
define the economic mineralised envelope and gain a feel for
viability. Further resource estimation is based on closer-spaced
drilling and underground development with limited bulk
sampling.
It is highly unlikely that a set of elements defined for one mine
or vein will be applicable to all. Indeed trials at some mines have
been completely unsuccessful in defining a GPI (Williams,
2002). This method can only be developed once substantial
sampling has taken place, linked with underground development
and mining. It may be possible in some cases to produce a
calibration from diamond drilling, though geological and grade
continuity must be clearly understood. The sampling and gold
assaying technique used to calibrate the GPI must be applicable
to the nature of the metal present in the deposit. In most cases,
particularly when coarse gold is present, the large assay-charge

TABLE 1
Gold grade classification, ranges and average.

Grade classification Description Grade range (g/t Au) Average grade (g/t Au)
Barren Sub-economic material, barren zones are encountered within oreshoots <2 0.5
and associated with either Type A or B mineralisation.
Low Low-grade ore within oreshoots and associated with either Type A or B 2-8 5.0
mineralisation.
Medium Medium-grade ore within oreshoots that can be encountered in both Type 8 - 20 14.0
A or B mineralisation.
High High-grade ore usually encountered in the rich portions of the Type B >20 25
mineralisation, occasionally found in Type A mineralisation.

TABLE 2
GPI values for differing element ratios representing the entire sample dataset. Values calculated from the true mean of the log-transformed
data for each element for each grade type.
Grade type/GPI [Mg Zn]/[Li Sb] [Mg Zn]/[Na Sb] [Rb Zn]/[Na Mo]
Barren 112 259 373
Low-grade 684 10 290 1859
Medium-grade 2542 43 339 7404
High-grade 4781 78 857 13 902

294 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 REDUCING GRADE UNCERTAINTY IN HIGH-NUGGET EFFECT GOLD VEINS

FIG 2 - Graph showing the increase in GPI values (log10 value) of three different ratios for the entire dataset (Type A and B mineralisation).
Grade ranges: 1 = barren; 2 = low-grade; 3 = medium-grade and 4 = high-grade.

FIG 3 - Graph showing the increase in GPI values (log10 values) of ratios for Type A and Type B mineralisation.
Grade ranges: 1 = barren; 2 = low-grade; 3 = medium-grade and 4 = high-grade.

ACE technique is recommended providing residues are also
assayed, at least during initial testing. The strength of the GPI
technique is that it requires standard laboratory methodology and
equipment, and the costs involved are not prohibitive.
NEW BENDIGO CASE STUDY
Introduction
The Bendigo Goldfield is located in the state of Victoria,
Australia, and is one of a number of fields that make up the
Central Victorian Goldfield. Since 1993 Bendigo Mining Ltd has
been carrying out an extensive re-evaluation of the field, and in
July 2004 announced its successful fund raising campaign to
develop an underground mine.
The sediment-hosted auriferous quartz reefs are structurally
complex and closely associated with anticlinal axes and reverse
faults. The gold within the quartz reefs is characterised by both its
very coarse nature (100 up to >10 000 m) and erratic distribution.
As a result, the Bendigo Goldfield is classed as an extreme-nugget
effect coarse-gold system (Dominy et al, 2001, 2003). This
characteristic ensures that the Bendigo mineralisation is very
challenging to sample, estimate grade and report resources and
reserves (Johansen et al, 2003; Dominy, Stephenson and Annels,
2003; Johansen, 2004; Dominy and Johansen, 2004). In this case,
diamond drilling is a good measure of structure and geological
continuity, but a poor measure of grade. At the current time,
drilling, combined with on-reef development, bulk sampling and
detailed geological studies are required for resource evaluation
(Johansen et al, 2003; Johansen, 2004).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 295

S C DOMINY and G F JOHANSEN

TABLE 4
Comparisons between underground core samples and GPI grade classes, with bulk sample values undertaken during subsequent
development. The drill intersection grades are diluted to allow direct comparison with the bulk sample data. Drill core recoveries
>90 per cent.
Hole ID Hole intersection GPI grade class from Bulk sample grade Agreement between bulk
gold assay (g/t Au) core analysis (g/t Au) sample and GPI grade class
103 0.2 Low-grade 3.8 Good
107A 0.08 High-grade 21.9 Good
108 0.4 Low-grade 6.9 Good
113 7.7 Medium-grade 18.9 Good
114C 1.3 Low-grade 7.4 Good
114D BDL Low-grade 8.3 Poor
116 0.04 Low-grade 0.1 Poor
099S BDL Low-grade 12.7 Poor
119 3.4 Medium-grade 9.6 Good
123 0.6 Medium-grade 3.2 Poor
124A 0.9 Medium-grade 12.4 Good
124B 1.5 Medium-grade 15.8 Good
125 3.7 High-Grade 64.3 Good
127 1.7 High-grade 26.7 Good
131 9.2 High-grade 55.4 Good

BDL = below detection limit ( 0.01 g/t Au).

The challenge
The challenge at Bendigo is to define zones of economic
mineralisation during the definition of Inferred Resources from
100 - 120 15 m spaced drill fans. This interpretation is based
upon core assay grades that:
1. display an extreme nugget effect (>90 per cent relative
nugget effect);
2. are highly skewed; and
3. generally under-represent mineable grades by 50 - 75 per
cent (Dominy, Johansen and Annels, 2001; Johansen et al,
2003; Johansen, 2004).
When reviewing drill sections, it is useful to assign a grade
estimate to each intersection rather than use the actual grade.
Bendigo Mining currently identifies mineralised envelopes
representing the boundary of broad zones within which, smaller
zones of potentially economic mineralisation are expected to
occur. The mineralised envelope probably represents an
approximation of the 1.5 g/t Au grade shell, ie everything inside
the mineralised envelope will be greater than 1.5 g/t Au, and will FIG 4a - New Bendigo Mine, Victoria, Australia: laminated
contain the high-grade zones that the historic miners extracted. gold-quartz vein in the well-developed east leg of the shale-hosted
At present, mineralised envelopes are selected on the basis of Upper Sheepshead S3 Reef. The laminated vein is about 0.5 m
key geological, textural/mineralogical and geochemical features wide at this point, and contains visible gold along its footwall
(eg proxies): contact. Beneath the main structure is a complex network of
known structural controls to quartz and gold; footwall veins, which can carry grade. In section, the definition
of a mineralised envelop for this type of reef is not difficult
quartz distribution;
(Source: Bendigo Mining Ltd).
visible gold distribution;
gold assays from drill core; several drill holes on a single section before it is possible to
alteration intensity; define the envelope limits (Figures 4b and 5).
sulfide mineralogy, content and distribution; and The grade assigned to a particular drill intersection represents
the expected bulk sample grade for a sample collected along the
micro- and macro-bulk sample results where available drill intercept over the full width of the mineralised envelope.
(Johansen et al, 2003). Though assigned grades will at best be a rough approximation of
For narrow and geologically distinct reefs the mineralised actual grades, and occasionally will be wrong, they are a much
envelope is not difficult to define (Figure 4a). The broad better approximation to actual grade than the drill assays. Six
stockwork zones often have diffuse boundaries, and require grade classes are assigned (Table 5).

296 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 REDUCING GRADE UNCERTAINTY IN HIGH-NUGGET EFFECT GOLD VEINS
FIG 4b - New Bendigo Mine, Victoria, Australia: stockwork/spur vein
zone of the Christines No 2 Reef, Lower Sheepshead (S3) Ribbon.
The controlling structures have generated a complex mineralised
zone composed of closely spaced sub-vertical quartz veins at the
base of the reef (this photo), more massive to stockwork-like quartz
in the central core, and a lower-density quartz vein stockwork in the
upper portion of the reef (see Figure 5a). Higher grades tend to be
associated with the high quartz core and the footwall and
hangingwall faults. The hangingwall and footwall are sharply defined
and easily recognised, whereas the upper and lower boundaries of
the reef are poorly defined reflecting reducing quartz content and
gold grades moving away from the reef core (Source: Bendigo
Mining Ltd).
TABLE 5
Grade classes.
Designation Grade (g/t Au) Colour
Barren <1 Grey
Low-grade 2 Blue
Sub-economic grade 5 Green
Economic grade 8 Yellow
Target mine grade 12 Orange
High-grade >25 Red
As the parameters for assigning grade to a particular drill
intersection must cater for both well-defined narrow structures
and diffuse stockworks, there are no hard and fast parameters for
assigning a particular grade. Each reef is treated to some degree
as a unique entity. The key parameters to check when assigning
grade are:
1. Gold assays average gold assay, grade of lower 50 per
cent of assays, and distribution of higher numbers (if any).
2. Visible gold number of drill metres/visible gold
occurrence, size of visible gold particles, and presence of
visible gold distribution in surrounding drill holes.
3. Arsenopyrite and arsenic values presence and amount of
arsenopyrite. Is it in host rock and/or quartz veins? Average
arsenic value for that intersection.
4. Sulfide species the number of sulfide species present
(generally sphalerite, galena, arsenopyrite, pyrrhotite and
pyrite). How many hosted by quartz veins? Percentage of
each sulfide species present.
5. Host quartz is it bucky-quartz or does it look well
mineralised? Does the quartz contain sulfide(s), rock
fragments, stylolites, carbonate and/or multiple generations
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
of quartz? What percentage of quartz is present and is this
style of veining associated with economic mineralisation
elsewhere?
6. Structural setting is the reef being drilled situated in good
structural setting? What are the other drill holes into the
reef like?
7. Alteration check intensity of alteration. Is this style of
alteration typically associated with significant
mineralisation? How big is the alteration system?
The proxy criteria developed for grade ranges at Bendigo is
given in Table 6, and Table 7 lists examples of actual reefs. The
assigned high-grade is ranked as >25 g/t Au. This allows some
flexibility to assign a higher grade (eg 60 g/t Au) to specific
intersections (especially in narrow reefs) if the drilling
information suggests this is likely. Based on the understanding of
grade distribution within the historically mined reefs, whole reef
intersections are unlikely to represent grades in excess of 30 g/t
Au. Within a single reef there will be numerous drill
intersections, each of which can be assigned a grade. If assigned
grades are done effectively, then the average of the assigned
grades from all the drill intersections are likely to approximate
the actual reef grade.
A spreadsheet-based scorecard was designed for assigning
grades to drill intersections (Figure 6). This attempts to cater for
all styles of mineralisation, and is likely to require further
refinement as additional sampling information becomes
available. Each intersection is ranked, and the total ranking used
to help to define the grade range type.
Concluding comments
This type of multi-proxy method has proved useful for defining
mineralised envelopes during preliminary resource modelling,
and to gain an appreciation of internal grade distribution. This
compliments the approach used in estimation, which involves an
extreme top-cut and grade factoring followed by block modelling
to define the final resource shape, prior to grade estimation
(Johansen, 2004).
RUSSIAN CASE STUDY
Introduction
The privately owned mine, which cannot be identified due to a
confidentiality agreement, produces about 150 000 tonnes per
annum at a grade of 13.5 g/t Au.
The deposit is a lode-gold system comprising a series of
subparallel, moderate to steeply dipping 1.2 to 5.5 m wide
(average 2.5 m) gold-quartz-sulfide reefs (Figure 7), which yield
run-of-mine grades of between 12 and 15 g/t Au. The host fault
structures are continuous over 2 km, with quartz-mineralised
segments continuous for up to 500 m along strike.
Mineable grades are localised within structurally controlled
high-grade oreshoots that are continuous along strike for about
100 m, and over 200 m down-dip. Gold grade within the
oreshoots is erratic, with closely spaced samples potentially
returning grades that are different by 10s g/t Au. These shoot
zones are related to bends and/or splays within the hosting
structure. Mineralisation is dominated by coarse visible gold
(>50 per cent +105 m), with some 65 per cent being gravity
recoverable.
The challenge
Drill hole intersections within the mineralised vein are often
low-grade (<1.5 g/t Au), with relatively few intersections above
8 g/t Au, though rarely reaching 100 g/t Au. The projected
down-dip geology of the vein below economically viable workings
297
S C DOMINY and G F JOHANSEN

A
Score Card for Assigning Grade to Drill Intersections

Drill Hole No:

. Score:
..
Drill Hole Interval:
. Ranking:
..
Reef Name:
.
Assigned Grade:
.

PARAMETER Assigned PARAMETER Assigned

Score Score
Gold Assays Arsenic
Highest assay <1g/t Au 0 Arsenopyrite Present 1
<2g/t Au 1 In quartz 2
<5g/t Au 2 Absent 0
<25g/t Au 3 Average As value <100ppm 0
<50g/t Au 4 <500ppm 1
>50g/t Au 5 <1000ppm 2
>1000ppm 3
Average assay <0.1g/t Au 0
<0.4g/t Au 1 Sulphides
<1g/t Au 2 Species present 2 0
<4g/t Au 3 3 1
<10g/t Au 4 >3 2
>10g/t Au 5 Species present in 2 1
quartz. 3 2
Bottom 50% of assays <0.05g/t Au -2 >3 3
<0.1g/t Au 0
<0.2g/t Au 2 Host Quartz
<0.3g/t Au 3 Percentage <10% 0
<0.5g/t Au 4 <20% 1
>0.5g/t Au 5 <30% 2
>30% 3
Visible Gold Degree mineralised Low 0
Vis Au present Yes 2 Med 2
No 0 High 4
Coarse vis Au present Yes 2
No 0
Multiple pieces present Yes 2 TOTAL SCORE:
No 0
Vis Au in adjacent holes Yes 2
No 0
Ranking Grade Score
Barren <1g/t Au <8
Low Grade 2g/t Au 8 to 15
Sub-economic 5g/t Au 15 to 20
Economic 8g/t Au 20 to 25
Target 12g/t Au 25 to 30
High Grade >25g/t Au >30

Score and Ranking completed by:

B FIG 6 - Scorecard used for assigning grade to drill intersections

(Source: Bendigo Mining Ltd).

FIG 5 - Idealised sections (looking north) through the (A) Christines

No 2 (10 10 m grid shown) and (B) Garrards Reefs, on the Lower
Sheepshead (S3) Ribbon. These show the internally complex
nature of the wider more stockwork-like orebodies at Bendigo, and
their amenability to modelling using the proxy criteria to help set the
mineralised envelope (Source: Bendigo Mining Ltd).
FIG 7 - Main Reef, undisclosed gold operation in the CIS.
Well-laminated quartz veins within the oreshoot zone
characterises this moderately dipping structure. Local
developments of stylolites indicate the presence of
coarse gold particles (Source: S C Dominy).

298 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

PACRIM 2004

TABLE 6
New Bendigo grade classes with characteristic proxy types.

Grade class Proxy type

Gold assays Visible gold Arsenopyrite and Sulfides Quartz Structure Alteration
arsenic assay
Barren Average <0.1 g/t, and/or Extremely rare, when Asp generally absent and Usually pyrite and Looks bucky and barren. Not in recognised setting, Weak and restricted.
(<1 g/t Au) 50 per cent below present as isolated specks confined to hostrocks. As confined to hostrock. eg away from recognised
detection. Only a few >0.3 and fine-grained. assays <100 ppm. ribbon position.
g/t.
Low-grade Average <0.4 g/t, and/or Extremely rare, when Asp generally rare and Usually pyrite and Often bucky and looks In recognised setting. Weak to moderate and/or
(2 g/t Au) most are <0.1 g/t. Assays present as isolated specks confined to hostrocks. As generally confined to barren with little sulfides, of restricted dimensions.
>0.5 g/t are rare. and fine-grained. assays <500 ppm. hostrock. stylolites or rock The type is associated
fragments. with known
mineralisation.
Sub-economic Average <1.0 g/t, with Relatively rare, when Asp relatively common in Pyrite and arsenopyrite May be bucky, more often In recognised setting for Moderate to strong. The
(5 g/t Au) majority <0.2 g/t. Assays present as isolated specks hostrocks and veins. As common. well mineralised with significant mineralisation. type is associated with
>0.5 g/t relatively and fine-grained. assays >1000 ppm. sulfides, stylolites and Nearby drill holes contain known mineralisation.
Adelaide, SA, 19 - 22 September 2004

common though >5 g/t rock fragments. Quartz significant mineralisation.

rare. content >15 per cent.
Economic Average <4 g/t (2 g/t for Relatively common, being Asp relatively common, Pyrite and arsenopyrite Well mineralised with In recognised setting for Moderate to strong. The

REDUCING GRADE UNCERTAINTY IN HIGH-NUGGET EFFECT GOLD VEINS

(8 g/t Au) intersections >10 m) with both fine- and with As assays >1000 common. Most sulfides in sulfides, stylolites and significant mineralisation. type is associated with
majority of assays <0.3 coarse-grained. Several ppm. In both hostrocks hostrock, though some rock fragments. Quartz Nearby drill holes contain known mineralisation.
g/t. Some >1 g/t common, particles may be present in and veins. will be in the veining. content >20 per cent. significant mineralisation.
with very rare >25 g/t a single metre of core. Quartz may look grey.
assays.
Target Average <10 g/t (5 g/t for Relatively common, being Asp relatively common, Pyrite and arsenopyrite Well mineralised with In recognised setting for Moderate to strong. The
(12 g/t Au) intersections >10 m) with both fine- and with As assays >1000 common. Most sulfides in sulfides, stylolites and significant mineralisation. type is associated with
majority of assays <0.5 coarse-grained. Several ppm. In both hostrocks hostrock, though some rock fragments. Quartz Nearby drill holes contain known mineralisation.
g/t. Some >1 g/t common, particles may be present in and veins. will be in the veining. content >20 per cent. significant mineralisation.
with rare >25 g/t assays. a single metre of core. Quartz may look grey.
High-grade Average >10 g/t (5 g/t for Relatively common, being Asp relatively common, Pyrite and arsenopyrite Well mineralised with In recognised setting for Moderate to strong. The
(>25 g/t Au) intersections >10 m) with both fine- and with As assays >1000 common. Most sulfides in sulfides, stylolites and significant mineralisation. type is associated with
majority of assays >0.5 coarse-grained. Multiple ppm. In both hostrocks hostrock, though some rock fragments. Quartz Nearby drill holes contain known mineralisation.
g/t. Some >1 g/t common, particles may be present in and veins. will be in the veining. content >20 per cent. significant mineralisation.
with >25 g/t assays a single metre of core. Quartz will look grey.
present.
299
S C DOMINY and G F JOHANSEN

TABLE 7
Examples of grade type at Bendigo.

Grade class Example

Barren Stockwork systems encountered in the Swan Decline, parts of the St Anthony Reef, and the majority of quartz located
away from recognised ribbons and anticlines.
Low-grade Majority of the St Anthony Reef, Bryant Reef, low-grade portions of both D3 East Leg and Shywolup Reef, and the
majority of Shywolup Eastern Spurs.
Sub-economic Ends of the Upper S3 Reef, some of the Mid-S3 reefs, and the margins of the central shoot in D3 East Leg/Shywolup
Reef.
Economic Upper S3 Reef, the central shoot in East Leg/Shywolup Reef, the Upper S3 Reef, and the central shoot in East
Leg/Shywolup Reef.
Target Higher-grade areas of the Upper S3 Reef, central shoot in the East Leg/Shywolup Reef, and the core of Christine No 2
Reef.
High-grade Small high-grade shoot in St Anthony Reef, southern shoot in East Leg/Shywolup Reef, and the high-grade area in east leg
of the Upper S3 Reef.

TABLE 8 Geochemistry
Summary of proxies. Linked with the mineralogy defined above, the oreshoots were
relatively enriched in As and Bi. As was identified in quantities up
Proxy Comment Grade indication
to about 30 000 ppm, with Bi up to about 5000 ppm. The
Arsenopyrite Coarse crystals to 3 mm, Indicates oreshoot zone correlation coefficients between As and Bi with gold were 89 per
(As content) sometimes enclosing or cent and 77 per cent respectively. Outside of the oreshoots, the
intimately associated with concentrations of these elements were extremely low (<100 ppm
coarse and/or fine gold
each).
Bismuthinite Generally microscopic, Indicates oreshoot zone
(Bi content) sometimes enclosed by gold
Textural
Laminations Unlaminated massive veins Indicates oreshoot zone
always barren and on the A clear link between gold grade and the degree of vein
periphery of oreshoot zones laminations and presence of stylolites was observed (Figure 7).
Stylolites Pressure solution features that Density of stylolites In fact, the presence of seven to ten or more stylolites within a
have a very clear relationship appear to be metre of core almost certainly guaranteed local grades above
to gold. They often contain proportional to the 30 g/t Au. The link between stylolites and gold grade has been
coarse visible gold particles presence of gold grade recorded in other high-grade gold quartz vein systems (Dominy,
Platten and Raine, 2003). Particles of gold up to 3 mm in size are
typically located within the stylolites.
generally gives an indication of the continuation of oreshoots. On
a number of occasions, the projected vein had continued Table 9 shows the comparison between eight original drill
physically, but grades were low and either sub-economic or intersection grades (over entire intersection) with proxies and
marginal. A methodology was required that would enable local grade. The local grade is based on bulk samples taken
geologists to have confidence in the continuation of these within 10 m of the original drill intersection. The results show
high-grade oreshoots, and make better informed recommendations the differences that can be seen between the drill grades and the
on underground development and thus capital investment. local grade (eg intersection X055).
One example of the issue in question was when a series of 20
(20 15 m grid) NQ2 drill holes intersected the down-dip Concluding comments
continuation of a known oreshoot. The drilling yielded In this case study the identification of proxies for gold has
79 one-metre samples with an un-cut weighted average grade of provided the geological team with a powerful method of
3.4 g/t Au, hardly indicative of economic viability when the verifying the presence of oreshoots from otherwise low-grade
breakeven cut-off grade is 6.2 g/t Au. The mined-out region drill intersections. The cost of developing the method was not
above the projected shoot yielded a grade of 17.4 g/t Au. Was great, and took approximately two calendar months of good
this economic shoot likely to continue below this level? Drilling geological data collection and interpretation. Results to date have
proved the reef continued, but does grade?
all been positive, with decisions to extend the depth of workings
resulting in mineable grades within the oreshoots.
Development of the proxies Referring back to the example given earlier in this case study
A series of studies were undertaken in accessible and where 20 drill holes yielded an average grade of 3.4 g/t Au. The
well-mineralised areas to identify any links between gold use of the proxies gave confidence that gold was likely to be
grade/mineralisation with textural, mineralogical and geochemical present, and eventual stoping yielded grades of between
features. A number of key relationships were identified as a result 11.8 - 15.3 g/t Au.
(Table 8).
DISCUSSION
Mineralogy
The application of proxies has clear potential to any high-nugget
Throughout the oreshoots arsenopyrite was present in variable gold vein where drill grades understate the likely true grade. This
quantities, and rarely up to about eight per cent. Rarely visible style of mineralisation is common throughout the Pacific Rim
quantities of bismuth minerals, usually bismuthinite, were region, and is generally seen as either mesothermal (lode-gold,
observed in polished sections. Outside of the oreshoot zones greenstone-hosted, etc) or epithermal quartz vein systems. In
arsenopyrite and bismuth minerals were not observed. most cases the development of proxies requires only careful

300 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 REDUCING GRADE UNCERTAINTY IN HIGH-NUGGET EFFECT GOLD VEINS

TABLE 9 REFERENCES
Comparison of original drill sample grade with proxies and Boyle, R W, 1979. The Geochemistry of Gold and its Deposits (Ottawa:
proximal bulk sample grade. Geological Survey of Canada), 584 p. Bulletin 280.
Charlet, J M and Quinif, Y, 1990. Quartz thermoluminesence (TL) in gold
Drill hole Drill Proxy Bulk Sample exploration: some applications, Mineralium Deposita, 25:S13-S20.
ID Grade Grade Cuffley, B W, Krokowski de Vickerod, J, Evans, T and Fraser, R, 1998. A
(g/t Au) (g/t Au) new structural model for fault-hosted gold mineralisation: an
X043 0.5 2 % arsenopyrite. Laminated 18.9 example from the Nick OTime oreshoot, Poverty Reef, Tarnagulla,
vein with stylolites. in The Second GPIC Conference on Developments in Victorian
Geology and Mineralisation, Bulletin No 24, pp 53-63 (Australian
X045 1.3 2.5 % arsenopyrite. Bismuthinite 28.6 Institute of Geoscientists: Perth).
observed. Very well-developed
laminations and stylolites. Dominy, S C, 1996a. Unpublished report, Clogau St Davids Gold Mines
Ltd, 8 p.
X049 1.4 3 % arsenopyrite. Laminated 22.3 Dominy, S C, 1996b. Unpublished report, Welsh Gold PLC, 9 p.
vein with stylolites.
Dominy, S C, 2001. Prediction of gold grade potential in erratic
X055 1.1 5.5 % arsenopyrite. Trace 137.5 vein-hosted deposits using a geochemical discrimination index,
bismuthinite. Very Trans Inst Min Metall (Section B), 110:B39-B43.
well-developed laminations and Dominy, S C, Phelps, R F G and Camm, G S, 1996. Geological controls
stylolites. on gold grade distribution in the Chidlaw Link Zone, Gwynfynydd
Z003 4.1 2 % arsenopyrite. Laminated 15.3 mine, Dolgellau, North Wales, UK, Trans Inst Min Metall (Section
vein with stylolites. B), 105:B151-B158.
Z033 3.3 3.5 % arsenopyrite. Bismuthinite 38.9 Dominy, S C, Cuffley, B W, Johansen, G F, Annels, A E and Platten, A E,
2000a. Resource evaluation of nuggety shale-hosted gold-quartz
observed. Very well-developed
reefs, in Proceedings Fourth International Mining Geology
laminations and stylolites.
Conference 2000, pp 169-186 (The Australasian Institute of Mining
Z035A 2.5 2 % arsenopyrite. Bismuthinite 19.6 and Metallurgy: Melbourne).
observed. Well-developed Dominy, S C, Johansen, G F, Annels, A E and Cuffley, B W, 2000b.
laminations and stylolites. General considerations of sampling and assaying in a coarse gold
Z036 2.7 1.5 % arsenopyrite. Laminated 9.7 environment, Trans Inst Min Metall (Section B), 109:B145-B167.
vein with stylolites. Dominy, S C, Johansen, G F and Annels, A E, 2001. Bulk sampling as a
tool for the grade estimation of gold-quartz reefs, Trans Inst Min
Metall, 110:B176-B191.
geological data collection and interpretation, which should be Dominy, S C, Johansen, G F, Cuffley, B W, Platten, I M and Annels, A E,
part of any resource development or grade control program. 2001. Estimation and reporting of mineral resources for coarse
It must be clearly remembered that proxies on the whole are gold-bearing veins, Explor Min Geol, 9(1):13-42.
not an alternative to gold assaying. They may permit the Dominy, S C, Platten, I M and Raine, M D, 2003. Grade and geological
calculation of grades for Mineral Resources, only in certain continuity in high-nugget effect gold-quartz reefs: implications for
well-constrained cases (Dominy, unpublished data). Any attempt resource estimation and reporting, Trans Inst Min Metall,
to use the method to report grades is unlikely to pass any due 112:B239-B259.
diligence study unless there is very good reconciliation with Dominy, S C, Stephenson, P R and Annels, A E, 2003. Classification and
mining. Even if this were to be the case, the method would only reporting of Mineral Resources for high-nugget effect gold vein
deposits, Explor Min Geol, 10(3):215-233.
allow the definition of specific grade ranges and not individual
grades. Proxies provide increased confidence to the evaluation Dominy, S C, Annels, A E, Platten, I M and Raine, M D, 2003. A review
of problems and challenges in the resource estimation of high-nugget
process, and may aid targeting for drilling and underground
effect lode-gold deposits, in Proceedings Fifth International Mining
development. Their use could yield benefits during the Geology Conference, pp 279-298 (The Australasian Institute of
exploration stage when otherwise apparently barren or low-grade Mining and Metallurgy: Melbourne).
gold intersections could be misleading. Dominy, S C, 2004. Fundamental sampling error and its relationship to
As with all vein-type gold deposits the exploration, evaluation the nugget effect in gold deposits, in Proceedings EGRU Mining and
and exploitation process must be guided by strong geological Resource Geology Symposium, James Cook University of North
input and understanding, and sample data of appropriate size, Queensland, EGRU Contribution No 62, pp 30-45.
density and quality (Dominy et al, 2000a, 2003). Dominy, S C and Johansen, G F, 2004. Sampling extreme coarse-gold
reefs at the New Bendigo Project, Central Victoria, Australia, in
AusIMM Critical Issues in Sampling Workshop, Perth, 8 - 9
ACKNOWLEDGEMENTS September, p 12.
The authors are grateful to the North American and Russian Dowling, K and Morrison G, 1989. Application of quartz textures to the
mine operators for permission to publish this paper, which has classification of gold deposits using North Queensland examples,
Econ Geol Monograph, 6:342-355.
been composed within confidentiality agreements. Bendigo
Mining Ltd has supported work at the New Bendigo Gold Garnsey, R, 1993. Ore reserves and reconciliation in an orebody
containing coarse gold, in Proceedings Second International Mining
Project, and in particular Rowly Hill and Paul Quigley are Geology Conference, pp 207-212 (The Australasian Institute of
thanked for their part in developing the proxy scheme. Dr Ian Mining and Metallurgy: Melbourne).
Platten (GGI Consulting UK), Dr Alwyn Annels (SRK Johansen, G F, 1997. Sampling, assaying and reporting in a coarse gold
Consulting UK Ltd), and Matt Raine and Mark Williams (James environment, in Proceedings Assaying and Reporting Standards
Cook University) are thanked for useful discussions on gold Conference (Singapore: AIC Conferences), paper 4, p 9.
proxies. A James Cook University Merit Research Grant 2002 Johansen, G F, Raine, M D, Dominy, S C and Bartlett, J K, 2003.
supported part of the work. The North American case study was Challenges of sampling extreme nugget-effect gold-quartz veins at
undertaken during 1999 - 2000, and the Russian and Bendigo the New Bendigo Project, Central Victoria, Australia, in Proceedings
case studies during 2001 - 2003. Simon Dominy acknowledges a Fifth International Mining Geology Conference, pp 299-310 (The
Senior Lectureship at James Cook University funded by the Australasian Institute of Mining and Metallurgy: Melbourne).
Minerals Council of Australia (MCA). Louisa Tomas (James Johansen, G F, 2004. A solution to grade estimation in a high nugget
environment the Bendigo experience, in Proceedings PACRIM
Cook University) is thanked for her editorial skills, and The Congress 2004, pp 317-322 (The Australasian Institute of Mining
AusIMM reviewers for their comments on the manuscript. and Metallurgy: Melbourne).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 301

S C DOMINY and G F JOHANSEN
Nugus, M J and Dominy, S C, 2003. Grade control methods at the Marvel
Loch gold mine, Southern Cross, WA, in Proceedings Fifth
International Mining Geology Conference, pp 279-298 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
Parsons, B D, 2000. Grade distribution, sampling and resource evaluation
of the Nick OTime Shoot, Tarnagulla, Victoria, Australia, MSc
Mineral Resources thesis (unpublished), Cardiff University, 190 p.
Pelham, D A, 1991. The evaluation of vein gold deposits: Part I, Min
Tech, 57-61.
Pelham, D A, 1992. The evaluation of vein gold deposits: Part II, Min
Quarrying Tech Int, 31-34.
Platten, I M and Dominy S C, 2003. The occurrence of high-grade gold
pockets in quartz reefs at the Gwynfynydd Mine, Wales, United
Kingdom: A geological explanation of the nugget effect, Explor Min
Geol, 10(4):249-272.
Roberts, L S, Dominy, S C and Nugus, M J, 2003. Problems of sampling
mesothermal lode-gold deposits: case studies from Australia and
North America, in Proceedings Fifth International Mining Geology
Conference, pp 387-400 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
302 Adelaide, SA, 19 - 22 September 2004
Royle, A G, 1989. Splitting gold assay pulps containing coarse gold,
Journ Leeds Univ Mining Assoc, 1989: 63-68.
Shepherd, T J, Bottrell, S H and Miller, M F, 1991. Fluid inclusion
volatiles as an exploration guide to black shale-hosted gold deposits,
Dolgellau gold belt, North Wales, UK, Journ Geochem Explor,
42:5-24.
Vearnacombe, J R, 1993. Quartz vein morphology and implications for
formation depth and classification of Archaean gold-vein deposits,
Ore Geol Rev, 8:407-424.
Williams, M J, 2002. Development of a gold proxy index for the
Hadleigh Castle B Lode, Charters Towers, North Queensland,
Economic Geology Honours Thesis (unpublished), James Cook
University.
Wooldridge, R, 1998. Sample preparation and assaying of coarse gold
ores, in Proceedings AusIMM Annual Conference 1998, pp 143-154
(The Australasian Institute of Mining and Metallurgy: Melbourne).
PACRIM 2004
Hydrothermal Alteration at the Bowdens Silver Deposit, NSW
J Elliot1, T Leach2 and I Pringle3
ABSTRACT
The Bowdens Silver Deposit is a low sulfidation epithermal silver-base
metal deposit that has formed on the northern margin of a hydrothermal
system within air-fall breccias, ignimbrites and crystal tuffs of rhyolitic
composition. The deposit lies near the north-eastern margin of the
Lachlan Fold Belt and the rhyolitic host rocks are of Early Permian age.
The silver mineralisation occurs as flat-lying zones of disseminations and
silicic fracture-filling within the Rylstone Volcanics and is closely
associated with sulfides of iron, arsenic, lead and zinc. High-grade silver
mineralisation is also found in steeply-dipping fracture zones which host
banded sulfide veins. Studies of the alteration products indicate that high
temperature fluids, flowing from south of the current project area,
resulted in quartz (adularia) illite/sericite alteration. Cool, low pH
steam-heated waters on the northern and eastern margins of the prospect
area resulted in siderite smectite-illite clay Fe-carbonate marcasite
kaolinite alteration. Mixing of the two fluid types resulted in the
deposition of iron and base metal sulfides, a decrease in both sulfur
activity and pH of the mineralising fluid, and subsequent silver
mineralisation.
LOCATION
The Bowdens Silver Deposit is located in the Central Tablelands
region of New South Wales, (NSW), Australia, 29 km by road
south-east of Mudgee and 23 km north-west of Rylstone. The
deposit is on the eastern margin of the Lachlan Fold Belt and the
mineralisation is hosted by silicic pyroclastics of Early Permian
age (Rylstone Volcanics).
EXPLORATION HISTORY
The deposit was discovered in 1988 by CRA Exploration
(CRAE), during regional stream sediment sampling. Between
1988 and 1992 follow up drilling defined the extent of
outcropping mineralisation. The property was originally called
Bowdens Gift but is now known as the Bowdens Silver Deposit
(BSD), or simply Bowdens. Golden Shamrock Mines acquired
the ground in 1994 and in 1997 the property was purchased by
Silver Standard Resources Inc of Vancouver and is currently held
under Exploration Licence 5920.
Drilling since 1994 has extended the mineralised sequence
down dip towards the north from the discovery outcrops. On the
basis of topography and inferred faulting the deposit has been
divided into four zones. The original Bowdens Gift zone defined
by CRAE has been renamed Main Zone South and the northern
extension, beneath sandstone cover, is called Main Zone North.
Further drilling some 400 m west of the Main Zone has defined
another prominent zone of mineralisation, extending N-S for
600 m, which has been divided into Bundarra North Zone and
Bundarra South Zone. All these zones comprise the Bowdens
Silver Deposit.
A pre-feasibility scoping study which included metallurgical
testwork and environmental studies was completed in 1997. In
1998 further extensive drilling, metallurgical studies, and
1. Geological Consultant, Anzeco Pty Ltd, 26 Casey Circuit, Bathurst
NSW 2795.
2. Geological Consultant, Terry Leach and Co, 2713 Rings Road,
Coromandel, New Zealand.
3. MAusIMM, Exploration Manager, Silver Standard Australia Pty
Limited, 556 Crown Street, Surry Hills NSW 2010.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
geological mapping were completed. Environmental baseline
studies continued and included surveys on flora, fauna,
archaeology, meteorology, noise control and water. Further
drilling and metallurgical work have been undertaken since 1998.
Since discovery of Bowdens 403 holes totalling 43 764 m have
been completed at an average drillhole spacing of approximately
30 m.
LOCAL GEOLOGY
Alteration and mineralisation at Bowdens is hosted in a sequence
of rhyolite pyroclastic units of the Rylstone Volcanics, that
unconformably overlie fine grained Ordovician meta-sediments,
which crop out in the southeast corner of the prospect (Figure 1).
Post-mineral sandstones of the Shoalhaven Group (Permian)
unconformably overlie the pyroclastic sequence in the northern
project area (Figure 1, Pringle and Elliot, 1998).
The rhyolite pyroclastic sequence comprises four distinct,
relatively flat lying units. A lower thick unit of fine-grained
crystal to crystal lithic tuff is overlain by a thinner ignimbrite
unit, which in turn is overlain by a unit of coarser lapilli tuff and
tuff breccia. A thin (few metres) vitric tuff unit locally occurs in
the uppermost portion of the pyroclastic sequence. The basement
meta-sediments consist of very fine-grained meta-siltstone that
alternates with lenses of coarser grained meta-sandstone.
The Main Zone of mineralisation is bounded to the east by a
north to NNW trending structure, whereas mineralisation at
Bundarra North and South are aligned along another north/NNW
fault zone in the western part of the prospect area (Figure 1). The
pyroclastic sequence has been downfaulted within a graben-like
feature associated with these fault zones. A series of ESE faults
are subsidiary to the N/NNW structures.
ALTERATION AND MINERALISATION
Style
The silver-base metal mineralisation is mainly hosted within
air-fall breccias, ignimbrites and crystal tuff units, with minor
mineralisation extending into the underlying sediments. The
ignimbrites, and to a lesser degree the tuffs, are frequently
welded and therefore are excellent hosts to crackle, mosaic and
rotational breccia-style mineralisation, in which ore minerals are
deposited within the matrix and the siliceous fill cementing the
breccias. The crystal tuffs, however, display less brecciation and
locally, the mineralisation may be finely disseminated within the
altered matrix. A later stage of mineralisation is represented by
crustiform veining, up to 100 mm in width, often within
subvertical shear or fault zones, and comprising
quartz-carbonate-sulfide assemblages. The zones of strong
silicification and fracturing, notably within the ignimbrite units,
are considered to have been the loci of mineralising fluids.
The mineralisation occurs as stacked lenses within a 20 - 80 m
thick zone extending from surface, and near surface, to vertical
depths of about 200 m and dipping northwards at less than
30 degrees. However, this zone is not uniformly mineralised and
irregular barren intervals are present. The BSD, as delineated at
present by drilling, extends 600 m east-west and 700 m
north-south. In the east and north the mineralisation is terminated
by inferred steeply dipping faults. The southern and western
limits have not yet been clearly determined.
303
J ELLIOT, T LEACH and I PRINGLE
FIG 1 - Geology plan of Bowdens Silver Deposit.
Sequence
The sequence of hydrothermal events at Bowdens is initiated by
an early stage of fluidised brecciation in which pyroclastic
fragments have been transported, milled and sealed in a
quartz-illitic clay-pyrite altered comminuted matrix (Figure 2).
The breccia event was followed by a polyphasal event that
grades from early to late stages of wallrock replacement and
cavity, fracture and breccia fills dominated by: quartz sulfides
carbonate clay. In some samples this sequence is repeated,
indicating a periodically pulsating hydrothermal system. This
sequence of events is identical to that recognised in the
South-West Pacific carbonate-base metal-gold systems (Corbett
and Leach, 1998).
Quartz replacement and deposition took place throughout all
hydrothermal events, but dominated in early events that
post-dated the formation of fluidised breccias. Early stage quartz
is fine grained and intergrown with illitic clay + pyrite, whereas
later quartz is coarser grained, clear and contains intergrowths of
illitic clay, pyrite, arsenopyrite and/or sphalerite. Primary
feldspar has undergone initial alteration to adularia, which also
lines some fractures and cavities, and is replaced by later illite
clays or sericite. Traces of hydrothermal apatite are locally
intergrown with the quartz.
The abundance of sulfides increases during later quartz events
and continues into subsequent carbonate-dominated events. In
many of the open spaces, quartz is overgrown by sulfide minerals
that exhibit an overall depositional sequence of pyrite
304 Adelaide, SA, 19 - 22 September 2004
arsenopyrite sphalerite galena tennantite (chalcopyrite)
silver minerals, although there is considerable overlap in this
sequence. In most cases the sulfide minerals are deposited in open
spaces; however they are also the product of the replacement
minerals, especially mafic crystal fragments and to a lesser degree
feldspar fragments and vitric ash.
Sphalerite is the most abundant base metal sulfide mineral, and
exhibits an overall change from Fe-rich (red-orange in thin
section) cores to Fe-depleted (pale yellow to yellow) rims,
indicative of progressive cooling during mineralisation. However
these zonations are polyphasal, again illustrating multiple pulses
of hydrothermal activity. Chalcopyrite occurs in only trace
amounts and reflects the copper-poor content of the
hydrothermal fluids at this level in the system.
Carbonate minerals are intergrown with, and commonly
overgrow the quartz and sulfide minerals. The early carbonate
species are manganiferous, whereas later carbonates are more
iron-rich. The carbonate becomes progressively finer grained
during later depositional events, and locally forms very fine
grained multiple colloform banding.
Clay minerals are the dominant wallrock replacement minerals
and are also deposited in fractures and breccias with quartz,
sulfide and carbonate minerals. Clay minerals also commonly fill
open spaces and overgrow all other depositional minerals.
Depositional kaolinite overgrows and partially replaces illitic
clay, and locally contains marcasite and traces of galena and
silver minerals.
PACRIM 2004
 HYDROTHERMAL ALTERATION AT THE BOWDENS SILVER DEPOSIT, NSW
FIG 2 - Interpreted paragenetic sequence at Bowdens.
Silver mineralisation
Silver mineralisation took place mainly during the carbonate
event and extended into the clay-rich depositional events. Silver
occurs in tennantite (McConachy, 1989), as silver sulfosalts
(pearceite-polybasite > pyrargyrite proustite), silver sulfide
(argentite, acanthite) and native silver. Microprobe analyses
(McConachy, 1989; Pontifex, 1995) also identified
stephanite/argyrodite, plagionite and cerargyrite. The silver
content of the galena has not been determined. The overall
sequence of deposition of the main silver minerals is:
(?Ag-galena) Ag-tennantite/freibergite native silver
argentite/acanthite pyrargyrite-proustite pearceite
polybasite.
Distribution in alteration and mineralisation
The distribution in illitic clay minerals in plan view at the 500 m
level is illustrated in Figure 3, as identified by x-ray diffraction
(XRD) analyses. Illite and well-crystalline sericite define a
relatively high temperature, N-S trending regime that coincides
with a major fault in the western portion of the prospect area and
broadens to the south. The clay minerals grade to interlayered
smectite illite (high smectite content) over short distances to
the west, suggesting rapidly cooling conditions in that direction.
To the east however, the broad zones of interlayered
illite-smectite, with a progressive increase in smectite content, to
the east is interpreted to reflect slowly cooling conditions,
possibly associated with an outflow zone. The smectite content
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
of the interlayered clays increases with depth in the northeastern
drillholes, suggestive of temperature reversals associated with the
postulated outflow zone.
Based on XRD data, Fe-rich carbonate minerals (siderite,
mangano-siderite) dominate in the eastern portion of the prospect
area, whereas manganese magnesium (rhodochroiste,
kutnahorite, ankerite) are the main carbonate minerals identified
in the western drillholes (Figure 3).
From polished thin section analyses, arsenopyrite occurs
mainly in the south and west, whereas abundant marcasite is
closely associated with Fe-carbonate formation in the north and
east. The distribution in the Fe-content of the sphalerite (based
on colour index in thin section) follows that of the illitic clays,
and is indirectly related to the temperature of mineralisation
(Simmons et al, 1988). Fe-poor sphalerite indicative of cooler
conditions occurs mainly in the northeastern drillholes and at
shallower levels, whereas darker, Fe-rich sphalerite is overall
restricted to the western drillholes. Galena (and higher lead
assays) dominates over sphalerite (and Zn-assays) as the main
base metal sulfide at shallower levels, especially in the
northeastern drillholes. Silver minerals are mainly restricted to
samples selected from the north and east drillholes, as would be
expected from the distribution in ore zones. Overall tennantite
(Cu-rich silver sulfosalt) occurs at deeper levels, whereas
Sb-As-Ag ore minerals (pearceite, pyrargyrite, argentite/
acanthite, native silver) occur in the northeast and at shallow
levels.
305
J ELLIOT, T LEACH and I PRINGLE
FIG 3 - Distinction in clay minerals and siderite at 500 m level.
MINERALISATION MODEL
The above distribution of the mineralisation and alteration
indicates the following:
1. The Bowdens Silver Deposit formed on the northernmost
margin of a hydrothermal system. High temperature fluids
originating south of Bowdens flowed northwards along a
major structure on the western side of the deposit and
northeast along open structures or fracturing. This resulted
in quartz (adularia) illite/sericite pyrite arsenopyrite
assemblages.
2. Cool, low-pH steam heated waters, comparable to that
encountered around the margins of many active geothermal
systems (eg Broadlands; Simmons and Brown, 2000),
permeated the volcanic units in the north and east, and
resulted in an alteration assemblage comprising siderite
smectite-rich illite clays Fe-carbonate marcasite
kaolinite.
3. The mixing of the two fluid types resulted in deposition of
iron and base metal sulfides, a subsequent decrease in both
sulfur activity and pH of the mineralised fluid, and
subsequent silver mineralisation. The clay minerals
associated with the ore zones indicate that silver base
metal mineralisation took place at less than 150 - 200C.
306 Adelaide, SA, 19 - 22 September 2004
The almost total absent of gold at Bowdens is enigmatic, and
may reflect a geological environment comparable to that
associated with silver-base metal provinces in southern USA and
Mexico (Simmons et al, 1988).
REFERENCES
Corbett, G J and Leach, T M, 1998. Southwest Pacific rim gold-copper
systems: structure, alteration and mineralization, SEG Special
Publication No 6, 236 p.
McConachy, T F, 1989. Bowdens Gift Prospect, NSW, Internal CRAE
report #16302.
Pontifex, I R, 1995. Mineralogical Report No 6931.
Pringle, I J and Elliot, J, 1998. Bowdens silver-lead-zinc deposit,
Mudgee, in Geology of Australian and Papua New Guinean Mineral
Deposits (Eds: D A Berkman and D H Mackenzie), pp 627- 634 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
Simmons, S F and Browne, P R L, 2000. Hydrothermal minerals and
precious metals in the Broadlands Ohaaki Geothermal System:
Implications for understanding low-sulphidation epithermal
environments, Economic Geology, 95:971-1000.
Simmons, S F, Gemmel, J B and Sawkins, F J, 1988. The Santo Nino
silver-lead-zinc vein, Fresnillo, Mexico, Part II. Physical and
Chemical Nature of Ore-Forming Solutions, Economic Geology,
83:1619-1641.
PACRIM 2004
Results From Induced Polarisation Surveys Over the Beruang
Copper-Gold Deposit in Central Kalimantan, Indonesia
P J Elliott1
ABSTRACT
The Beruang Copper-Gold Deposit is located very near the centre of
Kalimantan in Indonesia. It is a dacitic tuff hosted deposit currently being
explored by Kalimantan Gold Inc of Canada. Exploration activity to date
at Beruang has been focused on the Main Zone, which is defined by a
north-south elongate, 1 1.5 km zone of anomalous copper and gold, and
intense phyllic alteration. The disseminated sulfides responsible for the
anomalous copper values produce a moderate to high chargeability
signature.
Phyllic alteration is overlain at high levels, and on the western edge of
the Main Zone, by advanced argillic alteration. The dacite tuffs hosting
the mineralisation are cut by post-mineralisation dacite dykes that are
delineated by ground magnetics and geological mapping. Base metals,
principally zinc and lead, form broad halos around the Main Zone. These
metals are incorporated in massive NE-trending, polymetallic
mineralisation that outcrops to the north and south of the Main Zone.
IP survey results highlight the highly pyritic NNE-trending shear zone,
which hosts most of the copper mineralisation in the Main Zone. A high
chargeability anomaly at depth in the central-east part of the Main Zone
is interpreted to be the source for mineralising hydrothermal fluids. This
is suggested by zonations in alteration and mineralisation. A deep
chargeability anomaly can be correlated with the downward plunge of a
pyritic breccia, encountered at shallow levels in drillholes, and is a
potential breccia pipe target. Compilation and inversion modelling of the
IP results has yielded some interesting drill targets.
1. MAusIMM, Elliott Geophysics International Pty Ltd, PO Box 204,
Floreat WA 6014. E-mail: elliottgeophysic@aol.com
FIG 1 - Regional location plan for the Beruang Kanan Project.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
INTRODUCTION
Geophysical methods have been used to assist exploration for
gold and copper porphyry type deposits probably since the
1930s; when magnetics, gravity, and resistivity were starting to
be used on a frequent basis. Certainly by the 1960s there was a
boom in exploration for porphyry copper-gold deposits
throughout Indonesia, Papua New Guinea, and the South Pacific
Islands. The boom encouraged the use of the relatively new
induced polarisation (IP) method, developed to a fine art in North
America during the 1950s (Bleil, 1953; Wait, 1959; Madden and
Cantwell, 1967). Since the 1960s, IP has been used extensively
around the Pacific Rim in search of porphyry copper-gold targets
(Elliott 1994, 1999, Irvine and Robertson, 1987, Maula and
Levet, 1996, Meldrum et al, 1994) and has been used recently in
Central Kalimantan to outline the Beruang Kanan copper-gold
and base metal deposit.
A semi-detailed IP/resistivity survey completed over the
Beruang Kanan Project, in Central Kalimantan (Figure 1) used a
150 m dipole-dipole array. Some 16 lines (80 00N to 10 950N)
were surveyed with one line (10 350N) being surveyed twice with
two different dipole sizes (150 m and 300 m). The modelling of
the individual IP/resistivity pseudo-sections obtained from the
surveys was accomplished using RESIX IP2DIv4, an Interpex
software package for 2D IP/resistivity modelling. The program
includes options to perform 2D polygonal modelling (similar to
version 3), Zonge style inversion modelling, and smooth image
modelling. The models presented in this paper were all produced
using the Zonge style inversion modelling approach.
REGIONAL LOCATION PLAN
BERUANG KANAN PROJECT
KALIMANTAN GOLD
307
P J ELLIOTT
Various models for porphyry copper systems have been
proposed by specialists such as Sillitoe, 2000. More specific to
the Kalimantan geological setting are papers by Carlile and
Mitchell, 1994, Corbett and Leach, 1998, and Van Leeuwen et al,
1990. A porphyry copper-gold model can be proposed similar to
that presented in Figure 2, which shows the inferred positions of
the Central Kalimantan prospects on the right hand side of the
diagram (Kalimantan Gold Corp, 2004). Beruang Kanan is
shown to sit just above the main porphyry intrusive in the
propylitic alteration zone. These geological models assist the
interpretation of IP data collected over the project area.
GEOLOGY
Beruang Kanan Project
Beruang Kanan is located in the SE corner of the Kalimantan
Suraya Kencana CoW and can be reached from Palangka Raya,
the regional capital of Central Kalimantan, by speedboat and
logging road. A well-established base camp exists at Beruang
Kanan. The Beruang Kanan porphyry copper system, described
recently by Geiger et al, 2002, is hosted in a 10 - 12 km wide
circular topographic feature, which occurs at the intersection of
the regional NE and NNW trending structures. At Beruang
Kanan, various alteration styles include: propylitic alteration,
local phyllic alteration, and rare advanced argillic alteration.
Anomalous copper (>200 ppm) in soils extend over an area of
more than 16 km2 and is open in most directions. Alteration and
mineralisation are hosted in a sequence of dacite tuffs of
probable Oligocene age, which overlay lower Tertiary
volcanoclastic siltstones and sandstones in the eastern prospect
area. Pre-mineral Sintang dacite porphyry intrusions of probable
Oligocene age, and post-mineral andesite, dacite to
basalt-gabbros dykes, of possible Miocene age, are intruded into
the tuffs and sediments. Structures in the Beruang Kanan area are
dominated by the regional NE and NW lineaments, as well as
local E-W features.
FIG 2 - Porphyry copper model, Beruang Kanan Project, Central Kalimanatan. (Courtesy of Kalimantan Gold).
308 Adelaide, SA, 19 - 22 September 2004
Geological, geochemical, and geophysical (induced
polarisation and ground magnetic) surveys have delineated three
centres of porphyry-style alteration and mineralisation:
Beruang Kanan Main Zone,
Beruang Kanan Silver and Associated Minerals Prospect, and
Beruang Kanan Main South Zone.
Beruang Kanan Main Zone (BKMZ) (porphyry
copper)
Exploration activity to date at Beruang Kanan has been focused on
the Main Zone. The BKMZ is defined by: a north-south elongate,
1 1.5 km area of anomalous copper >0.1 per cent in rock chips
gold-molybdenum geochemistry; high chargeability; and intense
phyllic alteration (Figure 3).
The phyllic alteration is capped at high elevations on the
western margin of the BKMZ by advanced argillic alteration.
Alteration and mineralisation are hosted almost entirely in dacite
tuff. These are cut by post-mineral dacite dykes that are defined
by ground magnetics and geological mapping to be up to
100 - 200 m wide and radiate north-west and west through the
prospect area. Copper-gold-molybdenum in soils are aligned NE
within the BKMZ, possibly indicating that possible ore shoots
are aligned in this direction.
Zinc, and to a lesser extent lead, form broad halos around the
BKMZ, and occur as massive NE-trending, polymetallic
mineralisation that outcrops to the north and south of the BKMZ.
Limited drilling into these zones to the north intersected up to
16 m @ 2.8 per cent lead, 5.8 per cent zinc, 58 g/t silver, 0.65 g/t
gold and 0.17 per cent copper (DDH BKZ-1) associated with
quartz-chlorite-illite-base metal sulfide (sphalerite, galena,
tennantite, chalcopyrite-bornite) as wall rock disseminated grains,
in shear zones and as sheeted veins. Such mineral assemblages are
often associated with transitional phases between a high
sulfidation system and a porphyry.
PACRIM 2004
 RESULTS FROM INDUCED POLARISATION SURVEYS OVER THE BERUANG COPPER-GOLD DEPOSIT
FIG 3 - Copper contour analysis map, Beruang Kanan Project, Central Kalimantan. (Courtesy of Kalimantan Gold).
Between 1998 and 2001, 25 diamond drillholes drilled by
KGC to 280 m depth have intersected a NNW-trending zone of
intensely sheared and silicified, highly pyritic, zoned phyllic to
advanced argillic alteration, with copper grades up to 167 m @
0.59 per cent copper. Higher grade copper (> 0.5 per cent Cu)
occurs in two zones: a shallow flat lying zone (down to 375 m
RL) that is composed predominantly of supergene covellite and
chalcocite forming a sort of supergene blanket; and a deeper
zone below 300 m RL that is composed mostly of hypogene
chalcopyrite.
The IP surveys display well the highly pyritic NNE-trending
shear zone, which hosts most of the copper mineralisation in the
BKMZ. A high chargeability anomaly at depth in the central-east
part of the BKMZ is interpreted to be the source for mineralising
hydrothermal fluids. This is indicated by zonations in alteration
and mineralisation.
Chargeability anomalies at progressively shallower levels
reflect the inferred outflow to the NNE and also to the
south-west. It is postulated that the deep chargeability anomaly is
the downward plunge of the pyritic milled matrix breccia
encountered at shallow levels in DDH BK-8, and is a potential
target for a breccia pipe.
There are two copper mineralisation models that can be
conceptualised:
1. copper mineralisation may be genetically related to the
breccia pipe, with copper bearing fluids coming in late and
migrating upwards along the side to the quartz-pyrite shear
zone; or
2. copper mineralisation may be derived from a source other
than the inferred breccia pipe, eg represented by a high
chargeability anomaly at depth immediately north of the
BKMZ.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Beruang Kanan South Zone (BKSZ) (porphyry
copper)
The Beruang Kanan South Zone is located 2.5 km south of the
Beruang Kanan base camp with direct logging road access. The
BKSZ consists of a 1 km NE-striking central zone of locally
intense sericite-quartz-pyrite and elsewhere argillic alteration,
stockwork quartz + sulfide veins and veinlets, weak anomalous
gold-molydenum in soils (Figure 4) and a deep IP chargeability
anomaly. A broad outer halo of anomalous copper-zinc-arsenic-
antimony in soils occurs in intensely chlorite-pyrite altered tuff.
Although copper is depleted in the central silica-pyrite-clay
zones, leaching of copper from the veins in this zone is not
evident. However blebs of covellite/chalcocite were observed to
coat locally some of the vein pyrite indicating that copper has
been mobilised in this central zone.
Supergene copper sulfides are abundant in the zone of high
copper in soils, identify as an area 800 400 m with values of
greater than 350 ppm. It is postulated that much of the copper in
this soil anomaly has been remobilised from higher elevations.
However, quartz-pyrite-chalcopyrite veins do cut the
chlorite-altered tuff within the copper-soil anomaly at lower
elevation and to the south-west of the central silica-pyrite
alteration.
Petrological examinations show that native gold occurs as
inclusions in the chalcopyrite. Surface samples have returned
assays of up to 15 g/t gold and one per cent copper over 2 m,
2230 ppm copper over 12 m, with one rock chip sample
returning nine per cent copper. It is believed that these
copper-gold bearing quartz-pyrite veins have formed around the
margins of a hydrothermal system which is centered upon a soil
gold-molybdenum anomaly, intense quartz-sericite-pyrite and
argillic alteration, and quartz sheeted to stockwork veinlets. A
deep zone of high chargeability underlies this central zone.
309
P J ELLIOTT
FIG 4 - Soil geochemistry and geology structure, South Polymetallic Zone, Beruang Kanan, Central Kalimantan.
(Courtesy of Kalimantan Gold).
Beruang Kanan silver and associated minerals
prospect (BKSAMP)
The Beruang Kanan Silver and Associated Minerals Prospect is
situated between the margins of the BKMZ and BKSZ copper
porphyry centres of the Beruang Kanan Prospect. Lithologies
comprise dacitic tuff, breccia, dacite porphyry, andesite dykes
and fine grain tuffacious sediments. These rocks have undergone,
moderate to strong phyllic, and lesser argilic alteration, while the
southern boundary of the area, is propylitised dacite porphyry
(Figure 5).
Both soil and rock geochemistry results from the area are
sporadic but with some extremely high assays from rock samples
up to 3440 g/t silver, 26 per cent lead, 35 per cent zinc, 1 g/t
gold, and three per cent copper. These anomalies appear to be
structurally controlled and are considered the result of leakage
from an underlying mineralising source.
IP chargeability data suggests that a large, well-defined body
900 250 m underlies barren tuffs at a depth of 150 m. Low
resistivity (conductive) responses implying strong alteration, and
in association with the high chargeability anomaly, suggests a
mineralised body is located within a NE-SW structural corridor.
The SE end of this chargeability anomaly however forms a 400 m
diameter, spherical body of high resistivity. This is believed to be a
siliceous replacement body, containing mineralisation. Milled
breccia rocks are also observed to contain large 1 m diameter casts
of massive polymetallic mineralisation.
310 Adelaide, SA, 19 - 22 September 2004
This 900 250 m IP chargeability anomaly fits perfectly
within an intense set of parallel NE trending faults that have in
most cases, offset a series of equally intense NW faults. The
resultant jog effects, represents an excellent environment for
mineral deposition. High-grade polymetallic mineralisation can
be expected to occur within these dilational structures and at
their intersections.
The Beruang Kanan Silver and Associated Minerals Prospect
displays potential for hosting a large, high-grade polymetallic
deposit.
IP MODELLING RESULTS
Inversion models are presented as: sections (Figures 6 and 7);
interpreted sections (eg Figure 8); as image plans (Figures 9 and
10); and as wire frame perspective plots (Figures 11 and 12).
Some 17 models were completed in total, one for each of the
dipole-dipole IP/resistivity sections from lines 8000N to 10 950N.
The inversion models usually produce error fits to the original
data in the order of five per cent for resistivity and one per cent
or two per cent for chargeability. The error of fit can be seen by
comparing the modelled data to the original field data. Each
inversion model takes a few minutes to run depending on
required resolution and the size of the section being modelled,
and the speed of the computer being used. Most time is spent on
formatting data in preparation for modelling and on hard copy
presentation. Further time has been spent on preparing plans of
chargeability and resistivity at various depth ranges and 3D
perspective plots of the model results.
PACRIM 2004
PACRIM 2004

9934000 S
0 200 600 1000 M

LEGEND :

9933000 S
U PLEISTOCENE YDA Young dacite volcanics (YDA)
D
0.1 million years SCS Silica sand (SCS)

U
D PLIOCENE AND Andesite usually as dykes, sills, unaltered (AND)

RESULTS FROM INDUCED POLARISATION SURVEYS OVER THE BERUANG COPPER-GOLD DEPOSIT
D 1.8 million years HAD Horblende dacite, relatively unaltered (HAD)
U

D
U D MIOCENE DAP Dacite porphyry (DAP)
U U U
D
D
U 5.0 million years SLC Silica cap (SLC)
D
D
U U Collaps breccia (CBX) Hydrothermal breccia (HBX)
D
DU
U
BX Milled breccia (MBX) Fluidised breccia (BXF)
D

D
U U Diatreme breccia (DBX) Fault breccia (FBX)
D Tectonic breccia (TBX) Contact breccia (BXC)
SRY Rhyolite (SRY)
(SIN)
U
D SDA Dacite (SDA)

U
OLIGOCENE DAT Dacitic tuff (DAT)

9932000 S
23.5 million years
U
D
QDT Quartz dacitic tuff (QDT)
D
U D U
U
D D MQD Micro quartz diorite (MQD)
Adelaide, SA, 19 - 22 September 2004

U
D U

QZD Quartz Diorite (QZD)

D

U PALEOCENE SLT Siltstone (SLT)

D
D
U

58.0 million years SST Sandstone (SST)

TST Tuffaceous siltstone (TST)

U D

U D U
D
U
D U
TSS Tuffaceous sandstone (TSS)
D
D

U
TUF Tuff (TUF)

U
D

9931000 S
D

U
D

U
D

9930000 S
9929000 S
PT. KALIMANTAN SURYA KENCANA
GEOLOGY INTERPRETATIVE MAP
BERUANG KANAN PROJECT
CENTRAL KALIMANTAN

765000 E 766000 E 767000 E 768000 E 769000 E 770000 E 771000 E

FIG 5 - Geology Beruang Kanan, associated mineralisation, Central Kalimantan. (Courtesy of Kalimantan Gold).
311
P J ELLIOTT

FIG 6 - Inversion model of the 150 m dipole-dipole section for line 10 350N, Beruang Kanan Prospect. (Courtesy of Kalimantan Gold).

FIG 7 - Inversion model of the 300 m dipole-dipole section for line 10 350N, Beruang Kanan Prospect. (Courtesy of Kalimantan Gold).

312 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RESULTS FROM INDUCED POLARISATION SURVEYS OVER THE BERUANG COPPER-GOLD DEPOSIT
FIG 8 - Main Zone EW, interpreted IP chargeability section, Central Kalimantan. (Courtesy of Kalimantan Gold).
Line 10 350N was surveyed twice, once with 150 m dipoles
and again with 300 m dipoles. The inversion models for these
sections are presented here as Figures 6 and 7. There are a few
obvious anomalies, which have high chargeability and/or high
resistivity within the modelled sections. The discordant highly
resistive zones with low chargeability represent areas of intense
silica alteration. The zones of low resistivity and high
chargeability represent areas of clay-sulfide alteration. There are
noticeable vein structures also within the sections. A detailed
interpretation can be drafted for each section using a
combination of the known geology and the model IP/resistivity
sections.
It is interesting to note that the broad horizontal oval shaped
anomaly in the 150 m dipole section centred on 13 600E (Figure
6) is also well defined on the 300 m dipole section (Figure 7).
The 300 m dipole section, with its increased depth of
investigation shows a modelled root zone to the original
resistivity anomaly. It is presumed that this is a connection
between the underlying diorite porphyry pluton and the near
surface silica rich alteration zone. The higher chargeability zone
between 14 000E and 14 500E on the 150 m dipole section also
shows an extension to depth, with the peak response being
centred on about 14 500E for the 300 m dipole section. The
higher chargeability zone appears to bracket the central high
resistivity zone in the upper part of the section this is typical of
zoning in a mineralised diorite porphyry system.
Included as Figures 9 to 10, are some image plans of model
chargeability and resistivity. These model plans, were produced
by assigning numbers to each grid level in the models and
thereby producing data sets with numbers similar to n-spacings.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
These levels are not at constant depth or RL, but are effectively
draped under the topography. This can be seen in the posted data
for each of the modelled sections. The model data at different
levels are not equivalent to the original n-spacings in the pseudo
sections. The RL for each data point is exported with the
chargeability and resistivity information, so if a particular level is
useful for presentation of drillhole targets and positions, then an
average RL plan can be produced for that particular level. The
plan plots for Levels 1 and 2 are most useful for analysing the
near surface high resistivity zones associated with silica
alteration, and also the near surface chargeability variation. Level
7 on the other hand is the deepest level in the models, and defines
the deep high chargeability trends very well as well as the broad,
deep, high resistivity zones that may represent buried porphyries.
The resistivity plan for Level 4 (Figure 9), shows a very high
resistivity zone in the northern central area. The anomaly shows
up as a broad oval high with its main peak near the northern most
end. This is the same anomaly that is seen on the previously
presented model sections (Figures 6 and 7). The Level 4
chargeability plan (Figure 10), shows an anomalous eastern zone,
which is broken up into an agglomeration of individual
chargeability highs presenting a quasi-circular pattern. The
eastern chargeability zone is a broad anomalous zone about
1.5 km wide by 2.5 km long. The anomaly in the NW corner
appears more as a blob with a northerly trending linear zone.
The high resistivity is matched by a depression in the
chargeability. The modelled chargeability shows a broad
anomalous response surrounding the resistivity. This pattern is
typical for a mineralised porphyry system and probably
represents the different zones of alteration.
313
P J ELLIOTT

FIG 9 - Plan image of the resistivity inversion model for all 150 m dipole-dipole data from the Beruang Kanan Prospect.
(Courtesy of Kalimantan Gold).

FIG 10 - Plan image of the chargeability inversion model for all 150 m dipole-dipole data from the Beruang Kanan Prospect.
(Courtesy of Kalimantan Gold).

314 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 RESULTS FROM INDUCED POLARISATION SURVEYS OVER THE BERUANG COPPER-GOLD DEPOSIT

FIG 11 - Wireframe perspective of modelled chargeability at Level 4, Beruang Kanan Prospect, Central Kalimantan.
(Courtesy of Kalimantan Gold).

FIG 12 - Wireframe perspective of modelled resistivity at Level 4, Beruang Kanan Prospect, Central Kalimantan.
(Courtesy of Kalimantan Gold).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 315

P J ELLIOTT
CONCLUSIONS
Inversion modelling has been completed on all IP/resistivity
dipole-dipole sections surveyed during the period April to May
2001. The inversions are simple to achieve and relatively quick.
They provide the best estimate available from the field data for
the actual chargeability and resistivity variations below the
ground surface. The inversion model sections can be used
directly to design drillholes for testing the main IP/resistivity
anomalies.
During the modelling process a couple of large resistivity
highs were outlined that may represent relatively deep seated
porphyry intrusives, which are possible sources to the near
surface mineralisation discovered to date. The zones of near
surface silica alteration were defined as high resistivity zones in
the upper portions of each section. Possible feeder zones to the
near surface mineralisation were also defined as curvilinear high
resistivity zones rising from deep in the sections to near surface.
The chargeability highs to date, have represented sulfide
related mineralisation in the top 200 m of the drilled sections.
There appears to be a circular or oval shaped chargeability high
surrounding the main resistivity high in the central-northern part
of the survey area. This probably represents a sulfide rich
alteration zone around the main porphyry intrusive. The
chargeable halo has been tested at a number of locations and
found to contain significant mineralisation.
The IP/resistivity survey supports the existence of a deep
seated and mineralised porphyry complex, with a well developed
stockwork system extending above it. The porphyry intrusives
forming this complex are likely to contain disseminated
mineralisation, which would be related to the near surface
mineralisation discovered to date.
ACKNOWLEDGEMENTS
The author would like to thank PT Kalimantan Suraya Kencana
and its parent company Kalimantan Gold Incorporated of Canada
for allowing the publication of the results contained in this paper.
In particular, a special thanks to Mansur Geiger for supplying the
original data for modelling and giving feed back on the relevance
of the resultant inversion model to the known geology.
316 Adelaide, SA, 19 - 22 September 2004
REFERENCES
Bleil, D F, 1953. Induced polarisation: a method of geophysical
prospecting, Geophysics, 18:636-661.
Carlile, J C and Mitchell, A H G, 1994. Magmatic arcs and associated
gold and copper mineralization in Indonesia, Journal of Geochemical
Exploration, 50:91-142.
Corbett, G J and Leach, T M, 1998. Southwest Pacific Rim gold-copper
systems: Structure, alteration and mineralization, SEG Special
Publication No 6, 236 p.
Elliott, P J, 1994. Geophysical studies of the Nena massive sulphide and
Frieda River porphyry copper deposits, Papua New Guinea, in
Proceedings PNG Geology, Exploration and Mining Conference,
(Ed: R Rogerson) pp 138-146 (The Australasian Institute of Mining
and Metallurgy: Melbourne).
Elliott, P J, 1999. Geophysical methods over Pacific Rim Deposits, in
Proceedings Pacrim 99 Congress, pp 681-691 (The Australasian
Institute of Mining and Metallurgy: Melbourne).
Geiger, M, Prasetyo, D and Leach, T, 2002. Porphyry copper gold
systems in Central Kalimantan, in Proceedings PDAC, Canada,
March, p 8.
Irvine, R J and Robertson, I, 1987. Interpretation of airborne geophysical
data over the Ok Tedi porphyry copper-gold orebody using image
processing techniques, in Proceedings Fifth Australian Society of
Exploration Geophysicists Conference, pp 103-107.
Kalimantan Gold Corp, 2004. Description of the Beruang Kanan Deposit,
http://www.kalimantan.com
Madden, T R and Cantwell, T, 1967. Part D. Induced polarisation, a
review, in Mining Geophysics Volume II, Theory, pp 373-400 (SEG).
Maula, S and Levet, B, 1996. Porphyry copper gold signatures and the
discovery of the Batu Hijau Deposit, Sumbawa, Indonesia, in
Proceedings, Porphyry Related Copper Gold Deposits of the
Asia-Pacific Region, pp 8.1-8.13 (Australian Mineral Foundation).
Meldrum, S J, Aquino, R S, Gonzales, R I, Burke, R J, Suyadi, A, Irianto,
B and Clarke, D S, 1994. The Batu Hijau porphyry copper gold
deposit, Sumbawa Island, Indonesia, Journal Geochem Expl,
50:203-220.
Sillitoe, R H, 2000. Role of gold-rich porphyry models in exploration, in
Gold in 2000. Reviews in Economic Geology, Vol 13 (Eds: S G
Hagerman and P H Brown), pp 311-346.
Van Leeuwen, T M, Leach, T M, Hawke, M M and Hawke, A A, 1990.
The Kelian disseminated gold deposit, East Kalimantan, Indonesia,
Journal of Geochemical Exploration, 35:1-61.
Wait, J R (Ed), 1959. Overvoltage Research and Geophysical Applications
(Pergamon Press).
PACRIM 2004
A Solution to Grade Estimation in a High Nugget Environment
The Bendigo Experience
G F Johansen1
ABSTRACT
In a typical gold deposit each tonne of ore will contain tens of millions
of very fine gold particles, whereas; at Bendigo the majority of gold in
each tonne of ore is present in just five to 15 very coarse gold particles, a
distribution more akin to diamond deposits. At Bendigo the minimum
sample size required for good assay repeatability is estimated at 4.5 t for
each assay. Drill core samples are several orders of magnitude too small
to generate representative assays. The drilling data are characterised by a
nugget effect approaching 90 per cent, a very skewed distribution and, on
average, under-report in situ grade by 50 per cent to 75 per cent.
Since 2002 Bendigo Mining NL has completed 47 000 m of diamond
drilling, 1200 m of on reef development, collected and processed 100
bulk samples (150 t each), 800 mini bulk samples (50 kg each) and
processed in excess of 25 000 t of ore through a pilot plant. The results of
this work have provided the information required to develop and verify a
robust grade estimation procedure.
No standard geostatistical manipulation of the drilling assay data can
generate an accurate grade estimate as the samples and associated assays
are not representative of the ore being sampled. The recognition of this
fact was critical as a radical transformation of the raw drill assay data was
required to develop a workable methodology for grade estimation in such
a high nugget environment. The aim of the transformation was to reduce
the nugget effect and high variance and so approach a lognormal assay
distribution. Ideally the average drill grade would approximate actual in
situ grade and the assay distribution would reflect actual grade
distribution within the reefs.
Following the data transformation simple block modeling within the
defined resources allows grade estimates to be made that can be used in
the estimation of Indicated Resources and hence Probable Reserves. The
grade estimates have been checked by extensive bulk sampling and
processing of 25 000 t of ore (three per cent of total resource) through a
pilot plant with planned full scale mining to provide the ultimate
verification.
INTRODUCTION
The Bendigo Goldfield is centred on the urban area of Bendigo
in Central Victoria 130 km northwest of Melbourne. The
goldfield had a total historical production of 22 million ounces of
gold comprised of 18 million ounces from quartz reefs and four
million ounces from alluvials (see Figure 1).
Gold mineralisation is restricted to mesothermal quartz veins
associated with large scale chevron folding and associated
reverse faulting and is associated with minor amounts of coarse
grained pyrite, arsenopyrite and trace amounts of sphalerite and
galena. The gold mineralisation is unusually coarse grained
(60 per cent of the total metal is present as grains greater than
1 mm in diameter) and is very erratically distributed within the
reefs resulting in an extreme nugget effect approaching 90 per
cent (this is back transformed value estimated using a Gaussian
transformation of current drilling data). The metal distribution
has more affinity to a diamond deposit rather than a typical gold
mine and causes difficulties in the estimation of grade at all
scales.
Bendigo Mining NL (BMNL) has developed a means of
estimating grade in a high nugget environment and recently
released an Indicated Resource estimate of 720 000 t @ 10.0 g/t
Au for 236 000 ounces based on exploration within four reefs
beneath the historical workings.
1. MAusIMM, Exploration Manager, Bendigo Mining Ltd, PO Box
2113, Bendigo Vic 3554. E-mail: gjohansen@bendigomining.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
FIG 1 - Historical production for the Bendigo Goldfield.
THE CHALLENGES OF HIGH-NUGGET SYSTEMS
Mesothermal gold veins are often characterised by a high to
extreme nugget effect (>50 per cent), and the presence of coarse
gold particles (>100 m in size). The Central Victorian Goldfield,
which hosts the Bendigo Field, is well-known as a challenging
environment in which to work. These deposits rank amongst the
most difficult of ore deposits types, in terms of producing an
accurate and precise resource grade estimate. Their effective
sampling is generally difficult because of the relatively low
concentrations involved and erratic nature of the gold particles,
which lead to high sampling errors (Dominy et al, 2000).
Grade continuity is usually less consistent than gross
geological continuity, and is typified by the localisation of grades
within discrete ore shoots. In most cases drilling, particularly
diamond drilling is an effective measure of global geological
continuity, whereas grade can only be reliably resolved from
underground development. Usually only Inferred Mineral
Resources can be estimated from surface drilling alone with
underground development, in-fill drilling and, potentially, bulk
sampling or trial mining required to delineate Indicated and
Measured Mineral Resources and subsequent Ore Reserves.
Grade estimation will have to account for highly skewed and
often multi-modal populations, extreme values, a high nugget
effect and low geostatistical ranges.
HISTORICAL GRADE ESTIMATION AND
GRADE CONTROL
The sampling issues were recognised by the early miners and are
reflected in the almost complete absence of assaying throughout
the 100-year mining history at Bendigo. Grade control was
primarily through visual identification of gold in the
development and production faces supported by occasional bulk
sampling. Production was used as a form of continuous bulk
sampling with each stope or development heading processed
through a separate area of the stamp battery allowing a visual
check on the amount of gold present.
Reserves and resources were not estimated in front of
production as grade could not be easily projected away from
development. Despite this the field was successfully exploited
with mining from hundreds of individual reefs resulting in 18
million ounces of production over a 100-year period.
317
G F JOHANSEN
DEFINING THE NUGGET EFFECT AT BENDIGO
The nugget effect is a quantitative geostatistical term describing
the level of variability between samples in close proximity to
each other. It is a function of both the gold particle size and
particle distribution within the reef. In a typical gold deposit
(low nugget) each tonne of ore will contain tens of millions of
very fine gold particles whereas, at Bendigo the majority of gold
in each tonne of ore is present in just five to 15 very coarse gold
particles, a distribution more akin to diamond deposits.
Recognising the difficulty in sampling and grade estimation at
Bendigo (Johansen, 1997; Franois-Bongaron, 1998; Dominy et
al, 2000, 2001a, 2001b, 2003; Johansen, et al 2003) BMNL
undertook extensive sampling programs at a variety of sample
sizes to understand the nugget effect and develop a grade
estimation methodology. Sample sizes included drill core and
rock chip samples (2 kg), mini bulk samples (50 kg), bulk
samples (100 to 150 t) and trial ore parcels (1000 to 3000 t).
Drill core samples
Sawn half core samples (1 m intervals) weighing 2 kg each are
pulverised and assayed by fire assay (FA) with the entire pulp for
all assays >0.5 g/t Au subject to screened fire assay (SFA).
Indicated Resources are typically drilled out on 40 m sections
with drillholes at 5 m vertical separation. Figure 2 is a lognormal
plot of the 1805 drill core assays within the current Indicated
Resources. The assays exhibit a highly skewed distribution rather
than an expected bell curve (lognormal distribution). The top
three per cent of assays (56 samples above 16 g/t Au) contain
80 per cent of the total metal and 47 per cent of samples within
the resource assay <0.1 g/t Au providing a good indication of the
high nugget effect. The drill samples average 3.58 g/t Au
compared to an estimated resource grade of 10 g/t Au.
Variography (following a Gaussian transformation) indicates a
nugget effect approaching 90 per cent.
Lognormal Grade Distribution Plot
(in resource drilling)
50%
45%
40%
35%
% of Assays
30%
25%
20%
15%
10%
5%
0%
0.1
0.2
0.3
0.5
10
20
Grade (g/t Au)
FIG 2 - Lognormal plot of the 1805 drill core assays within the
current Indicated Resources.
The highly skewed distribution was expected as earlier
homogeneity test work using 2 kg samples generated a similar
distribution pattern and indicated a gold particle size
approaching 6 mm.
Mini bulk samples
Mini bulk samples are 50 kg samples of broken development ore
screened at -10 mm. The entire sample is crushed to -1 mm using
a rolls crusher and passes through a 7.5 inch Knelson
concentrator. The concentrate is panned and all gold coarser than
300 micron collected and weighed with the remainder of the
concentrate and a tail sample sent for assay. This allows the
318
estimation of total grade, percentage of gold recoverable by
simple gravity separation and percentage of total metal in
particles coarser than 300 micron. Two mini bulk samples are
collected from each development round.
Figure 3 summarises the data from 190 development rounds
(380 mini bulk samples). Gravity recoveries are between 75 per
cent and 85 per cent with recovery increasing gradually as grade
increases. Similarly the percentage of gold present as +300
micron grains increases slowly from 60 per cent to 75 per cent
with increasing head grade. The sudden drop in gravity recovery
and coarse gold content at low grade is a function of sample size
rather than any change in gold particle distribution. The coarse
gold fraction from the mini bulk samples were screened at 1 mm
and 2 mm to provide further information on gold particle size
distribution.
% to Tail
Gold Distribution in Mini BulkSamples % to gravity
% +300 micron
100
80
% Distribution
60
40
20
0
0 20 40 60 80 100 120 140 160
Sample Grade Range (g/t Au)
FIG 3 - Gold distribution in mini bulk samples.
Variography on the mini bulk samples indicates a nugget effect
of <10 per cent. The samples tend to overestimate grade as they
only represent the fine fraction of broken ore with gold
preferentially reporting to the fine fraction following blasting due
to the brittle nature of the host quartz.
Bendigo Raw Drilling Data
Bulk samples
Bulk samples, typically a whole development round weighing
approximately 150 t, are crushed to -10 mm with a statistically
valid sub-sample weighing approximately 14 t cut from the
sample using a mechanical cutter. The sub-sample is riffle split to
two 7 t duplicate samples. Each duplicate is reduced to -1 mm in a
hammer mill before passing through a gravity circuit consisting of
two Gekko spinners. A coarse gold trap within the hammer mill
circuit recovers all gold greater than 1 mm in diameter which is
then weighed. The gravity concentrate and a tail sample are sent
1000
30
50
100
200
300
500
for SFA. This allows the estimation of total grade, percentage of
gold recoverable by simple gravity separation and percentage of
total metal in particles coarser than 1 mm.
Figure 4 summarises the data from 63 bulk samples within the
current resources. Gravity recovery varies between 65 per cent
and 85 per cent with recovery increasing gradually as grade
increases. Similarly the percentage of gold present as +1 mm
grains increases slowly from 35 per cent to 55 per cent with
increasing head grade. Gold particle size distribution data from
the bulk sample circuit was considered less reliable than the mini
bulk samples as the hammer mill was very abrasive and
substantially changed the morphology of the gold particles.
Trial ore parcels
Trial ore parcels varied between 750 and 3000 t each and after
crushing to -10 mm were processed through a circuit comprised
of a ball mill (9 t/hour) generating a product with P80 of 450
Adelaide, SA, 19 - 22 September 2004 PACRIM 2004
 A SOLUTION TO GRADE ESTIMATION IN A HIGH NUGGET ENVIRONMENT THE BENDIGO EXPERIENCE

% Total Gravity TABLE 1

Gold Distribution in Bulk Samples % Tail Table of theoretical gold particle size distribution and number of
% +1mm gold particles present in a typical tonne of ore grading 10 g/t Au.
100
Size fraction Per cent of total No of particles
90 (mm) metal present in one tonne
80
<0.03 3 600 000
70
% Distribution

60
0.03 to 0.05 4 220 000
50 0.05 to 0.1 5 21 000
40 0.1 to 0.3 8 5000
30 0.3 to 0.5 10 420
20
0.5 to 1 10 53
10
0 1 to 2 20 13
2 4 6 10 15 20 >20 2 to 3 15 3
Sample Grade Range (g/t Au) 3 to 5 10 0.4

FIG 4 - Gold distribution in bulk samples. 5 to 10 10 0.078

micron before passing through a gravity circuit of jigs and a
Gekko spinner and a tail sampler. The gravity concentrates were
cleaned with tables and spirals with the final concentrate leached
to recover the contained gold. This provided valuable data for
grade reconciliation but little additional information on gold
particle size.
GOLD PARTICLE SIZE AT BENDIGO
Based on the information generated from this sampling it was
possible to accurately estimate the gold particle size distribution
for Bendigo (see Figure 5) for typical ore.
Gold Particle Size Distribution
at Bendigo
5 thus more representative. This would result in:
4.5
4 reduced nugget effect and variance so assays approach a log
Gold Particle Size (mm)
>10 5 0.016
within the resources are <1.0 g/t Au), a small percentage have
extremely high grades and relatively few assays reflect the actual
grade of mineralisation.
As the drill core samples are so unrepresentative no standard
geostatistical manipulation of the assays will generate a grade
estimate approaching the actual grade or a reasonable estimation
of grade distribution within the resource. This realisation opened
the way to a more radical geostatistical approach.
OVERCOMING THE NUGGET EFFECT AT
BENDIGO
Bendigo data
The goal of any statistical manipulation at Bendigo is to make
"Typical" gold deposit the samples (drill core) behave as though they were larger and
average drill grade approximating global in situ grade;

3.5 normal distribution; and

3 assay grade distribution approximating actual grade
2.5 distribution.
2

1.5
1 There are three characteristics of Bendigo mineralisation that
0.5
0 1.
0 10 20 30 40 50 60 70
Cumulative Metal %
FIG 5 - Graph of gold particle size distribution for Bendigo.
Knowing the percentage of gold present for each particle size
it is a simple matter to estimate the theoretical number of gold
particles present for each size fraction in a tonne of ore grading
10 g/t Au (see Table 1). The distribution of metal below 300
micron is extrapolated from known data.
At Bendigo 75 per cent of the total gold present is as particles
coarser than 300 micron with more than 60 per cent of gold in
particles larger than 1 mm. The 60 per cent of gold, present as
particles >1 mm, occurs as just 16.5 particles within a tonne of
10 g/t Au ore. This distribution requires a minimum sample size
(on crushed ore) of 4.5 t to give good assay repeatability.
Drill core samples (2 kg) are several orders of magnitude too
small to produce representative assays resulting in a nugget
effect approaching 90 per cent, significant underestimation of in
situ grade and have a very skewed distribution with the vast
majority of assays being very low grade (83 per cent of assays
Key characteristics
were critical to overcoming the extreme nugget effect:
As grade increases the relative abundance of coarse and
80 90 100 fine gold particles stays much the same. Figure 6 shows a
simple linear relationship between head grade and tail
grade (essentially gold particles <150 micron) for the mini
bulk samples. Each 0.5 g/t Au increase in tail grade equates
to an additional 850 000 fine gold particles per tonne. This
means as grade increases the number of fine gold particles
increases rapidly and is reflected in an increased number of
>0.1 g/t Au values (elevated background).
2. In general an increase in quartz content and/or arsenic
values reflects an increase in gold grade. At Bendigo gold
particles are restricted to quartz veins; therefore an
increased quartz content increases the opportunity for gold
to be present. Similarly economic gold mineralisation is
restricted to reefs containing arsenopyrite though the
presence of arsenopyrite in a reef does not necessarily
indicate the presence of economic mineralisation.
Statistical analysis of the entire drilling database (27 000
assays) confirms the observed relationships between gold,
quartz and arsenic (see Figures 7 and 8) despite the high
nugget effect.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 319

G F JOHANSEN

Tail grade vs Head Grade 3. The gold particle distribution at Bendigo has affinities with
that of diamonds in diamond deposits. Recognition that the
30.00 coarse average gold particle size and very limited fine gold
fraction at Bendigo produce a distribution closer to
25.00 Mini bulk diamond deposits (Rombouts, 1995) rather than most gold
samples deposits was critical as the methodology used to overcome
Tail Grade (g/t Au)

20.00
the nugget effect has a basis in the statistical treatment of
diamond deposits.
15.00
Transformation of raw drill assay data
10.00 Recognising that no standard geostatstical manipulation of the
raw drill assay data would generate a grade estimate approaching
5.00 actual grade opened the possibility that a radical statistical
treatment may produce a better estimate. Based on the identified
0.00 key characteristics of the Bendigo gold mineralisation a
0 20 40 60 80 100 120 140 160 transformation was applied to the raw drilling data involving
Head Grade (g/t Au) three steps:
1. Apply a top cut at the 92nd percentile. This equates to a top
FIG 6 - Plot of tail grade versus head grade for mini bulk samples. cut at 4 g/t Au and is designed to reduce the high nugget
and high variance in the raw data. It also has the effect of
reducing the average grade to <1 g/t Au.
Log-Log plot of Gold Assay vs Quartz % 2. Multiply all assays by 11. This enhances the more
(27,000 assays) representative but weak, elevated background improving
the metal distribution and brings the grade back towards
100
actual in situ grade.
3. Add 0.2 g/t Au for each ten per cent quartz and 1000 ppm
As. This adds an average of 1.2 g/t Au to the overall grade
10 but more importantly operates as a pseudo background
Quartz Percent

(accounts for mineralisation seen in bulk samples but is too

disperse to be detected by drilling) further reducing the
1 assay variance and nugget.
This transformation reduces the nugget effect to 70 per cent
0.1
and significantly reduces the variance producing a
pseudo-lognormal distribution (see Figure 9).

0.01
0.001 0.01 0.1 1 10 100 1000 10000
Log-Normal Grade Distribution Plot
Gold Assay (g/t Au)
(in resource drilling)
50%
FIG 7 - Plot of gold assay versus quartz per cent. Note the strong
trend despite the 90 per cent nugget effect. Bendigo Raw Drilling Data
40%
Bendigo Transformed (4*11)
% of Assays

30%
Log-Log plot of Gold Assay vs Arsenic Assay
(27,000 assays) 20%

1000000
10%
100000

10000 0%
Arsenic Assay

10

20

30

50

200

300

500

1000
0.1

0.2

0.3

0.5

100

1000
Grade (g/t Au)
100

10 FIG 9 - Lognormal plot of raw drill core assays and transformed

drill core assays from within the current Indicated Resources. The
1 transformed assays form a bell curve typical of a lognormal
distribution.
0.1
0.001 0.01 0.1 1 10 100 1000 10000
Gold Assay (g/t Au) GRADE ESTIMATION AT BENDIGO
A grade for each reef is estimated by block modelling the
FIG 8 - Plot of gold assay versus arsenic assay. Note the strong transformed drill data within the resource shape. Block
trend despite the 90 per cent nugget effect. modelling involves the following parameters:

320 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A SOLUTION TO GRADE ESTIMATION IN A HIGH NUGGET ENVIRONMENT THE BENDIGO EXPERIENCE

Very small blocks are used (1 m3 or 2 m3 blocks). This is Block Model Vs Development Grade
primarily to provide good visual grade definition on screen (for all in resource development)
and in hard copy plots. Testing using much larger blocks had 25
negligible impact on grade estimates.

Block Model Grade (g/t)

A 50 m 10 m 10 m search ellipse orientated along the reef. 20

Assay point selection is anisotropic matching the search

ellipse. 15

Given the still very high nugget effect, grade for each block
is the simple average of the nearest 25 assay points. 10

This provides a global grade for each reef and an indication of

5
grade distribution within the reef.

Calibration of block model grade estimates for 0

0 5 10 15 20 25
individual reefs
Development Grade (g/t)
There is some variability in gold distribution and gold particle
size between reefs. Currently each reef converted to Indicated FIG 10 - Comparison of actual grade of development subdivisions
Resource has extensive on reef development with associated bulk and grade for identical volumes derived from block modelling.
sampling and mini bulk sampling. The grade for the entire
development and subdivisions of development based on grade or
geographic distribution can be accurately estimated for each reef TABLE 3
from the bulk sample and mini bulk sample results. Comparison of grades derived from bulk sampling, block
Grade estimates for all development and the various modelling and processing through the New Moon mill.
subdivisions of development can also be estimated from the
block modelling of transformed drilling. By making minor Development Development
adjustments to the drill transformation parameters (top cut and/or reconciled with reconciled
block model with mill
multiplier) a best correlation between the two sets of grade
estimates is achieved for each reef. Table 2 give the adjusted Tonnage (t) 27 200 23 500
parameters providing the best match for each of the four reefs in Geology estimate (g/t Au) 12.9 9.4
the current resource estimate. Block model estimate (g/t Au) 12.7 -
Mill estimate (g/t Au) - 8.9
TABLE 2 Variance (%) 2 5
Transformation parameters that produce the best correlation
between block model and development grades estimates for each Based on bulk sample and mini bulk sample results.
reef. Based on transformed drilling.
Reef Topcut Multiplier
Christine 2 5 14.5
Garrard 5 13.5 CONCLUSIONS
Shywolup 4 11 The gold particle size and distribution at Bendigo are the causes
Upper S3 4.5 11 of the high nugget effect and make any standard geostatistical
treatment of drill core assays inappropriate for grade estimation.
Based on a detailed understanding of the gold particle size and
Figure 10 shows the comparison between development grade distribution at Bendigo the company applies a radical
and final block model grades for all development subdivisions transformation to the raw drilling data before block modelling
within the current Indicated Resources. The block modelling of and grade estimation. Extensive bulk sampling and processing of
transformed drill data is able to identify high- and low-grade over 25 000 t through a pilot plant have confirmed the efficiency
areas within the on reef development confirming that the of the transformation.
distribution of the small amount of fine, background gold reflects The grade estimation methodology outlined provides a
the overall grade distribution within the reefs. transparent, logical and auditable path to estimating grade,
attributes consistent with the JORC Code. The accuracy of the
Confirmation of grade estimation grade estimate is considered to be appropriate for the
classification of Indicated Resources.
As a final confirmation of the grade estimation procedure in Grade estimation at other gold deposits containing significant
excess of 25 000 t of reef material was processed through a pilot amounts of coarse gold may benefit from the application of a
plant (New Moon Mill) as individual trial ore parcels varying variant of the Bendigo transformation.
between 750 and 3000 t each. It was possible to compare the
grade from processing some 23 500 t against grade estimates
derived from bulk sampling and mini bulk sampling. Similarly
ACKNOWLEDGEMENTS
grades for 27 200 t of ore within the defined resources were The author would like to thank Bendigo Mining NL for
derived from bulk sampling and from block modelling permission to publish this paper and to exploration staff members
transformed drilling data as part of the calibration process. As Paul Quigley and Rowland Hill for there input. Dominique
can be seen from Table 3 there is reasonable agreement in the Franois-Bongaron and Daniel Guibal reviewed aspects of
grade estimates providing further support for the radical sampling and statistics treatment at Bendigo. Simon Dominy is
transformation of the raw drilling data. thanked for reviewing the paper.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 321

G F JOHANSEN
REFERENCES
Dominy, S C, Johansen G F, Annels, A E and Cuffley, B W, 2000.
General considerations of sampling and assaying in a coarse gold
environment, Trans Inst Min Metall (Section B), 109:145-167.
Dominy, S C, Johansen, G F and Annels, A E, 2001a. Bulk sampling as a
tool for the grade evaluation of gold-quartz reefs, Trans Inst Min
Metall (Section B), 110:176-191.
Dominy, S C, Johansen, G F, Cuffley, B W, Platten, I M and Annels, A E,
2001b. Estimation and reporting of mineral resources for coarse gold
bearing veins, Explor Min Geol, 9(1):13-42.
Dominy, S C, Annels, A E, Platten, I M and Raine, M D, 2003. A review
of problems and challenges in the resource estimation of high
nugget-effect lode-gold deposits, in Proceedings Fifth International
Mining Geology Conference, pp 279-298 (The Australasian Institute
of Mining and Metallurgy: Melbourne).
322 Adelaide, SA, 19 - 22 September 2004
Franois-Bongaron, D, 1998. Extensions to the demonstration of Gys
formula, Explor Min Geol, 7(1, 2):149-154.
Johansen, G F, 1997. Sampling, assaying and reporting in a coarse gold
environment, in Assaying and Reporting Standards Conference
(Singapore) notes, November (AIC Conferences).
Johansen, G F, Raine, M D, Dominy, S C and Bartlett, J K, 2003.
Challenges of sampling extreme nugget-effect gold-quartz reefs at
the New Bendigo Project, Central Victoria, Australia, in Proceedings
Fifth International Mining Geology Conference, pp 299-310 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
Rombouts, L, 1995. Sampling and statistical evaluation of diamond
deposits, Journal of Geochemical Exploration, 53:351-367.
PACRIM 2004
A Review of the Kapit Orebody, Lihir Island Group, Papua New
Guinea
R P Kidd1 and J R Robinson1
ABSTRACT
Prior to mine development, exploration drilling within the Luise Caldera
on Niolam Island encountered gold mineralisation to the northwest of the
main Minifie and Lienetz orebodies. This mineralisation was referred to
as the Kapit orebody, and ten diamond drill holes were subsequently
completed over the area to investigate its economic potential. Several
significant mineralised intercepts were returned. These were interpreted
using the orebody model developed for Lihir at that time which invoked a
strong horizontal control to mineralisation. Additional consideration was
given to the elevated geothermal temperatures encountered, the proximity
of the mineralisation to the ocean, and its perceived depth below sea
level, resulting in the Kapit area being given a low priority for follow up
investigation. As a result, no further work was completed over Kapit until
recently.
The Kapit orebody was the target of an intense diamond drilling
programme during 2002/2003. Data from this latest round of drilling has
significantly improved understanding of the mineralisation and associated
geology. The significant orebody now delineated at Kapit substantially
differs from past interpretations, and the latest data has upgraded its
economic potential.
A resource estimate and other critical studies completed in 2003
enabled an updated reserve statement to be issued in February 2004
including the Kapit mineralisation for the first time. Total Probable
Reserves at Kapit stand at 25.2 million tonnes (Mt) at 5.28 grams of gold
per tonne (g Au/t), for a total of 4.3 million ounces (Moz) of contained
gold.
REGIONAL GEOLOGY
The Lihir Island Group is part of the Tabar-Lihir-Tanga-Feni
island chain, located approximately 50 km north east of the
island of New Ireland, Papua New Guinea, roughly parallel to its
1. Lihir Management Company, GPO Box 905, Brisbane Qld 4001.
FIG 1 - Location of Lihir Island Group.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
northern coastline (Figure 1). The islands in the chain rise
sharply from depths of ~2000 m below sea level, located on
submarine ridges with north northeast trends bounded by steep
normal faults (Exon et al, 1986). The Lihir group, along with the
other islands of the Tabar-to-Feni chain, were formed essentially
by alkaline, silica-undersaturated, high potassic volcanism. The
volcanics display a varied but predominantly shoshonitic
composition, lacking normative andesite, and this differs strongly
from the calc-alkaline arc volcanism of similar age on the
adjacent islands of Bougainville and New Ireland (Patterson et
al, 1997). Figure 2 illustrates the major present day structural and
tectonic features of the New Ireland region.
Carbon-14, K-Ar, and Ar-Ar dating suggest an age progression
in volcanism from the north west to the south east of the
Tabar-to-Feni island chain. Lavas from Simberi in the Tabar
island group have been dated at 3.7 (0.2) to 1.9 (0.1) Ma
(Johnson, 1979, McInnes and Cameron, 1994), whilst trachytic
tuff-breccias from Ambitle Island in the Feni group have been
dated at 2300 (100) years (Licence et al, 1987). Present day
geothermal activity occurs within all four groups of the chain,
and significant epithermal gold mineralisation has been
discovered in the Tabar and Feni island groups, as well as in the
Lihir group.
Niolam Island (commonly referred to as Lihir Island) is the
largest island of the Lihir Group. It consists of five
Miocene-Pleistocene volcanic units of which three are
recognisable volcanic craters (including the Luise Caldera which
hosts the Ladolam gold deposit), and two of which are sequences
of mafic volcanic rock that predate the three volcanoes (Figure
3). The volcano which formed the Luise Caldera was developed
less than one million years ago (Davies and Ballantyne, 1987).
323
R P KIDD and J R ROBINSON

FIG 2 - Regional map of Papua New Guinea and the major structural and tectonic features (Carman, 1994).

FIG 4 - Shaded relief and bathymetry image, Niolam Island, Lihir

FIG 3 - Simplified geological map of Niolam Island, Lihir Island
Group. Main volcanic untis are highlighted. The pale unti fringing
the island is raised coral reef.
The Luise Caldera is a well defined, elliptically shaped
caldera, rising steeply to several peaks over 600 m above sea
level and is open to the northeast where it is breached by the sea
to form Luise Harbour. It is postulated to have formed by the
failure of the volcanic edifice in a Mt St Helens style eruption,
Island Group. Arrow indicates the seafloor debris flow formed by
the collapse of the Luise volcanic edifice.
with the seaward sector of the volcanic cone collapsing into
Luise Harbour (Figure 4). Remnants of the resulting debris flows
have been documented on the adjacent seafloor by scientific
cruise SO-133 of R/V Sonne (Herzig et al, 1999). The currently
defined area hosting the gold mineralisation of the Ladolam
Deposit occupies an area of about 2.0 by 1.5 km towards the
center of the caldera, which measures approximately 5.5 km by
3.5 km in total area.

324 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A REVIEW OF THE KAPIT OREBODY, LIHIR ISLAND GROUP, PAPUA NEW GUINEA

DEPOSIT GEOLOGY between trachybasaltic, latitic, andesitic, and trachytic

The gold deposits in the Luise Caldera are hosted within
volcanics, intrusives, and breccias which have undergone
extensive alteration. Two major alteration episodes have been
identified; an earlier and deeper porphyry style event resulting
in potassic alteration grading laterally into propylitic alteration,
and a later and higher level epithermal event producing argillic,
advanced argillic, phyllic, and lower temperature potassic
alteration. This intensive alteration has destroyed much of the
original host rock lithologies, and due to this an ore type
classification has been developed based essentially upon various
combinations of alteration, hardness, the degree of brecciation
and/or leaching of matrix material, and the presence of late stage
anhydrite veining. Whilst this is more a metallurgical
classification than a geologic one, it has proved useful in
determining many of the mining and processing characteristics
of the orebody and the host rocks. The ore types are roughly
subhorizontal in occurrence and form a fairly consistent vertical
sequence of clay-rich rock, grading into white mica-feldspar
rock, then feldspar-biotite and, at depth, into feldspar-biotite-
anhydrite rock. Within and at the boundaries of the ore types,
geological structure is also a major influence on the localisation
of higher gold grades in the orebody.
Original host rock lithologies
The original lithologies occurring within the Luise Caldera have
been documented in detail by Carman, 1994, Wallace et al, 1983,
Moyle et al, 1990, and Corbett et al, 2001, as summarised in
Table 1.
Forming the basement within the caldera are a series of
coherent mafic and minor intermediate lava flows that range
compositions (Cater, 2002). These flows have associated
interlayered deposits of epiclastic and pyroclastic material of
similar compositions. These basement volcanics are crosscut by
alkaline intrusives consisting predominantly of equigranular
biotite monzonite, but including equigranular to porphyritic
leucogabbro, monzonite, pyroxene diroite, and biotite syenite
(Davies and Ballantyne, 1987; Moyle et al, 1990; Carman, 1994).
The Ladolam breccia complex is developed in the central part
of the caldera, where there exists a complex series of breccia
units overlapping and grading into each other. The intricate
inter-relationships between units within this breccia complex
have been documented by Corbett et al, 2001. Slab, agglomerate,
pyrite injection, matrix supported, fluidised, crackle, and matrix
milled breccias are variously described, each having an
interrelated and/or overprinting relationship between each other.
Alteration
Intense alteration was intimately associated with ore forming
events. Early stage potassic alteration occurred as porphyry style
alteration associated with the emplacement of alkalic stocks
within the volcanic edifice, with peripheral and broadly
contemporaneous propylitic alteration. Sudden collapse of the
volcanic edifice is interpreted to have resulted in the rapid
depressurising of the system and subsequent telescoping of
epithermal alteration and associated gold mineralisation upon the
porphyry environment (Corbett et al, 2001). Argillic and
advanced argillic alteration assemblages developed through
continued geothermal activity, driven by post mineralisation
magmatism. Geothermal activity continues to this day. A
summary of alteration assemblages occurring within the Luise
Caldera is presented in Table 2.

TABLE 1
Summary of host rocks (modified from Cater, 2002).

Lithofacies Compositions Mineralogy Origin Age

Mafic Coherent lava (ankaramite, Phenocrysts of plagioclase, clinopyroxene, Subaqueous to subaerial mafic Plio-Pleistocene
Volcanic alkali basalt, trachybasalt, magnetite, amphibole and biotite. Groundmass intermediate eruptions
(volcanic trachyandesite) trachytic plagioclase with minor clinopyroxene,
basement) orthoclase, magnetite
Epiclastic facies Basement fragments Erosion of volcanic basement Post Pliocene
(sandstones and breccias)
Alkaline Coherent leucogabbro- Monzonite and porphyritic intrusives: Subvolcanic stocks Mid to late Pleistocene
Intrusives monzonite-diorite, syenite, plagioclase, biotite, clinopyroxene, interstitial (Plagioclase porphyry
intermediate porphyry, orthoclase, minor magnetite and apatite dikes post date intrusive
plagioclase porphyry Intermediate porphyry: plagioclase and stocks)
clinopyroxene (crystal rich and sugary textured)
Plagioclase porphyry: plagioclase (20 - 30 per
cent) with minor clinopyroxene, magnetite and
amphibole. Groundmass of orthoclase and
plagioclase.
Ladolam Volcanic breccia Fragments of mafic-volcanic. Matrix formed by Diatreme breccia Pre-ore
Breccia sand-to silt-sized lithic fragments
Complex Porphyry breccia Fragments of equigranular to porphyritic alkaline Intrusive breccia Pre-ore
intrusives. Matrix-formed by rock flour.
Minifie ore breccia Fragments of volcanic, intrusives, breccias, Near-surface explosive breccia Late Pleistocene
epilastics, tuff. Groundmass of rock flour and (pre to syn-ore)
hydrothermal K-feldspar and pyrite.
Silicic breccia Fragments of intrusives, porphyries and recycled Dissolution of earlier breccia?
hydrothermal minerals. Groundmass of quartz +
K-feldspar + sulfide.
Localised breccias Crackle, anhydrite sealed; Mill, rounded and Various formation processes
calcite sealed; Vein, sealed by calcite quartz; related to hydrothermal activity
Cavity filling, anhydrite cavity filling.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 325

R P KIDD and J R ROBINSON

Alteration
Potassic
High temperature propylitic
Low temperature propylitic
Potassic (2)
Silicic-intermediate argillic
Argillic-advanced argillic
TABLE 2
Summary of alteration assemblages (modified from Carman, 1994).
Assemblage Distribution
Biotite-orthoclase anhydrite albite magnetite Alkaline intrusives at Lienetz and Minifie deposits
tremolite pyrite chalcopyrite
Orthoclase-anhydrite-pyrite calcite chlorite rutile Overprints secondary biotite alteration at Lienetz , Minifie
magnetite pyrrhotite phlogopite and Kapit deposits
Chlorite-actinolite-calcite-anhydrite-epidote-albite-K-felds Weakly fractured lavas between southern Lietnetz and
par biotite pyrite rutile pectolite sphene Minifie deposits
Calcite-chlorite-adularia-albite-K-mica-pyrite-rutile Volcanic breccia (diatreme) lithofacies and volcanic
quartz epidote basement, envelops the Minifie, Lienetz and Kapit deposits
Adularia-pyrite-marcasite-rutile illite calcite Most intense in shallow breccia-hosted ores at Minifie,
anhydrite apatite quartz vermiculite albite close spatial association with Lietnez and Kapit ores
Quartz-mixed layer clay K-feldspar anhydrite pyrite Associated with the quartz, stockwork zone at Minifie,
overprints relatively deep breccia-hosted Minifie ore
Kaolinite alunite smectite pyrite marcasite Near-surface alteration, most intense subjacent to modern
enargite sulfur day hotsprings, overprints breccia-hosted near-surface lodes

Structure
Several structural trends appear important in localising and
confining individual breccia units as well as gold mineralisation.
Carman, 1994, recognised that the predominant regional
orientation of dykes and faults on Niolam Island are NNE
trending (~025), which he interpreted to be associated with
deep-seated tensional faults developed during translocation of the
islands away from the New Britain Subduction Zone, and which
may have controlled the long axis of the Luise volcanic edifice.
Other strong structural trends occurring within the caldera are
noted below:
ENE structures dipping moderately (60) to the north;
arcuate generally EW trending, north dipping, listric shaped
structures (believed to be associated with the collapse of the
volcanic edifice);
subvertical NW trending structures; and
steeply eastward dipping NS structures.
Mineralisation is localised along all structural trends noted
above, with intersections of structures creating a focus for major
ore deposition as evidenced by assay data returned from blast
hole samples (Corbett et al, 2001).

Mineralisation
Exploration work since 1983 has defined several adjacent and
partly overlapping orebodies in the Luise caldera, the principal
ones being Lienetz, Minifie, Kapit, Coastal, and Borefields.
Figure 5 is a plan view of the mine area showing their location.
Gold appears to be the only metal of economic significance
present within the Luise Caldera. Corbett, 2001 recognised a
number of mineralisation styles, ranging from early porphyry to
late stage epithermal mineralisation (Table 3). Two of these
represent economically significant phases of gold mineralisation FIG 5 - Plan showing the major orebodies of the Ladolam deposit.
Dashed lines indicate structural trends displaying strong controls
at Lihir. The most important is refractory K feldspar-sulfide
to high-grade mineralisation. Design pit is January 2004 final open
mineralisation. In this association, gold occurs primarily as
pit design.
submicron size particles in sulfide minerals. Overall sulfide
content is relatively high, with the average sulfur grade of the
reserves being above six per cent. The main sulfide mineral is localised portions of Minifie) this style of mineralisation has
pyrite, with the marcasite form present as an accessory mineral been associated with strong leaching of the original lithologies
and rare arsenopyrite. Gold also occurs as small (less than 100 creating pinhole to open, vughy textures. Cavities of up to 10 m
micron) blebs within fine pyrite crystals. The sulfides are in extent have been encountered in Lienetz. This secondary
characterised by their fine grained nature, and have been porosity is thought to have been the result of dissolution of host
deposited through wholesale flooding and deposition within all rock by hot alkaline fluids (Corbett, 1998), and previous workers
host rocks, imparting a sooty, dark grey colouring to the host have interpreted these zones as being the result of boiling
rocks. Within the Lienetz and Kapit orebodies (and in some (Davies and Ballantyne, 1987).

326 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A REVIEW OF THE KAPIT OREBODY, LIHIR ISLAND GROUP, PAPUA NEW GUINEA

TABLE 3
Ladolam Deposit gold mineralisation styles (documented by Corbett, 2001).

Style Characteristics Au Au:S

1A - Fracture controlled porphyry Anhydrite veins with silica-sericite halos and minor pyrite 1.5 - 3.5 2:1
1B - Disseminated porphyry Silicified intrusion with disseminated pyrite-marcasite ?2-3 ? 1.5:1
1C - Anhydrite seal Complex breccias with several generations of magnetite-orthoclase-biotite-anhydrite 1-3
alteration
2A - Lithologically controlled K Arsenian pyrite flood or line open space created by K feldspar leaching or anhydrite 2-5 1:1
feldspar-sulfide (leached and removal
pin hole textures)
2B - Structurally controlled K Sulfide flooding in association with fluidised breccias 8 - 12 1:1
feldspar-sulfide (Fbx)
2C - Pyrite-marcasite opal Fracture or breccia related pyrite-marcasite and opal/chalcedony 3 - 15 1:1
5:1 in high
Au
3 - Carbonate base metal Early quartz with base metal sulfides and late carbonate in vein breccias 10 - 20 10:1
4 - Quartz chlorite bladed anhydrite Bonanza gold with quartz chlorite lining fractures and bladed anhydrite Bonanza >10:1
5 - Prograde feeder structures Laminated anhydrite locally with sulfides >10 >10:1

FIG 6 - Genetic model for gold deposition within the Ladolam deposit. (Hunt, 2002)

The second significant style of gold mineralisation occurs as a
low sulfidation quartz-chlorite-bladed anhydrite association,
more typical of epithermal style mineralisation, deposited
through the mixing of magmatic fluids with oxidising
near-surface water (Corbett et al, 2001). Free gold up to several
millimetres in size has been observed by the authors associated
with this mineralisation style, though this is rare. Mineralisation
occurs as discrete fracture filled veins through all levels of the
deposits, and is inferred to be an overprinting style of epithermal
mineralisation associated with the cooling of the active
geothermal system within the Luise Caldera. It is also
characterised by relatively high though erratic gold grades when
compared with the high sulfidation style of mineralisation.
Figure 6 presents a model for gold emplacement at Lihir.
Listric faults (developed during the collapse of the Luise
volcanic edifice) acted as conduits for hot magmatic fluids rising
towards the surface. Extensive phreatomagmatic brecciation
which developed after the unloading of the system by edifice
collapse is interpreted to have provided sites of increased
permeability and fluid mixing (Corbett et al, 2001). High
sulfidation refractory gold deposition subsequently occurred
along these feeder structures and closely linked breccia units.
Clay rich, argillically altered material acted as a cap to the
system, with gold deposition penetrating into this overlying
material constrained to strongly developed structures only. As the
system cooled, low sulfidation, epithermal style gold
mineralisation developed as discrete, generally narrow (<10 cm),
structurally controlled mineralised veins or vein sets, closely
associated with the listric feeder zones of the earlier high
sulfidation mineralisation.
THE KAPIT OREBODY
Exploration history
The Kapit orebody lies between the Lienetz orebody and Luise
Harbour; approximately 500 m due north of the western limit of
Lienetz. It is linked to Lienetz by a subhorizontal zone of low
grade mineralisation (generally <2.0 g Au/t) which reaches a
thickness of up to 100 m (Figure 5).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 327

R P KIDD and J R ROBINSON
Soil sampling completed over the Kapit orebody failed to
delineate any significant gold anomalies. In contrast a
biogeochemical survey of gold and arsenic in bark, completed in
1990 over the Luise Caldera indicated a 750 m by 100 m
gold-arsenic anomaly over the Kapit area (Kennecott, 1992).
Significantly, the highest arsenic value of the survey was
returned from the Kapit anomaly, where it was surrounded by
other high values.
Prior to 2002, three diamond drill holes intersected the
present-day Kapit orebody. Of these, the best mineralised
intercept was returned from hole DDHL177. This hole was
drilled vertically, and encountered a mineralised interval of
143 m @ 5.11 g Au/t beginning at 262 m down hole. Whilst the
potential of the Kapit area to host significant mineralisation was
recognised from the drilling results, the widely accepted orebody
model for Ladolam at the time invoked a strong horizontal
control to all orebody interpretations. The mineralisation was
thus viewed as having too much overburden, and additionally
was considered to be too close to Luise harbour to allow for open
pit development. The area also has a co-incident, generally north
south corridor of surface geothermal activity, and it was thought
that the deep mineralisation in DDHL177 may be too hot to
enable extraction. The Kapit area was thus given a low priority
for follow-up investigation.
Diamond drilling recommenced in the Kapit area in 2002,
following on from a successful drilling program targeting
extensions to the Lienetz orebody, with 14 holes completed for a
total meterage of 6519 metres. Results from this program showed
that the Kapit orebody was of higher grade than what was
previously indicated, and that it appeared to be plunging steeply
northwards from quite shallow depths below the original surface.
Significantly, the orebody remained open to the east, north, and
west. Interpretation of this new data resulted in a re-assessment
of the orebody, and an upgrading of its resource potential.
In 2003 an intensified drilling effort followed, with a further
52 holes completed for a total meterage of 21 809 metres. Results
from this program provided a greater understanding of the nature
of the mineralisation, and allowed a detailed assessment of
resource inventory. Geotechnical, geothermal, and dewatering
studies were also completed during 2003 over the area, allowing
reserve estimates to be compiled on the Kapit orebody for the first
time.
Host lithologies
The lithologies of the Kapit area are identical to those occurring
elsewhere within the Luise Caldera and consist of fragmented
lithologies (pyroclastics), volcanic breccias (heterolithic
breccias), intrusives, and hydrothermal breccias. Fragmental
volcanics form the bulk of the original near surface lithologies,
accompanied by underlying intrusives. Hydrothermal breccias of
various types are associated with the emplacement of the
intrusives, and are predominantly localised and controlled by
major structural features.
Fragmented lithologies
These include unconsolidated surface colluvium material, plus
underlying volcanic breccias, agglomerates, and tuffaceous units.
Tuffaceous units include ash tuffs, lapilli tuffs, and
accretionary lapilli tuffs. Ash tuffs are composed of lithified,
fine-grained (<1 - 2 mm) fragments. Lapilli tuffs consist of
polymictic clasts ranging is size from 2 to 70 mm which are
supported by a fine-grained lithic matrix. Accretionary lapilli
tuffs are characterised by spheroidal accretionary structures
disseminated through fine-grained matrix material. Individual
accretions consist of an inner core of altered, fine-grained, lithic
grains, surrounded by a coating of clay. Accretionary lapilli tuffs
are inferred to be formed either by accretion of dust particles
328 Adelaide, SA, 19 - 22 September 2004
onto small fragments of rock, by water droplets which condense
due to adiabatic cooling of expanding eruption-clouds filled with
debris, or by flushing of an ash cloud by rain (Carman, 1994).
The volcanic breccias consist of heterolithic, angular to
subrounded, matrix supported clasts, ranging from less than
10 mm to 60 mm in diameter. The matrix consists of altered
fine-grained lithic material.
The agglomerates contain clasts of different lithological types,
supported by finer grained material. They are interpreted to have
provided important primary permeability for hydrothermal
fluids, resulting in alteration and associated mineralisation.
Agglomerates are distinguished from breccias by the component
clasts displaying fuzzy rather than distinct outlines.
Intrusives
The intrusives found at Kapit are of alkaline character as outlined
above. They occur as large undisrupted blocks, as dykes, or as
fragments within various breccias. In hand specimen the
intrusives appear medium grey-pale green in colour, and have
fine to medium grained, feldspar-biotite-mafics occurring in a
fine groundmass infilled with late stage anhydrite. The intrusives
may also form monolithic hydrothermal or crackle breccias, and
intrusive fragments also commonly occur within heterolithic,
matrix supported breccias.
Hyrdrothermal breccias
A diverse array of interrelated breccia types are recognised at
Kapit. These breccias occur within a funnel shaped breccia
complex which tapers and terminates at depth, occupying a
pronounced, generally symmetrical depression within the
underlying propylitically altered basement lithologies. The
depression tapers downwards to terminate at 450 m below sea
level, and displays a distinct U shape in cross-section (Figure
7). It is postulated that the Kapit breccia complex is diatreme
related, however further more detailed studies are required for
confirmation.
These brecciated rocks form an integral part of the mineralised
system at Kapit, forming units of greater porosity and
permeability to provide conduits for migrating mineralising
fluids. Breccia types are identified using fragment type, matrix
composition, fragment-matrix interactions, as well as their
overall relationship with the surrounding host rocks. The three
different breccia types recognised in the Kapit area are as
follows.
Crackle breccias occur generally towards the edges of the
Kapit breccia complex, restricted to more competent basement
units. They comprise angular clasts separated by random arrays
of fractures, which have been infilled by cemented rock flour and
pyrite as a result of fluidised injection. Some crackle breccias
display a more regular, subparallel fracture pattern, and these are
termed sheeted crackle breccias, with an overall stronger
structural control.
Fluidised breccias characteristically have a matrix of
cemented rock flour which is more pyrite rich than in the crackle
breccias. The matrix material is interpreted to have undergone
substantial transport as it regularly displays fluidised textures,
and the clasts are generally sub-rounded as the result of milling.
Additionally, the fluidised breccias commonly display a
dyke-like character, and are recognised cutting other lithological
and breccia units.
Vughy breccia units occur associated and restricted to the
epithermal potassic (adularia) style of alteration, and form the
core of the Kapit breccia complex. The vughy breccias have
undergone extensive leaching, and are characterised by open
vughy textures within the matrix material. Pinhole textures are
also developed within the clasts through leaching of mafics and
feldspar. Vughy breccias are distinctly texturally destructive and
PACRIM 2004
 A REVIEW OF THE KAPIT OREBODY, LIHIR ISLAND GROUP, PAPUA NEW GUINEA
FIG 7 - Schematic section of the Kapit orebody illustrating the breccia hosted mineralised zones.
alteration is generally pervasive. The open-spaced vughs within
the matrix are lined with pyrite, silica, and adularia, and at depth
this breccia unit becomes infilled or sealed by an
anhydrite-carbonate-silica assemblage.
Leaching resulting in the creation of pinhole textures is not
confined to vughy breccias, and also occurs within agglomerate,
brecciated agglomerate, and other breccia units. As with the
vughy breccias, this enhanced permeability within these units is
closely associated with gold mineralisation.
Alteration
Alteration styles present within the Kapit area are identical to
those recognised elsewhere within the Ladolam Deposit.
Porphyry style alteration
An early porphyry alteration style of potassic and propylitic
character at depth is confined to the basement units, and is
related to the emplacement of alkaline intrusive stocks at depth.
In hand specimen, the definitive alteration minerals are:
Potassic: potassium feldspars + biotite
Propylitic: calcite + chlorite + magnetite
Propylitic alteration occurs peripheral to potassic alteration.
Potassic alteration of porphyry style is overprinted by later lower
temperature potassic-silica alteration controlled by fractured
zones, or within brecciated units.
Epithermal-potassic alteration
Epithermal potassic alteration occurs immediately above the
porphyry style alteration and is generally confined to brecciated
units. It formed at lower temperatures than the porphyry potassic
event (Cater, 2002). The dominant potassium feldspar is adularia
and occurs intimately associated with pyrite and quartz.
This potassium feldspar-pyrite alteration is inferred to have
resulted from the dissolution of primary feldspars, biotite,
hornblende, and pyroxene by hot alkaline fluids, with subsequent
flooding and deposition of lower temperature potassium feldspar,
pyrite, and quartz (Corbett et al, 2001).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Epithermal alteration
The epithermal alteration suites overlie the porphyry alteration
and the epithermal-potassic alteration. The definitive alteration
minerals of each suite in hand specimen are:
Advanced
argillic: alunite + sulfur + opaline silica + kaolinite
Argillic: kaolinite + smectite + illite
Phyllic: illite + potassium feldspar + silica
Zones of phyllic alteration occur above the zone of epithermal
potassic alteration where illite is the indicator mineral. Other
minerals include pyrite, quartz and relic potassic alteration. In
drill core, phyllic altered rocks have a medium hardness with
pitted textures and are cream-grey coloured.
Argillic alteration occurs as a blanket over the upper zone of
the Kapit area. It is identified as soft, grey-white coloured clays
or at surface as orange-brown or cream coloured clays. In drill
core, argillic altered rocks have a swelling capacity and
characteristically have a greasy feel to them.
Advanced argillic alteration is associated with contemporary
hot spring activities. The original lithologies have become
extremely leached, and are white to light grey in colour and of
low density. The characteristic minerals are alunite, native sulfur,
and opaline silica. Advanced argillic alteration also occurs within
discrete structurally controlled zones, or as a leached condensate
zone immediately above the main potassic epithermal altered
core of the Kapit breccia complex .
Mineralisation
Gold mineralisation within the Kapit area is confined almost
entirely to the Kapit breccia complex, with a funnel-like shape
that is likely diatreme related. The high grade (+3.5 g Au/t) core
of the orebody is localised at the intersection of NS and NW
trending structural corridors (Figure 5). This intersection appears
to have been a favourable location for diatreme development
where enhanced porosity through breccia development has
created a favourable conduit for later mineralising fluid flow.
329
R P KIDD and J R ROBINSON

High gold grades are associated predominantly with epithermal
potassic (adularia)-silica-pyrite altered, vughy breccias which
form the core of the Kapit breccia complex. Overlying the vughy
breccia core is a zone of argillic clay-pyrite alteration,
corresponding to a highly leached condensate zone. This zone is
typified by gold grades that average greater than 10 g Au/t. The
upper limit of the orebody resembles a shallow conical depression
centred on the core of the breccia complex. The high grade
condensate zone can best be described as a 70 m diameter,
doughnut shaped, mineralised zone, encircling the conical
depression in the upper portion of the orebody (Figure 7).
A more extensive zone of lower grade (+1 g Au/t)
mineralisation hosted by fluidised and crackle breccias is located
peripheral to the core of the Kapit orebody (Figure 7). Outside the
breccia complex, low grade mineralisation extends only a short
distance (generally < 5 m) into the competent, propylitically
altered basement rocks. This mineralisation is fracture hosted,
with little penetration into enclosing host rock.
The Kapit orebody comprises at depth predominantly
K feldspar-sulfide style gold mineralisation, as is evidenced by
its generally high sulfide sulfur content (+6 per cent) and
consistent gold grades. Very minor fracture controlled late-stage
quartz-chlorite-anhydrite style mineralisation has been
recognised.
A unique feature of the Kapit orebody is the development of
an extensive and consistent zone of high grade (+10 g Au/t)
mineralisation within the near surface argillically altered rocks
(Figure 7). Elsewhere within the Ladolam deposit gold
mineralisation occurring within argillically altered rocks is
strongly structurally controlled, characteristically displaying high
grade variations within small localised zones.
Reserves
An updated resource estimate was completed in late 2003 that
included the latest drilling results from Kapit. Detailed open pit
design work based on the updated resource estimate indicated
that to develop the Kapit orebody, construction of a sea retaining
wall and coffer dam in the shallows of Luise Harbour would be
necessary to allow for resource conversion (Lihir Gold Ltd,
2004) (Figure 7). Data available from existing studies, indicated
that this would be both technically feasible and economically
viable. Additionally, no foreseeable regulatory impediments exist
for the granting of necessary approvals to allow for this to occur.
In light of this, an updated reserve statement was issued in
February 2004 including the Kapit mineralisation for the first
time.
Total Probable Reserves at Kapit stand at 25.2 million tonnes
(Mt) at 5.28 grams of gold per tonne (g Au/t), for a total of
4.3 million ounces (Moz) of contained gold (Hunt, 2004). Total
Resources along with Total Proved and Probable Reserves for
Lihir (including Kapit) are presented in Table 4.
Further drilling of the Kapit orebody is planned in 2004/2005.
The orebody remains open to the north and east.

TABLE 4
Total identified Mineral Resources and Ore Reserves, Ladolam Gold Deposit (Lihir Gold Limited, 2004).

Identified Mineral Resource1,5,6 (Resource includes reserves)

Resource category Tonnes Average grade Contained ounces3,4
(millions) (g Au/t) (millions)
Minifie, Lienetz, Borefields, Coastal Measured 36.5 2.86 3.4
and Kapit deposits
Indicated 273.1 3.49 30.7
Inferred 132.9 2.51 10.7
Total resource2 442.5 3.14 44.7

1. Cut-off grade 1.5 g Au/t (equivalent to $400 per ounce cut-off grade).
2. Inferred resource material falling within the planned mining pit, along with indicated material between the resource cut-off of 1.5 g Au/t and the
reserve cut-off of 1.7 g Au/t within the mine pit, is not included in stated resources. This identified material accounts for an additional 21.5 Mt @ 1.98
g Au/t containing 1.4 M oz.
3. Resource for the period has been depleted by mining. Substantial new resource was added in the Kapit and west Lienetz areas.
4. The number of contained ounces does not indicate the ounces that will ultimately be recovered. The ounces ultimately recovered and available for sale
depends on whether, and the degree to which, the mineral resource is converted to ore reserves and upon mining and processing efficiency.
5. 2002 resource was 404.1 Mt at 3.16 g Au/t for 41.1 M oz Au.
6. All $ references are to US dollars unless otherwise stated.

Ore Reserves1,5,6
Reserve category Tonnes Average grade Contained ounces4
(millions) (g Au/t) (millions)
Reserves at 31 December 20032 Proved 3.2 4.01 0.4
Probable 127.1 4.18 17.1
Sub total
Stockpiled ore3 Proved 33.2 2.75 2.9
Total reserves 163.5 3.88 20.4

1. Reserve tonnages have been depleted during 2003 by mining activity. Reserves quoted are those remaining below the mining surface as at
31 December 2003, within a revised ultimate pit design (January 2004), based on the December 2003 revised resource model.
2. Cut-off grade for the reserve is 1.7 g Au.
3. Stockpiled ore totals reflect ore above cut-off currently on stockpile at 31 December 2003.
4. The reserves reflect an assumed life-of-mine gold price of $340 per ounce. The number of contained ounces does not indicate the ounces that will
ultimately be recovered. The company currently estimates an overall recovery of 18.3 million ounces of gold from the 20.4 million contained ounces.
5. 2002 reserve was 143.0 Mt at 3.63 g Au/t for 16.7 M oz Au ($305 per oz).
6. All $ references are to US dollars unless otherwise stated.

330 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A REVIEW OF THE KAPIT OREBODY, LIHIR ISLAND GROUP, PAPUA NEW GUINEA
ACKNOWLEDGEMENTS
The authors wish to thank Lihir Gold Limited for permission to
publish this paper.
REFERENCES
Carman, G, 1994. Genesis of the Ladolam gold deposit, Lihir Island,
Papua New Guinea, PhD thesis (unpublished), Monash University,
Australia.
Cater, G, 2002. Deep hydrothermal alteration at the Ladolam Epitherlam
Gold Deposit, Lihir Island, Papua New Guinea, Masters thesis
(unpublished), The University of Auckland, New Zealand.
Corbett, G, 1998. Comments on the structural control to, and styles of
mineralization in the Minifie zone, Lihir Island, Unpublished
consultants report.
Corbett, G, 2001. Comments on the controls to gold mineralisation at the
Lienetz Zone, Ladolam Gold Deposit Lihir Island, Papua New
Guinea, Unpublished consultants report.
Corbett, G, Hunt, S, Cook, A, Tamaduk, P and Leach, T, 2001. Geology
of the Ladolam gold deposit, Lihir Island, from exposures in the
Minifie open pit, in Proceedings Geology, Exploration and Mining
Conference (Ed: G Hancock), pp 69-77 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
Davies, R M and Ballantyne, G H, 1987. Geology of the Ladolam gold
deposit Lihir Island, Papua New Guinea, in Proceedings Pacific Rim
Congress, Gold Coast, Queensland, August, pp 943-949.
Exon, N F, Stewart, W D, Sandy, M J and Tiffin, D L, 1986. Geology and
offshore petroleum prospects of the eastern New Ireland Basin,
northeastern Papua New Guinea, BMR Journal of Australian
Geology and Geophysics, 10:39-51.
Herzig, P M, Hannington, M D and Stoffers, P, 1999. Petrology, gold
mineralization and biological communities at shallow submarine
volcanoes of the New Ireland fore-arc (Papua New Guinea):
preliminary results of R/V Sonne cruise SO-133, InterRidge News,
7:34-38.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Hunt, S J, 2002. Powerpoint Presentation to Lihir Gold Limited Board,
Unpublished company document, Lihir Management Company.
Hunt, S J, 2004. Base case reserve calculations, Unpublished company
document, Lihir Management Company.
Johnson, R W, 1979. Geotectonic and volcanism in Papua New Guinea: a
review of the late Cainozoic, BMR Journal of Australian Geology
and Geophysics, 4:181-207.
Kennecott Exploration and Niugini Mining Joint Venture, 1992. Lihir
Project feasibility study report, volume 2 Geology and mineral
resources, Unpublished company report.
Licence, P S, Terrill, J E and Fergusson, L J, 1987. Epithermal gold
mineralisation, Ambitle Island, Papua New Guinea, in Proceedings
Pacific Rim Congress 87, pp 273-278 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
McInnes, B I A and Cameron, E M, 1994. Carbonated alkaline
hybridizing melts from a sub-arc environment: Mantle Wedge
samples from the Tabar-Lihir-Tanga-Feni arc, Papua New Guinea:
Earth and Planetary Sciences Letters, 122:125-141.
Moyle, M J, Doyle, B J, Hoogvliet, H and Ware, A R, 1990. Ladolam
Gold Deposit, Liir Island, in Geology of the Mineral Deposits of
Australia and Papua New Guinea, (Ed: F E Hughes), Vol 2, pp
1793-1805 (The Australasian Institute of Mining and Metallurgy:
Melbourne).
Patterson, D B, Farley, K A and McInnes, B I A, 1997. Helium isotropic
compositions of the Tabar-Lihir-Tanga-Feni island arc, Papua New
Guinea, Geochimica et Cosmochimica Acta, 61:2485-2496.
Wallace, D A, Johnson, R W, Chappell, B W, Arculus, R J, Perfit, M R
and Crick, I H, 1983. Cainozoic volcanism of Tabar, Lihir, Tanga and
Feni Islands, Papua New Guinea: Geology, whole-rock analyses, and
rock-forming mineral compositions, Bureau of Mineral Resources
(Australia), 243, 62 p.
331
The Tooloom Gold Project, NSW Forgotten But Not Gone!
Rediscovering a 147-Year-Old Gold Field
R M D Meares1, G G Lowder1, M J White2, B A Wake2 and M D Vickers2

ABSTRACT
The Tooloom gold field is located within the New England Fold Belt in
the far northeast of New South Wales, approximately 130 km south
southwest of Brisbane. Gold was first discovered at Tooloom in 1857.
Within a few years up to 10 000 people were mining placer gold deposits
in the Tooloom valley and a few tried their hand at mining hard rock
occurrences as well. Mining of alluvial gold continued in a small way at
Tooloom until very recently, but only in the last few years, since
Malachite Resources acquired its interest at Tooloom (now 100 per cent),
has the full potential of this large, reduced intrusion-related gold field
been appreciated. By the mid 1870s, once the alluvials were largely
worked out, most of the early miners had made the 270 km trek north to
the newly discovered Gympie gold field and Tooloom slipped into
obscurity. Such was this obscurity that the Tooloom gold field was not
recorded on published Government geological maps until 2001, four
years after Malachite became involved. The many hard rock prospects
that Malachite has identified in recent years were largely or completely
unknown.
Tooloom is a good example of blinkered thinking when it comes to
modern mineral exploration. The New England province in NSW has
long been the poor relation to the much better known and more highly
favoured Lachlan Fold Belt in NSW. It has also suffered serious neglect
compared with the Queensland part of the New England Fold Belt that
hosts such major gold deposits as Mt Morgan (>7 Moz), Gympie
(>3 Moz) and Cracow (>1 Moz). Tooloom is proof that prospective
geology does not stop at state borders. FIG 1 - Location of the Tooloom Gold Project.
When Malachite commenced exploration at Tooloom in 1997 there
was no prior work of substance on which the company could build its Tooloom consists of an entire gold field that, until Malachites
program, so it was effectively back to basics. The geological setting,
involvement, had escaped significant modern exploration and
steep terrain, rate of erosion and forest cover in many parts of Tooloom
are reminiscent of the southwest Pacific islands and Malachites had never been drilled. Numerous old hard rock and alluvial
exploration methodology has evolved from the exploration team workings are spread over an area of more than 100 km2 and
members experience in the search for gold and copper deposits in the historical newspaper accounts indicate that up to 10 000 people
islands. were actively working the field in the 1860s - 1870s. Small scale
Gold at Tooloom is characteristically coarse grained and the largest alluvial mining ceased at Eight Mile in 1997 and at Joes Gully in
nugget discovered historically at Tooloom contained 140 oz of gold. 2001. Old hard rock workings include shafts, adits, pits and
Many of the nuggets seen by Malachite geologists consist of intergrown trenches and are developed mainly on structures, with a variety
coarse gold and vein quartz. of host rocks and mineralisation styles. Few of these workings
After a systematic program involving geological mapping, are of significant size. Old alluvial workings occur in all the
helicopter-borne aeromagnetics, BLEG stream sediment sampling, main streams and many of the tributaries. Gold in the Tooloom
ridge-and-spur soil geochemistry, trenching, and mapping and sampling area is characteristically coarse grained and nuggets are
of old workings, Malachite has defined drill targets at ten prospects. The common, the largest recorded containing 140 oz of gold. Visible
BLEG technique in particular has been very successful in locating gold hosted by stockwork quartz veins occurs in outcrop and
outcropping gold sources.
many of the alluvial nuggets comprise fragments of eroded
Reconnaissance drilling at a key prospect, called Phoenix, took place gold-bearing quartz vein material.
in December 2003 and produced long intercepts of low grade gold
mineralisation. A comprehensive induced polarisation survey conducted Malachite first acquired an interest at Tooloom in mid 1997,
at Phoenix in March 2004 identified a large chargeability anomaly and when it recognised that the widespread alluvial gold in the area
showed that the initial drilling, targeted on surface geochemistry, had was shedding from numerous hard rock gold occurrences, some
intersected only the periphery of the IP anomaly. Follow up drilling at of which displayed evidence of strong hydrothermal alteration of
Phoenix began in May 2004. intrusive, highly pyritic host rocks. Gold was also noted to be
shedding from quartz vein stockwork systems in sericitised and
INTRODUCTION pyritic sandstone and similar rocks of the Emu Creek Formation.
Although only three hours drive from Brisbane, the Tooloom
The Tooloom Gold Project is owned 100 per cent by Malachite area is relatively remote, sparsely populated and quite rugged,
Resources NL (Malachite) and comprises tenements covering especially in the southern part. Government metallogenic
approximately 300 km2. The project is located in the far northeast mapping had not, in 1997, extended to this part of the state and
of NSW, within about 30 km of the Queensland border (Figure 1).
although the presence of gold at Tooloom was recorded by
Markham and Basden (1975), little was known about the nature
1. FAusIMM, Malachite Resources NL, PO Box 42, Lindfield NSW of the mineralisation. The large size potential of some of
2070. Toolooms gold occurrences was not recognised. It was the
intensity of hydrothermal activity, the widespread and varied
2. Malachite Resources NL, PO Box 42, Lindfield NSW 2070.
nature of the gold mineralisation, and the virtually unexplored

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 333

R M D MEARES et al
status of the Tooloom area that encouraged Malachite to
persevere with what has proven to be a challenging task for a
small company. Limited funding meant that progress was slow
but perseverance pays in mineral exploration and the company is
now set to see its efforts amply rewarded.
REGIONAL GEOLOGY
The Tooloom Project lies in the southern (NSW) section of the
Palaeozoic New England Fold Belt, within the Emu Creek Block,
which is overlain on the east by the Mesozoic Clarence-Moreton
Basin (Figure 2). The New England Fold Belt further north in
Queensland hosts the Mount Morgan, Gympie and Cracow gold
deposits that have produced over seven million, three million and
one million ounces of gold respectively.
TOOLOOM GOLD
PROJECT
FIG 2 - Simplified regional geology of north eastern NSW, showing
Tooloom Gold Project located in the eastern part of the Emu
Creek Block.
The main rock units at Tooloom (Figure 3) are gently folded
quartzo-feldspathic and volcaniclastic sandstone, siltstone,
conglomerate, carbonaceous mudstone and tuffaceous rocks of
the Emu Creek Formation, thought to be of Permo-Carboniferous
age. The sedimentary rocks are intruded by a variety of igneous
phases in the form of dykes, plugs and larger plutons, most of
which were not previously recorded. Lithologies include felsic
and intermediate porphyry, diorite, tonalite and dolerite (Figure
3). Post-orogenic, I-type, high-K, Permo-Triassic granitic
intrusions belonging to the Clarence River Supersuite (Bryant,
Arculus and Chappell, 1997) occur to the south of the Tooloom
area (eg Jenny Lind Granite) and the newly mapped intrusive
rocks at Tooloom probably belong to this group. Bryant, Cosca,
and Arculus (1997) dated the Clarence River Supersuite as of
Permo-Triassic age. Hydrothermal alteration and sulfide
mineralisation, commonly with gold, are well developed in many
of the intrusive bodies and in their sedimentary wall rocks. Some
of the intrusions represent significant magnetic highs and can be
mapped using detailed magnetic data.
334 Adelaide, SA, 19 - 22 September 2004
To the east and north, the Emu Creek Formation and the
intrusive rocks are overlain unconformably by Jurassic-aged
sediments of the Clarence Moreton Basin. Fossilised wood at
the base of the Jurassic is common in the area, in places
underlain by Jurassic palaeo-placer gold occurrences. The
Jurassic rocks are overlain locally by Tertiary basalt lava flows,
which occur on some of the high hilltops (inverse topography).
Perched fossil alluvial deposits (deep leads) occur on some of
the ridges beneath the Tertiary basalt and a few have been
worked for placer gold. Recent alluvial deposits in Tooloom
Creek and its tributaries have been extensively worked for placer
gold.
The region displays a strong structural grain, clearly visible in
aeromagnetic images. North northeast and northwest structures
predominate, commonly controlling mineralisation, affecting
drainage patterns and influencing the orientation of intrusions.
Dilational structural zones occur in many places and some
contain intrusions, hydrothermally altered breccia bodies or
sheeted auriferous quartz veining.
REDISCOVERING TOOLOOM
Exploration by the Malachite team for a major hydrothermal
gold deposit at Tooloom commenced in 1997 and started with a
clean slate. There were:
1. no gold occurrences marked on the published 1:100 000
geological map;
2. virtually no recorded previous exploration; and
3. no modern aeromagnetic coverage of the area.
The team recognised that, although there are Jurassic and
Tertiary cover rocks on the eastern and northern margins of the
area, a large part of the Emu Creek Block is well exposed. Given
the lack of previous exploration, if an economic ore body were to
be present it could very well occur in outcrop and thus
prospecting and drainage geochemistry were likely to be
effective search methods.
Several members of the exploration team had worked on gold
and porphyry copper exploration programs in the southwest
Pacific islands where the geological setting, steep terrain, rate of
erosion and forest cover are comparable to at least parts of
Tooloom. Consequently Malachites exploration methodology at
Tooloom evolved from that experience, and in addition had to
contend with thick, secondary re-growth, including much
lantana, in many of the creek beds and on adjacent hill slopes.
The initial approach, aimed at generating prospects for follow
up, was based on traversing the creeks to:
1. collect 2 kg BLEG (bulk leach extractable gold) stream
sediment samples to delineate anomalous drainages;
2. sample veined or altered rock outcrops and float in the
creeks; and
3. create a 1:25 000 scale geological map of the project area.
Information about the location of hard rock gold workings
was gleaned mainly from local landowners. Mapping of road
cuttings added to the geological picture.
Experience soon showed that most known hard rock gold
localities had strongly anomalous BLEG trains in streams
draining them, with BLEG values of the order of 20 to 60 ppb
Au, irrespective of whether there were old alluvial workings in
the stream banks. Regional background BLEG values are <1 ppb
Au. Further work revealed that values as low as 8 ppb Au could
indicate the presence of outcropping gold mineralisation
upstream. 5 ppb Au was therefore adopted as the anomalous
threshold level. A number of streams anomalous at the 5 to 15
ppb Au level are yet to be investigated further.
PACRIM 2004

FIG 3 - Simplified geological map of the Tooloom area, showing prospects identified by Malachite.
This stream sediment sampling program identified numerous
highly anomalous drainages warranting follow up. The
anomalous catchments tended to occur in clusters, some
associated with known hard rock gold occurrences and others
not. The most anomalous stream sediment values of all (up to
80 ppb Au) came from Phelps Creek, in the northern part of the
area, where no significant old workings were then known or have
been found since. This area has, however, emerged as the
companys most important prospect at Tooloom and is now
known as the Phoenix prospect. Phoenix is a true grassroots
discovery.
Initial follow up of anomalous drainages included more
detailed BLEG sampling, further rock chip sampling, geological
mapping and panning of the creek sediments for visible gold.
Once the interpreted hard rock source areas were established,
ridge-and-spur soil sampling was conducted at 25 m intervals
along selected ridges. Samples were assayed for gold by the
BLEG technique and for a range of other metals by the ICP
technique. The BLEG technique was used for gold in soils, rather
than fire assay of -80# samples, because orientation work had
shown that BLEG tended to smooth out the nugget effect
arising from the widespread occurrence of coarse gold. The soil
sampling stage was complemented by detailed prospecting and
geological mapping.
In 2000 a helicopter-borne aeromagnetic survey was
conducted over a substantial part of the Tooloom gold field,
producing an outstanding image that has been invaluable for
interpreting structure and intrusive relationships. Later that year,
three diamond holes were drilled into a dioritic intrusive body at
Frasers Gully, where a possible porphyry copper system had
been newly recognised. These holes intersected strongly altered,
reduced dioritic intrusive, with abundant pyrite and pyrrhotite in
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
THE TOOLOOM GOLD PROJECT, NSW
multi-phase quartz-carbonate veins. Copper and gold values in
the intrusive were generally low. Arsenopyrite was locally well
developed in adjoining hornfelsed sedimentary rocks. Streams
draining the diorite body have been worked extensively for
alluvial gold and significant potential remains to be tested in this
vicinity.
Since 2000 attention has been focussed exclusively on gold.
Using the systematic approach described above, drill targets have
been delineated at ten prospects, several of which have now
received at least reconnaissance drilling, with highly encouraging
results. The Phoenix prospect, discovered through anomalous gold
in Phelps Creek and further defined by an induced polarisation
geophysical survey, has now been confirmed as a major,
intrusive-related gold system with very large size potential.
STYLES OF GOLD MINERALISATION AT
TOOLOOM
At this stage it seems probable that all of the gold mineralisation
at Tooloom is broadly related to one or more of the various
intrusive phases identified in mapping and in drilling. The best
mineralisation tends to occur in rocks of the Emu Creek
Formation adjacent to such intrusive bodies, especially where
they have been hornfelsed, rather than in the intrusive rocks
themselves. In such cases gold can be found in both sheeted and
stockwork quartz vein systems, as well as in hydrothermal
breccia. Gold also occurs within the intrusive rocks in some
places, both disseminated with pyrite and in narrow quartz veins
or quartz vein stockworks. In a few cases gold is concentrated in
well defined faults or other dilational linear structural features
than can be traced for up to 1 km or more and have both north
northeast and northwest orientations.
335
R M D MEARES et al
The most important styles of gold mineralisation identified at
Tooloom to date are:
Breccia-hosted gold occurs in strongly fractured to
intensely brecciated rocks, including both sediments and
intrusives. The best known example of this type is the Phoenix
prospect (Figure 4), where there is a large polymictic,
hydrothermal breccia, with unsorted, angular fragments of
sandstone, diorite and dolerite enclosed in a matrix of
open-space quartz and sulfides (mainly pyrite, pyrrhotite and
arsenopyrite). The Phoenix breccia trends north northeast and
is at least 1.2 km long and up to 400 m wide. As described in
more detail below, drilling has confirmed that this breccia
body is the preferred host for the substantial body of gold
mineralisation newly discovered at Phoenix. Another
significant body of mineralised breccia occurs at Cullens
prospect, although in this case the brecciation may be tectonic,
filling a zone of dilation between faults.
Quartz vein stockworks and sheeted zones gold occurs
as particles of native gold, commonly quite coarse in grain
size, enclosed in quartz vein stockworks and sheeted zones
hosted by brittle-fractured sandstone and conglomerate of the
Emu Creek Formation. These tend to occur around the
margins of dioritic intrusions and the veins are generally of
the order of 10 mm thick, although some are up to 100 mm.
Vein intensities (number per metre) vary widely, but there are
significant areas of five to ten veins per metre.
Back Creek and Rise and Shine (stockworks) and Nine Mile
(sheeted veins) are examples of this type at Tooloom. At
FIG 4 - Soil geochemistry and location of the breccia at the Phoenix prospect.
336 Adelaide, SA, 19 - 22 September 2004
Back Creek particles of gold up to about 5 mm across have
been observed within stockwork quartz veins in outcrop. At
Nine Mile, trenching through several strong soil anomalies
(where panning of soil revealed gold particles up to 1 mm in
size) revealed sheeted, auriferous quartz veins in zones up to
about 20 m wide. At Joes Gully some of the nuggets (up to
3 cm in size) recovered from recent alluvial mining consist
of fragments of vein quartz with coarse intergrown gold.
These are possibly eroded from stockwork quartz zones
observed in outcrops of hornfelsed Emu Creek sandstone
mapped upstream and upslope from the alluvial workings.
Carbonate-base metal veins gold occurs in discrete,
structurally-controlled (ie linear), carbonate-rich veins,
together with abundant sulfides, mainly pyrite, sphalerite and
galena. High silver values may also be associated with the
gold and base metals. The best example of this type is the
Dairy Shaft workings at Frasers Gully, which is probably the
largest old hard rock working in the Tooloom area, where
this type of vein is developed in a quartz-sericite altered,
pyritic diorite host rock.
Porphyry-hosted stockwork systems weak gold
mineralisation is present in quartz vein/pyrite-pyrrhotite
stockworks hosted by diorite porphyry at Frasers Gully, in
part enclosed within a broader halo of weak copper
mineralisation. At Watsons prospect a much more felsic
porphyry intrudes the Cullens Creek Granite and hosts a very
well developed quartz vein stockwork with gold, silver,
minor chalcopyrite and pyrite in the quartz veins.
PACRIM 2004

Pseudo-epithermal veins gold occurs in veins with some
pyrite and a clay-carbonate alteration assemblage but lacking
major base metals. These veins are typically banded in a
manner reminiscent of epithermal veins seen in much
younger volcanic terrains elsewhere in the southwest Pacific.
The best example at Tooloom is the Wakey Wakey workings,
where the mineralisation is described as pseudo-epithermal,
as it is believed that mineralisation took place a depth too
great to allow classification as true epithermal.
Mineralised dilation zones gold occurs in quartz veins or
quartz-carbonate veins introduced into various host rocks
where a flexure or off-set in a fault creates a dilational
environment (eg Pine Gully and Mt Amy) or where folding of
Emu Creek sedimentary rocks creates a dilational environment
along bedding planes (eg Pioneer adit at Joes Gully). At Pine
Gully the dilational veins are individually up to 5 cm wide but
occur in anastomosing form and carry 30 - 60 g/t Au and
300 - 400 g/t Ag. They are enclosed in highly altered,
brecciated quartz diorite porphyry that carries up to 5 g/t Au.
EXPLORATION RESULTS
Currently the most interesting target at Tooloom is undoubtedly
the Phoenix prospect, where a large body of sulfide-bearing rock,
including both sedimentary and intrusive lithologies, surrounds
and incorporates a breccia zone that is of the order of 1.2 km in
length and from 100 to 400 m wide (Figure 4). Five
reconnaissance drill holes were drilled at Phoenix in December
2003, targeted mainly on soil geochemistry and outcropping
breccia. However, due to limited access available at the time,
only the northeastern part of the system, as now known, had been
mapped and sampled at surface. This meant that the initial
drilling tested mainly the northern and eastern margins of the
Phoenix sulfide system as it is now delineated. Even so, all five
drill holes intersected long intervals of low grade gold
mineralisation. The best result was obtained in PHRC02, which
Drill holes
Sea level
100 m
Elevation looking SE
Drill holes
Oblique view looking ESE
FIG 5 - Model of the IP chargeability anomaly at the Phoenix prospect, showing location of initial drill holes (PHRC01 to PHRC05).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
THE TOOLOOM GOLD PROJECT, NSW
intersected 162 m of 0.42 g/t Au, including 79 m of 0.72 g/t Au
which correlated with the intercept of strongly brecciated rock in
the hole. Significantly, gold was present in every metre of
intercepted breccia, generally in the range 0.3 to 2.8 g/t Au.
Elsewhere in this hole, many of the gold values in non-brecciated
wall rock were also very anomalous (>1 g/t Au) but the average
was distinctly lower than in the breccia, with sections of <0.1 g/t
Au over one to several metres.
The large size of the Phoenix sulfide system was apparent
from surface mapping but became much more clearly defined as
a result of an induced polarisation geophysical survey completed
in March, 2004. This survey utilised the off-set pole-dipole
method described by White, Collins and Loke (2003) and
produced an outstanding three-dimensional model of the Phoenix
sulfide system (Figure 5). The more intense part of the IP
chargeability anomaly has a well defined doughnut shape, is
approximately 1 km across and at least 400 m deep. When the
locations of the five initial drill holes are entered into the model
it is apparent that the early drilling did not test the IP anomaly at
all adequately. The best hole (PHRC02) mentioned above came
closest to intersecting the most chargeable rock but in fact just
clipped the outer edge of the IP doughnut.
Elsewhere at Tooloom attractive targets have been defined at
Joes Gully, Cullens and Watsons prospects. Joes Gully is of
particular interest as coarse grained gold was being mined there
from alluvial deposits in the creek bed as recently as 2001. Of
particular note is that many of the nuggets recovered by recent
alluvial mining consist of composites of gold and vein quartz
(Figure 6a). These appear to have been eroded from outcrops of
altered sandstone displaying well developed quartz vein
stockworks (Figure 6b) located just upstream and upslope from
the mined area.
A new drilling program, targeting geochemistry, breccia and
the IP anomaly, commenced at Phoenix in May 2004.
Reconnaissance RC percussion drilling was also being conducted
at Joes Gully and several other prospects in the Tooloom area.
N
200 m
Plan
Phoenix
Prospect
Model of IP
survey results
337
R M D MEARES et al
2 cm
A B
FIG 6 - Joes Gully prospect. (A) Composite gold/vein quartz nuggets recovered by recent alluvial mining
(B) Altered sandstone with well developed quartz vein stockwork possible source of nuggets shown in (A).
CONCLUSIONS
Exploration at Tooloom is still at too early a stage to make
definitive statements about the nature and economic potential of
gold mineralisation discovered there by Malachite. However, it is
clear that Tooloom should be seen as a large and well
mineralised district, with gold occurrences extending for at least
20 km in a north-south direction. Multiple styles and multiple
centres of mineralisation have been identified. Some of these,
such as Watsons and Joes Gully, are associated with minor old
hard rock workings or alluvial mining, but others, like Phoenix,
are true grass roots discoveries. Phoenix in particular is a very
large body of mineralisation with clear potential to host a
multi-million ounce gold deposit.
Definitive classification of the Tooloom mineralisation as to
ore deposit type will come as more data are generated by
drilling. At this stage, nevertheless, there is good evidence to link
the mineralisation seen to date to those gold deposits related to
reduced granitic intrusions, as described by Thompson and
Newberry (2000). Such deposits are characterised by a metal
association involving gold with one or more of Bi, Te, As, W, Mo
and Sb, while sulfide assemblages include prominent pyrrhotite
and arsenopyrite, together with pyrite, and alteration is marked
by early albite or K feldspar and later sericite-carbonate. The
gold at Phoenix prospect is accompanied by pyrrhotite, pyrite,
arsenopyrite, stibnite and minor molybdenite and chalcopyrite,
and the host rock displays sericite-carbonate alteration,
consistent with the reduced intrusion model. Similarly, drilling
for copper at Frasers Gully in 2000 intersected sericite-carbonate
altered diorite (possibly with earlier secondary albite) and
abundant pyrrhotite as well as pyrite, arsenopyrite and minor
chalcopyrite, also consistent with the reduced intrusion model.
On this basis, it may be pertinent to compare Tooloom and in
particular Phoenix with Kidston in north Queensland (Baker and
Tullemans, 1990) and perhaps with the Pogo gold deposit in
Alaska (Thompson and Newberry, 2000).
Much work remains to be done at Tooloom but the results
achieved to date provide strong encouragement that a major new
Australian gold province is in the process of being unveiled.
338 Adelaide, SA, 19 - 22 September 2004
Malachites exploration methodology has been very effective in
identifying individual prospects but it has taken considerable
perseverance on the part of the company to continue exploration
through very lean periods for exploration funding and in spite of
rejection of the project as a joint venture opportunity by a
number of Australias larger gold mining companies. That a
previously unknown gold-bearing sulfide system as large as
Phoenix can be found in outcrop only three hours drive from
Brisbane seems scarcely credible to some, but similar attitudes
were to be found when the Cadia Hill gold discovery, only three
hours drive from Sydney, was first reported in the early 1990s.
Gold was first discovered at Tooloom in 1857, making it one
of the oldest gold fields in Australia. In rediscovering Tooloom,
Malachite is making it one of Australias newest gold fields and
the company believes that the best is yet to come for Tooloom.
REFERENCES
Baker, E M and Tullemans, F J, 1990. Kidston gold deposit, in Geology
of the Mineral Deposits of Australia and Papua New Guinea (Ed: F E
Hughes) pp 1461-1465 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Bryant, C J, Arculus, R J and Chappell, B W, 1997. Clarence River
Supersuite: 250 Ma Cordilleran tonalitic I-type intrusions in Eastern
Australia, Journal of Petrology, 38(8):975-1001.
Bryant, C J, Cosca, M A and Arculus, R J, 1997. 40Ar/39Ar ages of
Clarence River Supersuite intrusions from the northern portion of the
New England Batholith, southern New England Orogen, in Tectonics
and Metallogenesis of the New England Orogen: Alan H Voisey
Memorial Volume (Eds: P M Ashley and P G Flood) Geological
Society of Australia Special Publication 19, pp 242-253 (Geological
Society of Australia).
Markham, N L and Basden, H, (Eds) 1975. The Mineral Deposits of New
South Wales, (NSW Department of Mineral Resources: Sydney).
Thompson, J F H and Newberry, R J, 2000. Gold deposits related to
reduced granitic intrusions, in SEG Reviews, 13:377-400.
White, R M S, Collins, S and Loke, M H, 2003. Resistivity and IP arrays,
optimised for data collection and inversion, Exploration Geophysics,
34(4):229-232.
PACRIM 2004
The Perseverance Gold Deposit The Next Step at Peak
S Munro1 and R Berthelsen2
ABSTRACT
The Peak Gold Mines Pty Ltd (PGM) operations are located
approximately 9 km south of Cobar, NSW. PGM currently mines and
treats ore from three separate orebodies within 5 km of the original plant.
Production from the original Peak orebody ceased in April 2002, with ore
currently sourced from underground at New Occidental and Perseverance,
while New Cobar supplies ore from an open cut mine. In March 2003,
Wheaton Minerals Asia Pacific purchased the operation from Rio Tinto
who operated the mine for around ten years.
The Perseverance deposit is situated between 600 and 1000 m south of
the Peak orebody. Access is via shaft from Peak to 630 m below surface
and then by decline to the northern end of the deposit locally known as
Zone A. Development of the resource commenced mid 2002 to set up for
a long hole open stoping operation at 25 m level spacing. The first
production stope was successfully brought on line December 2003.
The Perseverance deposit is hosted by sediments and volcanics along
the eastern margin of the Devonian Cobar basin and is one of many
deposits with the Cobar Gold Field. The Cobar Gold Field has produced
approximately 2.5 million ounces of gold since 1880. The Perseverance
area had long been considered prospective due to the presence of old
workings at the southern extensions of Peak line of lode. Significant
mineralisation was first discovered at Perseverance in 1994 when PGM
drilled a hole targeted 950 m below the historic workings. The target was
based on an apparent plunge to the stratigraphic controls to the Peak
Orebody. Subsequent exploration within and around the greater
Perseverance area has been significantly aided through the use of
downhole electro-magnetic surveys due to the close spatial association
between gold and sulfide mineralisation.
Ore grade mineralisation commences approximately 850 m below
surface and the resource remains open at depth (>1400 m). Close spaced
drilling indicates that multiple gold lenses are stacked, subparallel to a
NW cleavage in en-echelon fashion to form a general NNW orientation.
Studies reveal a paragenetic sequence of early gold subsequently
overprinted and partially remobilised by chalcopyrite-pyrrhotite
mineralisation and again overprinted by galena-sphalerite dominated
mineralisation.
Three main mineralisation styles are evident:
1. rhyolite vein hosted Au with strong Cu occurs above the
rhyolite/shear intersection generally just inside the rhyolite margin;
2. contact shear hosted Au and Cu at the sheared contact of rhyolite
with sediments within the higher levels of the deposit; and
3. sediment hosted Au low Cu below the rhyolite/shear intersection.
Similar to the Peak gold deposit, the mineralisation at the Perseverance
deposit is principally controlled by the intersection of the Peak shear zone
with fold closures in a stratabound rhyolite unit. Relying on the
competency contrast of rhyolite juxtaposed against softer sediments, the
fold closures act as fluid traps to allow for the precipitation of gold and
sulfide bearing ores.
INTRODUCTION
Concealed by 830 m of country rock, the Perseverance fault
breccia and vein hosted Au-Cu (Pb-Zn-Ag) deposit is located
approximately 9 km south of Cobar, central NSW (Figure 1).
The deposit forms part of the gold and base metal deposits
associated with the Peak Shear system, a group of north-south
trending, subvertical faults that also host the exhausted Peak
Gold Deposit 600 m to the north. The Perseverance deposit is
operated by Peak Gold Mines, a wholly owned subsidiary of
1. Project Geologist, Peak Gold Mines, Hillston Road, Cobar NSW 2835.
2. MAusIMM, Chief Geologist, Peak Gold Mines, Hillston Road, Cobar
NSW 2835.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Canadian based Wheaton River Minerals Ltd. Initially
discovered in 1994, the deposit went into production in late
2003. As at 31 December 2003, the JORC compliant resource
inclusive of reserves total 2.0 Mt @ 11 g/t Au, 0.9 per cent Cu,
0.1 per cent Pb, 0.1 per cent Zn and 9 g/t Ag. Remaining open at
depth, Perseverance forms the platform for the future of Peak
Gold Mines with current planned production from the resource to
year 2009 at approximately 150 000 ore tonnes per annum.
REGIONAL GEOLOGY
The Cobar Gold Field is located on the eastern margin of the
Early Devonian Cobar Basin (Glen, 1991). The Cobar Basin lies
within the northern part of the Central Belt of the Lachlan
Orogen, as defined by Glen (1995). Regional crustal extension of
the Lachlan Orogen in the late Silurian created a series of
north-south trending deepwater troughs and basins in the Cobar
region. These include the Cobar Basin with the Raast and Mt
Hope Troughs further to the south. The basin contains a
two-stage fill, regarded as two sequences representing syn-rift
and post-rift stages (Glen et al, 1996). The lower unit, the Nurri
Group, is a fining upwards sequence comprising initial
shallow-water sediments that pass rapidly up into more extensive
siliciclastic turbidites. The upper unit, the Amphitheatre Group,
comprises much more extensive lower energy turbidites.
Glen (1991) indicates that volcanics form only a small
proportion of the exposed syn-rift sequence in the Cobar Basin,
although volcanic units may be more prevalent at depth in the
basin. Small bodies of felsic porphyry are exposed near the
Queen Bee mine to the south of the Gold Field and bodies of
intrusive flow-banded rhyolite are exposed in the underground
workings at the Peak mine.
The Cobar Basin was inverted in the late Early Devonian
(395 - 400 Ma) in response to northeast-southwest compression
(Glen, 1992). This compression was partitioned into a high strain
zone along the eastern margin of the basin and a low strain zone
in the central and western parts of the basin. Deformation of the
basin sediments was controlled by the re-activation of the early
basin-forming extensional faults and by the formation of new
faults (Glen, 1991; 1995). Inversion of the western margin of the
basin, at a considerably lesser intensity than that experienced by
the eastern margin, subsequently occurred in the Carboniferous
(Glen, 1991). Cobar Basin sediments are generally
metamorphosed to chlorite grade (Glen, 1995). Glen et al, (1992)
constrain the age of metamorphism to between 395-400 Ma, on
the basis of 40Ar/39Ar dating of white micas formed during
cleavage development in the higher strain zone along the eastern
margin of the Cobar Basin.
Locally, the structurally controlled Cobar Gold Field is located
on the western limb of the Chesney-Narri Anticline, a moderate
to tight south-plunging anticline cored by sandstones of the
Chesney Formation (Glen, 1987b). Gold mineralisation is
associated with two principal shear zones, the Peak Shear and
Great Chesney Fault Zone, both of which juxtapose the Chesney
Formation with the Great Cobar Slate. The deposits usually
consist of multiple, steep-dipping lenses with short strike lengths,
narrow widths and strong vertical extents. A strong subvertical
regional white-mica S2 cleavage and a steep north-plunging
mineral and extension lineation L2 are present throughout the
Gold Field. This cleavage is subparallel to the major faults in the
Gold Field area. 40Ar/39Ar dating of alteration assemblages at
Peak Gold Mine indicate the main stage of mineralisation
occurred at 401.5 1.0 Ma (Perkins et al, 1994).
339
S MUNRO and R BERTHELSEN

FIG 1 - Cobar Goldfield geology and perseverance location plan.

340 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


ROAD FROM DISCOVERY TO DEVELOPMENT
A re-appraisal of the Cobar Gold field during 1993 highlighted
the potential for the Perseverance area to contain Peak style
mineralisation at depth below the shallow Perseverance
workings. It was principally based on the apparent southerly
plunge of the stratigraphic controls to the Peak orebody.
CRAE-Peak Gold Mines followed up the Perseverance target
with deep surface drilling in 1994. The fourth of a five-hole
program, DD94PK57B intersected 7 m @ 4.5g/t Au and one per
cent Cu within rhyolite at approximately 950 m depth. On the
strength of this intersection together with the prohibitive costs
and logistical difficulties of effectively drilling the target from
surface, an 800 m long underground exploration drive was
excavated from the base of the Peak workings at an approximate
cost of $2.4 million. This decline could then serve as an access to
any future mining at Perseverance. The drive was completed in
late 1997. Underground drilling of the Perseverance orebody (to
1500 m depth) down dip from the discovery hole commenced
soon after with immediate success, UD97PP971 11 m @ 4 g/t
Au and 0.5 per cent Cu, UD98PP974 18 m @ 80 g/t Au and 1.4
per cent Cu and UD98PP957 8 m @ 19 g/t Au and 0.5 per cent
Cu. During this time a series of deep surface holes were designed
to test the depth extensions to mineralisation approximately
300 m south of the discovery hole. The drilling intersected weak
mineralisation within the Perseverance shear before passing into
barren rhyolite, only 50 m above significant gold and copper
mineralisation. As yet, no attention had been drawn to the upper
northern extensions to the discovery hole now known to host the
high grade A zone (0.7 Mt @ 10.3 g/t Au and 1.34 per cent Cu)
within the Perseverance deposit.
In mid to late 1998, a number of the deeper peripheral surface
and underground holes were selected for downhole
electro-magnetics due to the strong spatial association between
the gold and copper mineralisation (Mutton, 1999). The result
was several modelled conductors that accurately depicted the
outline of the deeper sections of the resource. The results also
indicated that the discovery section was drilled close to the
mid-strike position of the resource. Drilling subsequently
focussed on the modelled conductors at an approximate
100 100 m pattern. In 1999, the upper high grade Zone A area
of Perseverance was discovered. From 2000 through to 2001
drilling initially focussed on defining Zone A to a 50 50 m
pattern, for which development approval was granted in early
2002. Since then drilling concentrated on the deeper, southern
Perseverance area, known as Zone D. The resource remains
open at depth (>1400 m) to the south. The first production stope
from Perseverance was fired late 2003.
Since discovery, the combination of progressively closer
spaced drilling with a greater understanding of the nature of the
mineralisation has resulted in the continual growth of the
resource (Table 1). Over time drill patterns have progressed from
an initial 100 100 m pattern (gross exploration pattern), down
to 50 50 m (evaluation pattern) and finally 15 15 m (mine
delineation pattern). These results suggest that as the mine
delineation pattern progresses to the deeper sections of the
deposit further increases to the resource should be expected.
TABLE 1
Previous resource estimates for Perseverance.
Resource estimate Tonnes (Mt) G/t Au % Cu Oz
May 2000 1.2 11.8 1.5 440 000
Nov 2001 1.0 11.0 1.2 355 000
June 2003 1.4 12.3 1.0 550 000
Dec 2003 1.9 11.3 0.9 690 000
Sweeney, 2000; Dettbarn, 2001.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
THE PERSEVERANCE GOLD DEPOSIT THE NEXT STEP AT PEAK
HOST ROCKS
The Perseverance deposit is hosted by Early Devonian aged
sediments and subvolcanic intrusives that form part of the Nurri
Group, known to represent a syn-rift basin fill sequence on the
western limb of the regional south plunging Chesney-Narri
Anticline.
Sediments
Bounding the deposit to the west is a turbidite package of graded,
well bedded sandstone and siltstone (Figure 2). Individual beds
vary in thickness from several cm through to 50 cm with
sandstone forming the predominant lithology. The package dips
steeply west and strikes NW. The package is generally poorly
cleaved. The trace of this unit with the Perseverance fault
plunges moderately south.
Bounding the east margin of the deposit is a thick monotonous
sequence of poorly bedded shale with rare thin sandy beds.
Bedding is vertical to steep west dipping and strikes NW.
Moderately to well cleaved, this unit is interpreted to exhibit a
shallow south plunging trace against the Perseverance fault.
Subvolcanic intrusive
Described as rhyolite, the subvolcanic intrusive occurs within the
sequence of shale on the east margin of the deposit. Felsic in
composition, the rhyolite can be described as either coherent or
incoherent. The coherent rhyolite consists of a fine-grained felsic
mass that displays flow banded, massive and perlitic textures.
The incoherent rhyolite generally consists of silicified, variably
clast and matrix supported rhyolite breccia within a siliceous
fine-grained siltstone matrix. It is considered that the coherent
rhyolite represents the core of the rhyolite body within an
incoherent brecciated rhyolite margin. The various textures
indicate that the rhyolite intruded wet unconsolidated sediments
at subvolcanic levels with rapid quenching causing brecciation.
The rhyolite body is interpreted to be folded around a gently
south plunging anticlinal fold that represents a minor fold on the
western limb of the regional Chesney-Narri Anticline.
STRUCTURAL SETTING
Quartz breccia and vein hosted gold and base metal
mineralisation is intimately associated either within or adjacent
to the Perseverance Shear and several smaller parallel structures
to the east. The Perseverance Shear strikes NNW and dips from
steep west above the rhyolite to steep east below the rhyolite.
The shear varies in width from 1 m up to 30 m and refracts
strongly where it comes into contact with the rhyolite (Figure 2).
Individual ore lenses mimic the orientation of the Perseverance
Shear and vary in thickness from four to 30 m, averaging 8 m. In
general, the most intense gold and more so, base metal
mineralisation occurs where the rhyolite is sheared against the
more sandstone rich portions of the Chesney Formation (Figure
3). Three highly mineralised zones have been defined within the
global deposit to date, a northern Zone A lens (shear hosted
gold), Zone B (vein hosted gold) and the southern Zone D
lens (mixture of shear and vein hosted).
A strong, pervasive, steep westerly dipping cleavage strikes
NW to NNW averaging 335. Underground exposures combined
with diamond drilling shows that foliation parallel en-echelon
lenses of high grade gold are stacked in a gross NNW
orientation.
ALTERATION
Successive overprinting generations of quartz and Fe-rich
chlorite are the main constituent alteration minerals associated
with the Perseverance deposit. Over time quartz in veins
341
S MUNRO and R BERTHELSEN

25600E 25800E 26000E

> 7g/t Au

> 1g/t Au

Rhyolite breccia

Coherent rhyolite

DNG top of sandstone dominant litho

9400 mRL
Perseverance shear

Bedding traces

DNC31

9200 mRL

DNC2

?
?
?
?
??
?

9000 mRL

FIG 2 - Cross-section through upper Perseverance Zone A on 9890N (mine grid). DNG-massive laminated shale,
DNC2-sandstone dominated turbidite, DNC31-siltstone dominated turbidite.

becomes coarser-grained and chlorite becomes increasingly
Mg-rich. Sericite commonly occurs as a peripheral halo around
more intense alteration. Alteration paragenesis and style is as
follows:
1. Early pre-gold, pervasive crypto-crystalline silicification
with peripheral pervasive sericite. Is most intense along the
Peak shear/rhyolite interface.
2. Multiple stages of cleavage parallel quartz-minor Fe-chlorite
veining and brecciation with Fe-chlorite alteration. Early
stages of veining and brecciation were synchronous with
regional deformation and the main gold and copper
mineralisation event. These veins are commonly folded and/
or boudinaged. Several exploration holes into the deepest
portions of the resource contained instances of fine biotite
within vein complex zones footwall to the Perseverance
shear. Reynolds (2000) described this feature as the initial
pro-grade metamorphic alteration of chlorite to biotite.
3. Late black Mg-rich chlorite talc and carbonate alteration
occurs as small subvertical overprinting shears. Strongly
associated with late silver-galena-sphalerite mineralisation.
4. Late buck quartz and feldspar in subhorizontal tension veins.
ORE ELEMENT ASSOCIATIONS AND
MINERALISATION STYLES
Associations
Three distinct gold-metal associations occur throughout the
deposit as defined by Reynolds (2000). The gold is most
commonly intergrown with sulfides with the majority of gold
grains <25 m in size and the remainder generally >60 m. As
the system evolved earlier metallic gold was continuously
re-mobilised by base metal bearing fluids and re-precipitated as
increasingly silver rich electrum.
Early gold-bismuth
This association represents the earliest known and consists of
fine Ag poor metallic gold generally <25 m, with native
bismuth and minor maldonite (Au2Bi) in fine-grained quartz
veins. The absence of bi-sulfide indicates low sulfur fugacities
during this stage (Reynolds, 2000). Due to later overprinting
stages this association is relatively minor and only occurs in
remnant patches.

342 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE PERSEVERANCE GOLD DEPOSIT THE NEXT STEP AT PEAK
9500 mE
DNG
Ryolite/shear intersection
> 10 g/t (Au + Cu equiv)
> 5 g/t (Au + Cu equiv)
>1 g/t (Au + Cu equiv)
sandstone/shear
intersection
FIG 3 - North/South long section though Perseverance deposit. DNG-massive laminated shale,
DNC2-sandstone dominated turbidite, DNC31-siltstone dominated turbidite.
Gold-silver-copper-iron
This is the dominant style of mineralisation and occurs within
and around coarser-grained quartz veins than the earlier
gold-bismuth association. Gold occurs as electrum with 30 - 50
per cent Ag. Chalcopyrite with minor cubanite as lamellar
intergrowths and pyrrhotite are the main sulfide minerals.
Late gold-silver-lead
This is a relatively minor association best developed with late
black Mg-rich chlorite alteration. A small proportion of gold
occurs in Agrich (>60 per cent Ag) electrum intergrown with
galena and minor sphalerite.
Mineralisation style
Several mineralisation styles occur throughout Perseverance. A
marked gradation in style is evident in the southern plunge
direction from predominantly shear hosted to vein hosted
mineralisation at depth. The lenses can be divided into three
groups according to the style of mineralisation.
1. Sediment shear hosted strongly silicified and brecciated
sediments with foliation parallel quartz sulfide veins
below the rhyolite intersection with Perseverance shear.
Generally contains lower sulfide contents relative to
mineralisation associated with rhyolite. There is no
sediment hosted mineralisation above the rhyolite
intersection with the Perseverance shear.
2. Contact shear hosted intensely silicified and brecciated
contact between the rhyolite and sediments that forms the
loci for the highest gold and base metal grades. The
alteration intensity is such that recognition of primary rock
textures is impossible. This style is restricted to the higher
levels of the deposit.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
10000 mE
9500 mRL
DNC2
9000 mRL
DNC31
3. Rhyolite vein hosted appears as a series of narrow
(10 cm) quartz-chlorite-chalcopyrite-pyrrhotite veins within
silicified and weakly chlorite altered incoherent rhyolite
breccia. It forms the dominant style of mineralisation in the
deeper reaches of the deposit where the veins follow the
lithological rhyolite contact adjacent to the Perseverance
shear. It also occurs within the eastern rhyolite lenses
footwall to the Perseverance shear.
CONTROL ON LOCATION OF MINERALISATION
There are three significant structural features that control the
location of mineralisation at Perseverance. The Primary control
is the Perseverance Shear. The shear provided the pathway for
ore-metal bearing fluids to ascend from depth.
The secondary control is the intersection of the Perseverance
Shear with the rhyolite. The rhyolite bodies display significant
competency contrasts with the enclosing sediments. When
intersected by a major shear-zone, the rhyolite causes refraction
of the shear and with subsequent movement becomes a locus for
dilation. Above the rhyolite/Shear intersection, the Perseverance
shear tightens considerably in shale-dominated lithologies. This
feature further enhances the effectiveness of the rhyolite/shear
intersection to act as a trap for ascending fluids.
The third control is with the location of anti-formal flexures
in the trace of the rhyolite against the Perseverance shear (see
Figure 3). The flexures are either the intersection of the
Perseverance shear with smaller parasitic folds in the rhyolite or
natural variations in the contacts of the rhyolite. These flexures
appear to have added an extra dimension to the entrapment of
ascending fluids whereby they have been focussed into what
appear to be the hinge areas of smaller parasitic anti-formal
features within the rhyolite.
343
S MUNRO and R BERTHELSEN
A fourth, though minor control is the distribution of
metamorphic facies changes. The presence of metamorphic
biotite replacing chlorite at depth suggests that as the early Au
bearing fluids ascended into a lower grade metamorphic
environment ideal temp and pressure conditions were realised for
the precipitation of Au. This suggestion requires further work but
may also be a contributing factor to location of mineralisation.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the permission of Peak Gold
Mines and Wheaton Minerals Asia Pacific to release the
information and ideas contained within this paper. The
Perseverance project has seen significant input from many
geologists since discovery. Peter Leah and Mark Dufty are
credited with the discovery of the deposit in 1994 while Peak
Gold Mines was owned by CRA-Riotinto. Since then, geological
contributions are also acknowledged from Michael Priestly and
Kai Dettbarn.
REFERENCES
Dettbarn, K, 2001. Perseverance Zone A Project Study, Nov 2001,
Geology and Resource Estimation Study, PGM Report No 1686.
Glen, R A, 1987a. Geology of the Wrightville 1:100 000 sheet 8034,
Geological Survey of New South Wales (Department of Mineral
Resources).
Glen, R A, 1987b. Copper and Gold deposits in deformed turbidites at
Cobar, Australia: Their structural control and hydrothermal origin,
Economic Geology, 82:124-140.
344 Adelaide, SA, 19 - 22 September 2004
Glen, R A, 1991. Inverted trans-tensional basin setting for gold and
basemetal deposits at Cobar, New South Wales, BMR Journal of
Geology and Geophysics, 120:13-24.
Glen, R A, 1992. Thrust, extension and strike slip tectonics in an evolving
Palaeozoic orogen a synthesis of the Lachlan Orogen of
south-eastern Australia, Tectonophysics, 214:341-380.
Glen, R A, 1995. Thrusts and thrust-associated mineralisation in the
Lachlan Orogen, Economic Geology, 90:1402-1429.
Glen, R A, Clare, A and Spencer, R, 1996. Extrapolation of the Cobar
Basin model to the regional scale: Devonian basin formation and
inversion in western New South Wales, in The Cobar Mineral Field
A 1996 Perspective (Eds: W G Cook, A J H Ford, J J McDermott, P
N Standish, C L Dilbert and T M Stegman), pp 43-83 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
Mutton, A, 1999. Peak Mine and Perseverance Area: Review of DHEM
surveys, PGM Report No 1173.
Perkins, C, Hinman, M C and Walshe, J L, 1994. Timing of
mineralisation and deformation, Peak Au mine, Cobar, NSW,
Australian Journal of Earth Sciences, 41:509-522.
Reynolds, I M, 1999. Mineralogy and ore petrography of the suite of
samples from the Perseverance Gold Deposit, Cobar, PGM Report
No 1105.
Reynolds, I M, 2000. Mineralogy and ore petrography of a second suite
of samples from the Perseverance Gold Deposit, Cobar, New South
Wales, Australia, PGM Report No 1426.
Sweeney, M, 2000. Perseverance Deposit Preliminary resource
estimates May 2000, PGM Report No 1617.
PACRIM 2004
Geology, Mining and Grade Control at the Batu Hijau Porphyry
Copper-Gold Deposits, Sumbawa, Indonesia
A Prihananto1

ABSTRACT

Batu Hijau is an island arc porphyry copper-gold deposit located in the southwest corner of Sumbawa
Island in the Banda archipelago of Indonesia. Southwest Sumbawa Island is underlain by Early
Tertiary, low-K calc-alkaline to weakly alkaline volcanic and intrusive rocks. Within the mine area
the oldest exposed rocks are thick-bedded, andesitic tuff breccia, thin bedded, fine-grained
sandstones and mudstones and hypabyssal andesite porphyry. In the eastern portion of the mine area
there is a thick sequence of quartz diorite that postdates the andesites.
Copper and gold mineralisation is directly related to quartz veining and wall rock alteration that
developed with multiple tonalite porphyry intrusions (Clode, 1999). The tonalites were emplaced
along the contact between the andesitic volcaniclastic rocks and quartz diorite. Copper minerals
consist of chalcopyrite, bornite and lesser of digenite mostly associated in the biotite alteration and in
the transition zone of overprinting chlorite-sericite. Most of native gold grains occur within the quartz
veins, attached to sulfide or as free gold along quartz or silicate grain boundaries (Arif, 2003). The
native gold grains are dominantly round in shape and mostly 2 - 12 m in size. The majority of gold
was deposited during the formation of the early A veins and dominantly associated with bornite
rather than chalcopyrite.
Batu Hijau deposit has been exploited since 1998, while the concentrator was started up a year
after that. Resource estimation was calculated based on 152 000 metres of drilling containing
137 863 copper assays to October 2003. Drill spacing is typically 50 50 m in the central ore and
100 100 m in the marginal zone. A total of 765 million metric tonnes grading 0.578 per cent copper
(9.75 billion pounds) and 0.478 g/t gold (11.755 million ounces) was reported as Measured and
Indicated resources until the LOM.
The deposit is mined by open pit methods using P&H 4100, 47 cubic metre capacity electric
shovels and 240 tonne haulage trucks for delivery. The current mine plan is scheduled for a
production rate of 600 Ktonnes material moved and about 135 Ktonnes ore milled per day. Mining is
monitored by Dispatch system on 24 hours continuous operation.
Grade control plays an important role in optimising mill production. Ore definition is determined
using 7 8 m blasthole assay data to generate a dollar revenue block model. Polygonal ore
boundaries are delineated based on certain geologic features of each area. This provides some options
to allow the engineer or operator blending ores to meet particular requirements of the process plant.
The understanding of geology and the ore characteristic revealed by mapping becomes a crucial issue
in the management of production.

1. Mine Geology Department, Batu Hijau Mine, PT Newmont Nusa Tenggara, Jl Sriwijaya 258 Mataram, Lombok NTB 83125, Indonesia.
E-mail: apri0946@corp.newmont.com

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 345

The Bong Mieu Gold Project, Phuoc Son Tam Ky Suture,
Central Vietnam
N N Quynh1, R H Murfitt2, T Sirinawin3 and W Shywolup4
ABSTRACT
The Bong Mieu Project lies within the Pre-Palaeozoic Kontum Craton
near the southern margin of the regional Phuoc Son Tam Ky Suture.
The Suture and accompanying Truongson Fold Belt of Indosinian age
host significant gold and base-metal bearing districts including Olympus
other project at Phuoc Son to the west of Bong Mieu and Oxianas Sepon
project in Laos.
The Bong Mieu Project is located some 90 km south of Da Nang and
some 30 km south-west of Tam Ky, the provincial capital of Quang Nam
province in central Vietnam. The project area is covered by an Investment
Licence of some 30 km2 which was granted in 1991 and a mining permit
of some 365 ha granted in 1992.
Several mesothermal, gold-silver deposits with assigned resources are
present on the property. Deposits with assigned resources contain some
615 000 ounces of gold in situ. These are at Ho Gan and Ho Ray which
are potentially exploitable by open-pit mining and Nui Kem which is
potentially exploitable by underground mining. Newly discovered
extensions to the known deposits at Ho Gan, Ho Ray and Nui Kem, and
several other new prospects, are currently being investigated by drilling.
The Ho Gan mineralisation is of mesothermal, shear and shear breccia
quartz vein, as well as intrusive breccia hosted style. Nui Kem
mineralisation is of mesothermal, shear and shear breccia quartz vein
style. The Ho Ray mineralisation is of skarn (gold-bismuth-tungsten) and
lesser quartz vein style. The deposits straddle and flank the east-west
trending Bong Mieu Anticline which trends through the central portion of
the Investment Licence.
In this paper the geology of the deposits, details of resources, results of
recent exploration on the Investment Licence and recent engineering and
mining studies at Ho Gan are briefly reviewed.
TENURE, LOCATION AND ACCESS
Investment Licence 140/GP (IL140/GP) covering 30 km2 issued
on 5 March 1991 and Mining Permit ML582/CNNG/KTM
covering some 365 ha issued on 22 July 1992 comprise Olympus
Pacific Minerals Incs (Olympus) Bong Mieu Project.
The tenements are located within the Tam Lanh Commune,
Tien Phuoc District, some 20 km south-east and inland of Tam
Ky, the Quang Nam provincial capital, central Vietnam. Tam Ky
is some 60 km south of Da Nang along Highway 1.
REGIONAL GEOLOGY AND STRUCTURE
The Bong Mieu Project is situated some 20 km south of one of
the main branches of the Phuoc Son Tam Ky Suture, which is
probably one of the most important structural controls of gold
metallogeny in central Vietnam. The Phuoc Son Suture
structurally juxtaposes Archaean to Proterozoic high-grade
crystalline metamorphics of the Kontum Craton against a
Palaeozoic sequence of lower grade metamorphics of the Truong
Son Geosyncline. This regional structure trends from the South
China Sea near Tam Ky for almost 100 km to Kham Duc, where
it is transected by a series of complex cross folds and faults. It
1. Bong Mieu Gold Mining Company Ltd, 113/2 Nguyen Chi Thanh
Street, Da Nang, Vietnam.
2. MAusIMM, Olympus Pacific Minerals Inc, Level 3, Bridgecorp
House, 36 Kitchener Street, Auckland, New Zealand.
3. Consulting Geologist, 68/76 Pradipat 25, Bangkok 10400, Thailand.
4. Consulting Geologist, PO Box 3612, Hermit Park Qld 4812.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
then trends northwest for a further 100 km into and across Laos
(Wallis and Ward, 1997). Regional east-west foliation and
extensive shearing are associated with the Phuoc Son Suture. The
Suture and accompanying Truong Son Fold Belt of Indosinian
age host significant gold and base-metal bearing districts
including Olympus other project at Phuoc Son (near Kham Duc)
to the west of Bong Mieu and Oxianas Sepon project in Laos
(Figure 1).
The Kontum Craton comprises Archaean metamorphic rocks
to granulite facies, Proterozoic rocks of middle amphibolite
facies and intrusive complexes that range in composition from
mafic to felsic and in age from Archaean to Late Triassic/Early
Jurassic. The geosynclinal rocks of the Truong Son syncline
comprise mainly sericite, chlorite and graphite schist and
intercalated greenstone and carbonate rocks and host numerous
intrusive phases, including hornblende diorite and quartz diorite
plutons and sills. South of the Phuoc Son Suture and trending
east-west through the central part of IL140/GP is the axis of the
Bong Mieu anticline. This gently open fold, which in the licence
area plunges to the east then south-east, is also intermittently
traceable to the west almost to Kham Duc.
PROPERTY GEOLOGY AND STRUCTURE
The east-west trending Bong Mieu anticline is an open fold with
shallow dipping limbs. Metamorphic foliation parallels the
metamorphic (and possibly the original) compositional layering.
This foliation has a shallow dip of between 5 and 45 on the
southern flank of the anticline and a slightly steeper dip on the
northern flank of the anticline.
The Proterozoic Kham Duc Formation underlies the area
covered by IL140/GP. The rocks are crystalline and of middle
amphibolite facies. The eroded core of the Bong Mieu anticline
comprises biotite-sillimanite gneiss and structurally (and/or
stratigraphically) above this lithology is a sericite-biotite schist.
Lithologies higher in the sequence comprise of quartz-feldspar
-biotite schist and gneiss and these cover a large part of the
licence area and are particularly prevalent on the southern flank
of the Bong Mieu anticline as at the Nui Kem to Suoi Tre area.
Intercalated with the quartz-feldspar-biotite schist at higher
levels are amphibolite schist units.
A package of lithologies compositionally different to those
already mentioned occurs on the northern flank of the anticline at
the Ho Ray gold deposit. These lithologies comprise
biotite-garnet schist, calc-silicate, quartzite and amphibolite.
Additionally augen gneiss, foliated granite sheets, pegmatite and
quartz veins are interlayered with and cross-cut the metamorphic
stratigraphy. Granite, granite gneiss and pegmatite are also
present in the core area of the anticline at the Ho Gan deposit,
another of the gold deposits.
The lithologies of the licence area are transected by numerous
prominent northwest-southeast and northeast-southwest trending
faults.
DEPOSIT TYPES
Gold deposits with ascribed resources occur at three localities on
IL140/GP, namely Ho Gan, Ho Ray (both potentially exploitable
by open-pit mining) and Nui Kem (potentially exploitable by
underground mining). Additionally there are a number of other
347
N N QUYNH et al

prospects which have been explored to various levels of detail. deposit (Areas 7 and 8, 11 and 1 and 2) and prospects Thac
These prospects are Bong Mieu West, Bong Mieu North, Ho Ray Trang and Rung De (Ho Ray Extensions), Ho Gan West (Ho Gan
West, Thac Trang, Rung De, Nui Kem North and South, Saro Extensions) and Nui Kem North (Nui Kem Extensions) have
Hill, Nha Tung and Suoi Tre (Figure 2). Part of the Ho Gan been recently investigated by diamond core drilling.

FIG 1 - Bong Mieu location with regional geology and structure.

FIG 2 - Investment licence 140GP_Prospects.

348 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE BONG MIEU GOLD PROJECT, PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM
The gold deposits and prospects lie within the core and on
both limbs of the Bong Mieu anticline. Mineralisation generally
occurs within shears that host quartz+sulfide veins, brecciated
quartz+sulfide+schist and/or sulfide bearing schist. The shears
have developed within and dip parallel to the metamorphic
foliation. They occur as parallel stacked structures in some areas
and have been traced for up to 2 km laterally and/or down-dip
(eg as at Nui Kem and Ho Gan). Recent work at Ho Gan work
has also identified an intrusive sill that carries sheeted veins, is
greisenised, contains miariolitic cavities, USD (brain rock)
textures and disseminated sulfides.
Mineralogy of the gold bearing veins comprises
quartz+pyrite+calcite with accessory graphite, galena, haematite,
magnetite, chalcopyrite, sphalerite, pyrrhotite, molybdenite and
cassiterite. Higher grades occur within shoots which are controlled
by the intersection with other shears/faults. Veins and shoots are
accompanied by silica, chlorite and sericite alteration. The only
age date available for the mineralisation is a galena derived age of
200 Ma from Nui Kem (Dzung, 1988 in Baxter et al, 1992).
The Ho Ray and Thac Trang mineralisation occurs within
calc-silicate lithologies and underlying quartzite, biotite schist
and gneiss. The mineralisation has an apparent stratabound
distribution that favours these lithologies that have a gentle to
moderate stratigraphic and metamorphic northerly dip. The gold
occurs as free grain disseminations within magnesium dolomite,
and is more often associated with pyrrhotite+pyrite+galena
+chalcopyrite+bismuthinite+scheelite. Alteration mineralogy
comprises biotite, actinolite, chlorite and carbonate. Recent
petrology from Ho Ray has identified both endo- and exo-skarn
altered lithologies. USD textures have also been identified in
hand specimen samples.
The style of the mineralisation on the Bong Mieu licence area
appears to indicate emplacement at mesothermal depths. It is
considered that hydrothermal fluids driven by intrusives provided
the mineralisation within the shear and breccia hosted
mineralisation at Ho Gan, within the shear hosted mineralisation
at Nui Kem and other areas, and also within the calc-silicates at
Ho Ray and Thac Trang.
FIG 3 - Ho Gan deposit areal extent.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
While direct analogues for Olympus Bong Mieu deposits are
not readily apparent, it is worthy of note that on a worldwide
scale many, if not the majority, of significant gold deposits are
associated with major crustal structures such as the Phuoc Son
Suture Zone, which lies close to the deposits. Based on the
observed geological setting, styles of alteration and
mineralisation at Bong Mieu an Intrusive Related Gold
environment is suggested.
DEPOSITS
Geology and mineralisation
Ho Gan deposit
The Ho Gan gold deposit area (worked since historical times by
Cham miners and later referred to by the French as Bong Mieu or
Champ DOr) comprises the original prospects Ho Gan, Xa Kok
Sau, Suoi Trang and Tru Sat. The deposit is located some 2 km
south of the Bong Mieu camp and offices on the southern side of
the Bong Mieu River.
The Ho Gan deposit is underlain by a gently west dipping
sequence of schist and granitoid gneiss intruded by granite and
pegmatite within the core area of the Bong Mieu anticline. The
gold mineralisation is located within several closely stacked
shallow dipping shears and shear-associated breccia zones as
well as intrusive breccia and granitic intrusive sill within the
metamorphic rocks. The shear, breccia zones and intrusive sill
occur within and dip parallel to the metamorphic foliation. The
shear structures have lateral extent to some 2 km highlighted by
the 0.5 g/t Au contour of the deposit (Figure 3). The shears host
quartz+sulfide veins (Figure 4), brecciated quartz+sulfide+schist
(Figure 5), and/or sulfide bearing schist (occasionally oxidised
and carrying limonite/goethite). The most common sulfide is
pyrite, and galena, arsenopyrite and pyrrhotite also occur.
349
N N QUYNH et al
FIG 4 - Ho Gan quartz sulfide veins.
FIG 5 - Ho Gan schist quartz sulfide breccia.
The western part of the deposit is intruded by a granite sill and
the gold mineralisation is hosted within an intrusive related
breccia (Figure 6) and the altered intrusive. The intrusive carries
sheeted veins (Figure 7), is greisenised, contains miariolitic
cavities, USD (brain rock) textures and disseminated sulfides.
The style of the mineralisation at Ho Gan appears to indicate
emplacement at mesothermal depths.
Much of the deposit outcrops at surface or occurs shallowly
beneath the surface and varies in thickness from a few
centimetres to 12 m. Previous mining by the Cham, French and
Vietnamese produced substantial thickness of mullock over parts
of the deposit. The deposit mineralised shears and metamorphic
foliation subparallel the present day topography which is also
dissected by numerous gullies and steep valleys. The near
parallelism of slope and mineralisation necessitates accurate
surveyed topography for geological and grade correlation, and
resource estimation.
Ho Ray deposit
The Ho Ray deposit area and the adjacent Thac Trang prospect
to the east are located on the northern flank of the Bong Mieu
anticline (Figure 8). Metamorphic foliation is slightly steeper
than on the southern limb of the anticline and varies from 10 to
40 and averages 30 to the north and northeast (Figure 9).
The area is underlain by lithologies seen elsewhere on the
licence area, namely quartz+mica schist, gneiss, granite and
pegmatite. Additionally the area hosts a package of lithologies
with composition markedly different to those seen elsewhere on
the licence area. These comprise biotite+garnet schist,
calc-silicate, quartzite and amphibolite.
350 Adelaide, SA, 19 - 22 September 2004
FIG 6 - Ho Gan greisenised granite breccia.
FIG 7 - Ho Gan sheeted quartz sulfide veins in granite.
The calc-silicate and enclosing lithologies have been
previously separated into four metamorphic stratigraphic units
which likely reflect the original composition of the sequence.
These units were described from the base up as follows:
Unit 5 a coarsely porphyritic quartz+feldspar gneiss,
commonly brown coloured due to abundant garnet,
interbanded with sericite schist, biotite schist and minor
amphibolite. This unit is not gold mineralised.
Unit 6 a mixed amphibolite (sometimes altered to
phlogopite schist) grading to biotite schist, and quartz+
feldspar gneiss grading to quartzite. This unit contains
approximately half the gold resource.
Unit 7 a massive mottled light green to dark green skarn
with alteration mineralogy being hornblende+clinopyroxene
+fluorite +sphene skarn. The unit also comprises irregular
quartz+feldspar cavity fills, quartzite and dolomite. Recent
petrology from Ho Ray has identified both endo- and
exo-skarn altered lithologies. USD textures have also been
identified in hand specimen samples. This unit contains the
remainder of the gold resource. It is strongly anomalous in
fluorine and carries disseminated and cavity fill scheelite.
PACRIM 2004
 THE BONG MIEU GOLD PROJECT, PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM

FIG 8 - Ho Ray deposit and extensions area.

FIG 9 - Ho Ray deposit cross-section 10150E.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 351

N N QUYNH et al
Unit 8 a maroon schist with quartz+feldpsar porphyroblasts
and common garnet, cm-scale interbands of green skarn,
gneissic bands, and dolomite. The unit commonly carries
anomalous Au values which range 0.05 g/t to 0.15 g/t Au.
Irregularly occurring within the above stratigraphy is
porphyroblastic quartz+feldspar mafic gneiss which has been
interpreted as migmatitic and intrusive into the stratigraphic
units.
Faults, some represented by filled breccias, and mylonitic
zones of various ages both transgress and parallel the
metamorphic foliation or stratigraphy.
The gold+sulfide mineralisation within Units 6 and 7 occurs as
disseminated grains and aggregates along foliation (Figure 10),
as fill accumulations in fractures within brittle lithologies such as
quartz veins (Figure 10), and as fill in fractures termed spider
veins (Figure 11). The spider veins occur at lithological
contacts where a rock competency contrast is evident. Some,
though not predominant, mineralisation occurs within hairline
fractures that cut the foliation at a shallow angle to the core axis.
FIG 10 - Ho Ray Unit 6 in core.
FIG 11 - Ho Ray spider veins in core.
The gold is often associated with pyrrhotite+pyrite+galena+
chalcopyrite+bismuthinite+scheelite+cassiterite+molybdenite.
Tungsten mineralisation as scheelite occurs as fine particles
within the metamorphic foliation and as coarse grains parallel to
some of the interpreted steeply dipping structures (Laubsch et al,
1996). Alteration mineralogy comprises biotite, chlorite and
carbonate.
352 Adelaide, SA, 19 - 22 September 2004
Nui Kem deposit
The Nui Kem (Zinc Mountain) Mine is a former underground
mine developed by French engineers and operated during the
period 1896 to 1942. More than 30 adits access some 18 000 m
of drives and crosscuts on 15 levels between 200 and 400 m RL
(Figure 12). Previous production by the French was estimated to
be 0.42 Mt @ 5.6 g/t Au (recovered) and 1.5 g/t Ag (recovered)
(Wallis and Ward, 1997).
The Nui Kem area incorporates the Saro Hill prospect and is
located on the southern limb of the Bong Mieu anticline. The
underlying geology comprises quartz+feldspar+biotite schist and
gneiss. Metamorphic foliation dips variably but gently generally
to the south at between 5 and 45. The gold mineralisation is
located within quartz+sulfide veins that occupy shears and that
have been worked over a strike length of some 2 km and down
dip for at least 350 m (Figure 12). The common sulfide minerals
are pyrite, galena, sphalerite, lesser pyrrhotite and chalcopyrite
(observed in oxidising crushed rock to occur more towards the
eastern part of the area).
The veins and lodes strike at about 058 and they vary in
thickness from 0 to 2.2 m and average 1.0 m. The shoots plunge
at about 019 towards 176 and the shoots within these lodes
vary in thickness from 0 to 3.2 m and average 1.0 m (Baxter et
al, 1992).
Several major north to north-northwest trending faults crosscut
the main underground workings. These faults from west to east
are the AD2, Tunnel, Mullock and French Office faults. Previous
structural analysis of measurements indicated that these oblique
slip faults have relative dextral movement. Horizontal
displacement on the respective faults was estimated as: AD2
fault 230 m, Tunnel fault some 25 m, Mullock fault 90 m,
and the French Office fault 35 m (Chisholm and Baxter, 1993).
The faults are sometimes mineralised, the most notable
example being the Pyrite Stope mineralisation within the
Mullock fault.
Resources
Estimates of the resources at Ho Gan, Ho Ray and Nui Kem were
completed as part of a pre-feasibility study compiled in 1996 by
Continental Resource Management Pty Ltd acting as independent
resource consultants (Anon, 1996). These estimates have been
reviewed by Watts Griffis McOuat of Canada in 1997 and were
found to have been carried out in a manner consistent with
standard practice in the industry. The results of these estimates are
listed in Table 1.
RECENT EXPLORATION AND RESULTS
2003 program
In May 2003 exploration resumed on the Bong Mieu Project.
The previous magnetic survey was reprocessed and interpreted
and reconnaissance mapping and sampling carried out to
investigate magnetic anomalies and certain soil geochemical
anomalies outlined in past programs. The 2003 reconnaissance
exploration program identified several areas as meriting
follow-up work. The most important were Ho Ray East/Thac
Trang and Rung De (Ho Ray Extensions), Ho Gan West (Ho Gan
Extensions) and Nui Kem North and Nui Kem South (Nui Kem
Extensions). Bong Mieu West and Nha Tung East also show
promise but are at an earlier stage of investigation. Ho Ray West
and Bong Mieu North were drilled in the mid-1990s and are
being reassessed as is the Suoi Tre prospect to the south-west of
Nui Kem.
PACRIM 2004
 THE BONG MIEU GOLD PROJECT, PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM

FIG 12 - Nui Kem.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 353

N N QUYNH et al

with exo- and endo-skarn rocks being identified. The

TABLE 1
mineralisation at Ho Gan was stated to be quartz-sulfide, low
Summary of Bong Mieu gold mineral resource estimates.
temperature, mesothermal style (emplaced into sheared and
Category/deposit Tonnes Gold Contained brecciated zones), likely from reduced fluids. A greisenised,
(t) grade (g/t) gold (oz) sheeted quartz veined granite carrying disseminated pyrite was
Measured and indicated resources identified from Ho Gan. Elsewhere, at Thac Trang, Rung De and
Ho Gan1 687 000 1.98 43 700
Bong Mieu West, calc-silicate lithologies were described and
from the northern part of the property various intrusive rocks
Ho Ray2 878 000 1.90 53 600 (tonalites to granodiorite) were also described.
Nui Kem3 216 700 6.51 45 300
Measured and indicated 1 565 000 1.94 97 300 Ho Gan deposit
Resources and and and
216 700 6.51 45 300 Additionally 50 rock samples (ranging from approximately 7 kg to
Total 1 781 700 142 600
47 kg) were collected from the Ho Gan Deposit for metallurgical
studies and grade substantiation. Average assay values were
Additional inferred resources 2.93 g/t Au (individual assays cut to 10.00 g/t Au) and 4.73 g/t Ag.
Ho Gan1 1 280 000 1.82 74 800 A portion of the material collected was sent to SGS NZ Ltd
Ho Ray2 1 380 000 1.80 79 800 (SGS) in Auckland, New Zealand in December 2003 for column
Nui Kem3 1 220 000 8.10 317 300 leach testing as an indication of amenability to heap leaching.
Initial bottle roll tests showed generally only moderate recoveries
Additional inferred resources 2 660 000 1.81 154 600
and and and and planned column-leach testwork was suspended in March
1 220 000 8.10 317 300 2004.
Total 3 880 000 471 000 Fifty kilograms of material, 100 per cent passing 25 mm, from
the Ho Gan Area 7 and 8 was despatched to Gekko Systems Pty
Ltd, Ballarat, Victoria, Australia for gravity and gravity in
Inferred resources are in addition to measured and indicated resources.
conjunction with flotation recovery testwork. This testwork
Notes: indicated that the material is amenable to gravity in conjunction
1. Employed a lower 0.8 g/t grade cut-off and 10 g/t grade upper cut-off. with flotation separation (93 per cent recovery) followed by gold
extraction using intense cyanidation conditions (in-line leach
2. Employed a lower 0.8 g/t grade cut-off and 10 g/t grade upper cut-off. reactor 98.4 per cent recovery) with an overall resultant gold
3. Employed a lower 3.0 g/t grade cut-off and 30 g/t grade upper cut-off. recovery of 91.5 per cent.
4. This resource was estimated in compliance with the former National
Policy 2A standard. Ho Ray deposit
The technical report to accompany the resource estimate was the
pre-feasibility study compiled in 1996 by Continental Resource Six pits or shafts, varying in depth from 7 m to 16 m were
Management Pty Ltd (CRM), acting as independent resource excavated in order to obtain samples of oxide and sulfide
consultants. These estimates have been reviewed by Watts Griffis material for metallurgical testing. Twenty-four channel or chip
McOuat (WGM) in 1997 and were found to have been carried out samples were sent assay. Assays for oxide material averaged
in a manner consistent with standard practice in the industry. This 1.50 g/t Au and for sulfide material averaged 2.54 g/t Au.
historical estimate is relevant as it is the most recent estimate
A portion of the material collected was also sent to SGS in
completed for the project. The reliability of the estimate is
compatible with the standards set by National Instrument 43-101. December 2003 for column leach testing however initial bottle
roll tests showed generally poor recoveries and planned
column-leach testwork was suspended in March 2004.
During the 2003 program 145 rock chip and float samples for
geochemical analysis and 36 petrological samples were 2004 program
collected. Forty-nine of the sample assays returned values of
0.1 Au/t. Samples from the Ho Ray Extensions extended the Ho Gan deposit
identified occurrence of auriferous and/or skarn altered
calc-silicate lithologies over 500 m of strike length in the Thac Engineering, metallurgical and environmental studies to
Trang area and this trend remains open to the southeast. Surface determine the feasibility of developing an open pit at the Ho Gan
chip sample assays ranged to 34.67 g/t Au. Cal-silicate 7 and 8 zone commenced and are continuing. Preliminary
lithologies anomalous in gold (to 1.21 g/t) were also found over scoping based on possible start-up in early 2005 at an initial
an intermittent strike length of some 900 m in the Rung De area mining rate of 500 tpd for three years with subsequent mining
though this is much less well defined compared to Thac Trang. rates stepped up to 800 tpd indicates a likely good return on
Anomalous gold mineralisation (to 4.36 g/t) was identified to the investment. Ore would initially be drawn from the Ho Gan
southwest of the Ho Gan deposit suggesting a possible extension deposit thence would be also sourced from other deposits already
of the Ho Gan mineralisation (Ho Gan Extensions). The samples identified and those to be fully delineated following more
from the Nui Kem Extensions identified auriferous quartz vein detailed studies.
and gossan material in float and outcrop. Surface rock chip and A diamond drill program was completed in April 2004 on Ho
float sample assays ranged to 19.99 g/t Au. Gan 7 and 8. This consisted of 34 PQ core size holes totalling
The petrological results of the metamorphic rocks that underlie 520.50 m (Figure 13) and was designed to provide infill
the majority of the project area indicated these had undergone information at a closer drill spacing as well as test the extremities
intense metamorphism at high temperatures (>710C) and some of the shallow, flat lying mineralised zone. The holes were
considerable depth (8.2 kb), being the top of the amphibolite or drilled on approximately 25 m centres. The program confirmed
bottom of the granulite facies, near the base of normal-thickness the higher grade nature of the Ho Gan 7 and 8 zone and was
crust. The petrological results of samples from Ho Ray indicated successful in establishing the continuity of mineralisation
that the mineralisation at this deposit is clearly Au-Bi skarn type, between the Ho Gan 7 and Ho Gan 8 zones over an area of some

354 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE BONG MIEU GOLD PROJECT, PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM

FIG 13 - Ho Gan areas 7 and 8 drill plan.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 355

N N QUYNH et al

TABLE 2
Ho Gan 7/8 area in-fill drill program results.
Hole ID From (m) To (m) Core width (m) True width (m) True width (Ft) Gold grade (g/t)
52 0.0 1.0 1.0 0.9 3.1 0.70
53 0.0 7.0 7.0 6.6 21.6 2.79
including 0.0 1.0 1.0 0.9 3.1 3.34
and 4.5 7.0 2.5 2.4 7.7 5.27
54 0.0 5.0 5.0 4.7 15.4 2.32
including 0.0 2.0 2.0 1.9 6.2 4.66
55 0.0 1.5 1.5 1.4 4.6 2.76
56 0.0 3.0 3.0 2.8 9.3 1.15
57 0.0 4.0 4.0 3.8 12.3 2.10
including 0.0 0.7 0.7 0.7 2.2 3.79
58 0.0 2.0 2.0 1.9 6.2 1.59
5.0 6.0 1.0 0.9 3.1 2.40
59 2.0 4.0 2.0 1.9 6.2 1.39
60 0.0 2.0 2.0 1.9 6.2 1.46
8.0 11.0 3.0 2.8 9.3 6.89
61 0.0 2.0 2.0 1.9 6.2 2.15
9.0 10.4 1.4 1.3 4.3 2.08
62 0.0 1.0 1.0 0.9 3.1 1.46
6.0 8.4 2.4 2.3 7.4 1.05
63 1.0 2.0 1.0 0.9 3.1 1.12
64 0.0 2.0 2.0 1.9 6.2 2.46
65 No Significant Values
66 2.1 4.2 2.1 2.0 6.5 1.63
67 0.0 3.0 3.0 2.8 9.3 2.65
including 0.0 1.0 1.0 0.9 3.1 3.77
and 2.3 3.0 0.7 0.7 2.2 4.43
68 No Significant Values
69 No Significant Values
70 2.8 3.4 0.6 0.6 1.9 2.18
71 No Significant Values
72 0.0 1.0 1.0 0.9 3.1 1.13
73 0.0 1.7 1.7 1.6 5.2 1.95
74 0.0 4.0 4.0 3.8 12.3 1.58
including 3.0 4.0 1.0 0.9 3.1 3.43
75 5.0 7.0 2.0 1.9 6.2 4.68
including 6.0 7.0 1.0 0.9 3.1 8.33
76 0.0 4.0 4.0 3.8 12.3 1.98
including 0.0 1.0 1.0 0.9 3.1 4.16
77 0.0 4.0 4.0 3.8 12.3 3.82
78 0.0 1.0 1.0 0.9 3.1 1.11
79 0.0 0.7 0.7 0.7 2.2 0.59
80 0.0 1.0 1.0 0.9 3.1 3.43
81 0.0 4.3 4.3 4.0 13.3 1.54
including 0.0 2.0 2.0 1.9 6.2 2.49
7.0 8.0 1.0 0.9 3.1 2.39
82 No Significant Values
83 0.0 12.0 12.0 11.3 37.0 2.87
including 0.0 5.0 5.0 4.7 15.4 5.61
or 0.0 7.0 7.0 6.6 21.6 4.41
84 0.0 1.0 1.0 0.9 3.1 2.28
85 No Significant Values

Note: All assays performed by the Mineral Assay and Service Co Ltd (MAS Laboratory) located in Bangkok, Thailand using the Fire Assay method
on 50 grams of prepared sample. The MAS Laboratory is certified by the Thailand Department of Industrial Works and Ministry of Industry. All
holes were drilled at an inclination of -90.

356 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE BONG MIEU GOLD PROJECT, PHUOC SON TAM KY SUTURE, CENTRAL VIETNAM

400 m by 150 m. The flat lying, gold bearing zone occurs on

TABLE 3
surface and extends to a known vertical depth of up to 13 m. The
Thac Trang exploration drill program results.
gold mineralisation is hosted within a shear and intrusive related
breccia and altered intrusive rocks. Hole ID From To True width Gold grade
Highlights of the program included intercepts (core widths) of (m) (m) (m) (g/t)
2.87 g/t Au over 12.0 m including 5.61 g/t Au over 5.0 m, HRDD134 0.00 4.00 4.00 2.07
6.89 g/t Au over 3.0 m, 2.79 g/t Au over 7.0 m including 5.27 g/t
and 11.00 16.00 5.000 0.71
Au over 2.5 m and 3.82 g/t Au over 4.0 m from drillholes
HGDD083, HGDD060, HGDD053 and HGDD077 respectively. HRDD135 7.00 11.00 4.00 0.79
Complete results are set out in Table 2. including 10.00 11.00 1.00 1.48
HRDD136 0.00 15.00 15.00 1.67
Other prospects including 0.00 2.00 2.00 8.59
Initial reconnaissance scale mapping and prospecting in late HRDD137 1.00 7.00 6.00 1.28
2003 discovered both quartz vein and skarn style mineralisation HRDD138 0.00 6.70 6.70 1.34
at the Thac Trang prospect area to the southeast of the Ho Ray
including 1.00 3.00 2.00 2.32
deposit (Ho Ray Extensions). Recent trenching and drill road
construction uncovered significant new mineralisation in this and 31.40 32.60 1.20 1.36
area hosted by veined and sulfide bearing skarn-altered HRDD139 18.00 19.00 1.00 0.86
calc-silicate, granite gneiss and schist rocks. Surface continuous HRDD140 15.00 19.00 4.00 1.40
chip samples returned assay results of 2.71 g/t Au over 36 m
HRDD141 0.00 10.00 10.00 0.53
including 4.45 g/t Au over 18 m from a bulldozed exposure
located approximately 400 m to the east of the Ho Ray deposit. including 7.00 8.00 1.00 1.72
A diamond drilling program completed in May 2004 at Thac and 25.00 29.00 4.00 1.34
Trang comprised 15 HQ/NQ core size holes totalling 771.66 m HRDD142 42.00 44.00 2.00 3.50
tested the newly discovered surface gold mineralisation over HRDD143 10.00 15.30 5.30 0.83
some 200 m strike. Highlights of the program included intercepts
and 18.00 24.00 6.00 11.69
(close to true widths) of 11.69 g/t Au over 6.0 m and 3.31 g/t Au
over 4.4 m in HRDD143 and 1.67 g/t Au over 15.0 m in and 26.60 31.00 4.40 3.31
HRDD136 (Figure 14). Complete results are set out in Table 3. including 29.00 30.00 1.00 11.70
Drilling at Ho Gan West was completed in June 2004 and HRDD144 4.00 23.00 19.00 0.69
comprised 12 HQ/NQ core size holes (some of which were including 4.00 6.00 2.00 1.48
drilled for sterilisation and geotechnical purposes as this area is
currently the preferred site for a tailings storage facility once including 13.00 16.00 3.00 1.38
mining development commences) totalling 1051.70 m. including 22.00 23.00 1.00 2.25
Moderately encouraging results included 3.41 g/t Au over 1.9 m HRDD145 3.00 5.00 2.00 0.85
in HGWDD002, 1.76 g/t Au over 1.0 m in HGWDD003, 0.94 g/t and 35.00 38.00 3.00 0.81
Au over 3.8 m and 1.43 g/t Au over 2.0 m in HGWDD004, and
3.14 g/t Au over 1.0 m in HGWDD005. and 42.00 47.20 5.20 2.54
HRDD146 34.40 37.00 2.60 1.18
FUTURE DIRECTION OF THE BONG MIEU HRDD147 40.65 54.00 13.35 0.77
PROJECT including 49.00 50.30 1.30 2.79

Open pit mining development, initially at the Ho Gan deposit, is and 60.00 61.00 1.00 1.71
anticipated to commence in early 2005. Meanwhile, exploration HRDD148 6.00 10.00 4.00 2.25
will continue at other deposits and prospects on the property. including 6.00 6.70 0.70 3.02
This work will largely comprise drilling (plus associated including 9.00 10.00 1.00 4.20
assaying and petrology), resource definition and estimation,
mining, metallurgical, environmental and engineering studies.
Exploration and the other studies will lead towards the larger Notes:
goals of a significantly expanded resource base and production at 1. All the holes were drilled to the west at an azimuth of 270 and
Bong Mieu together with the corporate goal of building Olympus inclined at -60 with the exception of hole HRDD134 which was
into a successful, profitable mining company and a major player drilled at an azimuth of 140/inclination of -60.
in Vietnam and the surrounding area. 2. Based on current geological interpretation of the drilling results, core
widths accurately reflect true widths.
ACKNOWLEDGEMENTS 3. All assays were performed by the Mineral Assay and Service Co Ltd
(MAS Laboratory) located in Bangkok, Thailand using the Fire
The authors would like to thank many individuals, in fact too Assay method on 50 grams of prepared sample. The MAS
many to name, who have worked on the Bong Mieu Project over Laboratory is certified by the Thailand Department of Industrial
the last 15 years. Particular thanks to the current management, Works and Ministry of Industry.
technical and field staff employed by Olympus Pacific Minerals
Inc, namely Roger Dahn, Jean Bailly, Le Van Hai, Henry
Castillo, Duc Hung, Dang Xuan Phu, Mai Ngoc Tuyen, Pham
Dinh De and Tran Dang Khoa and also to the office
administrative staff Mrs Phan My Dung and Mrs Xuan Mai for
their constant support in the exploration effort. Thanks also to
consulting petrologist Richard England.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 357

N N QUYNH et al

FIG 14 - Thac Trang drill section.

REFERENCES Laubsch, N, Baxter, J L and Gans, L, 1996. Review logging of diamond

core from the Ho Ray Prospect Bong Mieu Vietnam, in Continental
Anon, 1996. Bong Mieu gold project prefeasibility study, Lycopodium Resource Management Pty Ltd, Report Number WA96/080,
Pty Ltd, July (unpublished). September (unpublished).
Baxter, J L, Chisholm, J M and Morrison, A D, 1992. Resource Wallis, R and Ward, M-C, 1997. A review of the mineral properties
assessment at 10 June 1992, Nui Kem project, Socialist Republic of Olympus Pacific Minerals in Vietnam (October 1997), in Watts,
Vietnam, in Continental Resource Management Pty Ltd, Report Griffis and McOuat Limited, Toronto (unpublished).
Number WA92/025, June (unpublished).
Chisholm, J M and Baxter, J L, 1993. Bong Mieu Gold Program proposed
exploration program and budget for period ending 31 March 1993,
Socialist Republic of Vietnam, in Continental Resource Management
Pty Ltd, Report Number WA93/001 (unpublished).

358 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

Discovery of the Toguraci Epithermal Au-Ag Deposits,
Gosowong Goldfield, Halmahera Island, East Indonesia
T H Richards1 and M D Basuki Dwi Priyono2

ABSTRACT A number of quartz-adularia epithermal Au/Ag shoots have been

documented to date at Toguraci. Two shoots, Damar and Midas, are now
The Toguraci low sulfidation epithermal gold vein deposits are located 2 being mined.
km WSW from the Gosowong mine and form part of the Gosowong
Goldfield. The goldfield lies within a Neogene magmatic arc on The discovery (and mine permitting) of Toguraci was an exercise in
Halmahera Island, North Maluku Province, East Indonesia. The terrain is persistence. Key exploration tools were: geological observation,
moderate to steep, rainforested, and poorly outcropping. Exploration and large-scale mapping, stream sediment and quartz float sampling with
mining operations are being carried out within the PT Nusa Halmahera follow-up trenching and close spaced drilling. Soil sampling led to the
Minerals Contract of Work (exploration tenement) jointly owned by early recognition of the Damar shoot, however its usefulness could have
Newcrest Mining Ltd (82.5 per cent, operator) and PT Aneka Tambang been enhanced if closer sample spacing along the lines had been
Tbk. (17.5 per cent). employed as the target veins are mostly narrow near surface and show
only minor Au dispersion in soil.
First pass exploration of the Gosowong area began late 1992 resulting
in the discovery of weakly developed porphyry copper mineralisation
(named Bora) at Toguraci in 1993. Early in the evaluation of this INTRODUCTION
porphyry an overprinting epithermal Au/Ag mineralisation event was
recognised, however exploration focus remained on the Bora porphyry The Toguraci epithermal vein deposits are part of the Gosowong
deposit for some time. The discovery of the nearby Gosowong epithermal Operations located on the north arm of Halmahera Island, in the
gold deposit and the exploration of other prospects in the area further North Maluku province, East Indonesia. They are located
diverted attention and funds from Toguraci. It was not until year 2000 that approximately 1 in latitude north of the equator and 2400 km
Toguraci was seriously reevaluated as part of a detailed appraisal of the northeast of the national capital, Jakarta. Access from Jakarta is
epithermal gold potential within a 5 km radius of the Gosowong gold by daily commercial flight to Manado City in North Sulawesi
mine. A fly-camp based drilling campaign using man-portable rigs was (Figure 1). A one-hour private charter flight links Manado to the
undertaken between August 2000 and March 2002. All exploration at Gosowong Operations airstrip.
Toguraci was then stopped pending government approvals to mine. These
were finally issued in May 2003 and immediately drilling restarted and
pre-mine approvals (eg the environmental impact study) were sought and
granted. The first Toguraci gold bar was poured in February 2004.
Exploration at Toguraci and elsewhere in the Gosowong Goldfield
continues today.
The geological setting of Toguraci is well described in Olberg et al
(1999). Epithermal gold mineralisation discoveries in the district to date
are restricted to the Miocene(?) bedded andesite flows and volcaniclastics
of the Gosowong Formation. These are unconformably overlain by a
similar suite of rocks of the Pliocene Kayasa Formation, through which
the Gosowong Formation is exposed as an inlier. In the late Pliocene, the
Gosowong and Kayasa volcanic sequences were locally intruded by
andesite porphyry and quartz diorite. Quaternary pyroclastic flows and
airfall deposits unconformably overlie the Gosowong Formation to the
east and north. Halmahera Island remains volcanically active.
Epithermal alteration is similar throughout the Gosowong Goldfield.
However at Toguraci the alteration halo is complicated because the
epithermal event overprinted the earlier Bora porphyry copper event.
Biotite-magnetitechlorite and chloriteepidotecalcitepyrite alteration
of the Bora event is superimposed by: quartz-adularia-illite-pyrite
alteration within the epithermal vein zone, enveloped by an
illite-quartz-pyrite set within a widespread chlorite-epidote-albite-calcite-
pyrite zone. Adularia at the Gosowong deposit has been dated as
~2.8 Ma, (40Ar/39Ar, Vasconcelos, 1998), and it is presumed that quartz
adularia alteration at Toguraci is of similar age.
Most deposits are hosted within north striking, subvertical faults. Their
location on the fault is controlled by both lithology, preferentially hosted
in diorite, and by cross-cutting NE faults (which also carry some
mineralisation). A major WNW striking, moderate NE dipping fault,
named the T-fault, traverses the northern end of the recognised Toguraci
vein field. Work to the north of T-Fault is ongoing to understand its
impact on vein and gold distribution at Toguraci. The T-Fault appears to
have been first activated pre-mineral, as it is gold/quartz bearing in
places. Additional faults, previously only seen as lineaments in air
photographs, are now being exposed by mining operations.

1. MAusIMM, Regional Exploration Manager Indonesia, Newcrest

Mining Ltd, Level 9, 600 St Kilda Road, Melbourne Vic 3004.
2. Senior Exploration Geologist Gosowong Operations, Newcrest FIG 1 - Location map: Halmahera Island and Gosowong
Mining Ltd, Level 9, 600 St Kilda Road, Melbourne Vic 3004. operations.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 359

T H RICHARDS and M D BASUKI DWI PRIYONO
Topography in the area is moderate by Indonesian standards
with elevations of between 80 m and 260 m above sea level.
Vegetation is mostly primary tropical rainforest. The area
experiences around 3000 mm of rain annually, the majority of
which falls between October and April most years.
The Indonesian exploration and mining tenement, PT Nusa
Halmahera Minerals (PT NHM) Contract of Work, covers the
Gosowong Goldfield and is a joint venture between Newcrest
Mining Limited (82.5 per cent, operator) and the Indonesian
company PT Aneka Tambang Tbk (17.5 per cent). The Gosowong
Operations began with the mining of the Gosowong gold deposit,
which commenced in 1998. This open pit undertaking produced
around 760 000 ounces at a head grade of 27 g/t Au before mining
ceased in 2002. Now the Toguraci deposits are supplying ore to
the fully written-down Gosowong mill.
Natural outcrop at Toguraci is sparse and the rapid increase in
mine exposures is resulting in a marked expansion in the
understanding of Toguraci geology. It is therefore considered
premature here to address the geology and mineralisation at
Toguraci in detail; as such an undertaking may in part have
become obsolete by the time this work is published. More
appropriate here is to present the history of the discovery and
provide an insight into the exploration techniques and beliefs that
led to it. This paper also briefly reviews the non-geological
events that have become an integral part this history.
REGIONAL SETTING
The Maluku Sea, west of Halmahera, is a zone of collision
between the opposing Sangihe and Halmahera volcanic arcs.
Eastward subduction of the Maluku Sea plate beneath the
Halmahera and the Philippine Sea plates since the Paleogene has
produced four superimposed volcanic arcs in west Halmahera.
These four volcano-sedimentary formations have been termed
the Bacan Formation (?Paleogene), the Gosowong Formation
(?Upper Miocene) the Kayasa Formation (Pliocene), and the
Quaternary Volcanic Formation which remains active
(Marjoribanks, 1997). The formations are separated by major
regional angular unconformities (Marjoribanks, 1997).
FIG 2 - Gosowong operations.
360 Adelaide, SA, 19 - 22 September 2004
The Gosowong Formation is dominated by basaltic to dacitic
volcanic and volcaniclastic rocks and intermediate intrusions.
Volcanic rocks of the Pliocene Kayasa Formation unconformably
overlie this. In the late Pliocene, the Gosowong and Kayasa
volcanic sequences were locally intruded by andesite porphyry
and quartz diorite. Post-mineral Quaternary pyroclastic flows and
airfall tuff deposits cover part of the Gosowong Formation
lithologies in the east and north.
Two styles of mineralisation are recognised to date within the
Gosowong Goldfields, low-grade copper-gold porphyry and
high-grade gold-silver epithermal veining. Four sub-economic
porphyries, Bora, Tobobo, Ngoali, and Matat and a number of
epithermal vein deposits including Gosowong, Toguraci, Ruwait,
Gosowong North, Dongak, Tobobo, Langsat and Gosowong
Extended (Kencana) have been recognised (Figure 2).
Epithermal mineralisation is multi-phased and an early
quartz-adularia episode has been dated at ~2.8 Ma (40Ar/39Ar,
Vasconcelos, 1998). All discoveries to date are contained within
a 10 km long, 4 km wide north-trending mineralised corridor as
described in Olberg et al (1999).
TOGURACI GEOLOGY AND GOLD
MINERALISATION
Toguraci veins are hosted within Gosowong Formation, as is the
recently mined Gosowong deposit. However the lithological and
structural setting at Toguraci varies markedly from that of
Gosowong (Figure 3a, b and c). Readers interested in pursuing
this comparison further are invited to also read the Olberg et al
(1999) account of the Gosowong deposit.
Two distinct mineralising events have occurred at Toguraci. A
weak porphyry copper-gold mineralisation and associated
alteration has been overprinted by the Gosowong Goldfield
epithermal gold-silver event. The porphyry copper gold alteration
is biotite-magnetitechlorite within a chloriteepidotecalcite
pyrite envelope. Epithermal gold quartz veins clearly cross-cut the
porphyry style mineralisation and the superimposed epithermal
alteration is typical of the Gosowong Goldfield: quartz-adularia-
illite-pyrite alteration in the epithermal vein zone, enveloped by an
illite-quartz-pyrite which in turn is surrounded by a widespread
chlorite-epidote-albite -calcite-pyrite halo.
PACRIM 2004
 DISCOVERY OF THE TOGURACI EPITHERMAL AU-AG DEPOSITS, GOSOWONG GOLDFIELD

FIG 3a - Gosowong shoot: type cross-section, 10 200 mN (Olberg et al, 1999).

FIG 3b - Midas shoot: type cross-section, 9850 mN.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 361

T H RICHARDS and M D BASUKI DWI PRIYONO
FIG 3c - Damar shoot: type cross-section, 9650 mN.
Most epithermal veins at Toguraci are controlled by a set of
north striking subvertical faults. High-grade gold mineralisation
zones (termed shoots here) are confined to discrete dilation
zones along these faults. Dilation has developed preferentially in
diorite (Figures 3b and c; and 4). NE cross-cutting faults (Figure
4) also hosts lower grade gold-quartz mineralisation with small
high grade shoots in some dilational jog positions. A major
WNW fault named T-Fault discontinues the Toguraci
mineralisation to the north however the area immediately north
of the T-Fault is at this stage, under-explored and this
discontinuity not well understood. The T-fault is considered of
importance both because it is an integral part of the stress regime
that developed the predominantly N-S host structures and
because it is recognised as a fertile structure. A number of
additional faults, visible in air photos but generally not well
presented in natural outcrop, are now being exposed in the
Toguraci pit and haul roads.
Toguraci veins display typical epithermal textures. Quartz vein
breccia with clasts of both wall rock and earlier quartz veins
demonstrate a multiphase history. Chlorite and black sulfide
bands and rims up to 0.5 cm thick, commonly with fine grained
visible gold, overprint a quartz adularia event and are present in
bonanza gold grade zones. This quartz-chlorite-sulfide
mineralisation phase although volumetrically minor, contributes
significantly to the gold inventory (McLeod, 2004). Early
low-grade vuggy quartz veins and stockwork, and late stage
cross-cutting carbonate veinlets are common.
Within individual shoots high gold grades are the norm.
However shoots discovered to date at Toguraci are small and show
sharp boundaries with low-grade mineralisation. Generally vein
widths range between 2 m to 6 m and are set in a wider, earlier
subgrade stockwork zone. (Wide low-grade stockwork zones
without a significant high-grade gold overprint are also present.)
362 Adelaide, SA, 19 - 22 September 2004
EXPLORATION HISTORY
The discovery history of the Toguraci deposits commenced in
1992 when Newcrest began grassroots exploration along the island
arc defining the north and south arms of Halmahera Island.
Porphyry copper gold mineralisation was the primary target.
First pass exploration in what is now called the Gosowong
Goldfield was initiated in November 1992 using reconnaissance
stream sediment and float sampling and mapping of major
drainages. In 1993 anomalous gold-in-stream results brought
attention to the Toguraci area, resulting in the discovery of poorly
outcropping porphyry copper-gold mineralisation (named the Bora
prospect) together with an overprinting epithermal quartz vein
event. Epithermal quartz vein float collected in the area returned
moderate gold grades (9 - 15 g/t Au) (Basuki, 2003).
Outcrop at Toguraci is poor due to the thick soil and rainforest
cover. To add to mapping difficulties the Toguraci high-grade
veins are brecciated and are clay and adularia (weathering to
clay) rich. As a result high-grade vein outcrop decomposes
rapidly and is therefore rare. It does not form quartz scree
deposits or ridges.
North striking subvertical faults host the bulk of epithermal
quartz mineralisation. These faults are cross-cut by a NE striking
set of faults. Shoots located to date are positioned near these
intersections. Mineralised quartz outcrop and subcrop were
originally mapped at these intersections and also along the NE
faults, leading the workers to interpret the veins to have a NE
strike. As a result early gridding, drilling and other investigations
were mis-directed for a period as it was assumed that the veins
were discontinuous along the inferred NE strike. Early IP and
ground magnetic surveys did not clarify the situation (Wilson,
1995).
PACRIM 2004
 DISCOVERY OF THE TOGURACI EPITHERMAL AU-AG DEPOSITS, GOSOWONG GOLDFIELD
FIG 4 - Toguraci veins and geology.
In May 1994, 2 km NNE of Toguraci the high-grade Gosowong
Deposit was discovered and exposed by trenching. (There the best
trench exposed quartz adularia veining with an average channel
sampling result of 35 m @ 64 g/t Au, Langmead, 1995.) A
trenching program undertaken at Toguraci in early 1995 exposed
narrow quartz veins within an earlier weak quartz stockwork.
Assays returned included 3 m @ 20 g/t Au and 2 m @ 39 g/t Au.
Although interesting these results were clearly not in the same
class as those at Gosowong, and given the apparent discontinuous
nature of the Toguraci veins, the exploration focus in the Toguraci
area remained on the Bora porphyry.
During this time a grid based soil sampling program using
25 m 25 m spacing was completed with the surface delineation
of porphyry mineralisation the aim. This spacing was appropriate
for delineating porphyry copper gold mineralisation however the
across-vein spacing proved insufficient in testing the epithermal
vein occurrences because the veins are narrow and have a poor
dispersion of gold and other pathfinder elements in the soil.
Despite this the soil program did lead to the discovery of
high-grade gold mineralisation along the Damar structure. A
trench across a gold spot high (1.2 g/t Au) disclosed a very
narrow, slumped epithermal vein. The trench was later revisited
and deepened exposing 30 cm epithermal quartz boulder subcrop
assaying 10 g/t Au. Because of the limited nature of the exposure
and the prevailing view that epithermal mineralisation was
aligned NE, Damar subcrop was interpreted to connect with the
Matoa vein subcrop to the NE (Figure 4) (Basuki, 2003).
Drilling at the Gosowong deposit diverted further work away
from Toguraci until September 1996 when a scout drilling
program to test the Bora porphyry was conducted. The first hole,
BOD001, intersected a broad zone of epithermal quartz breccia
veining assaying 17.8 metres down hole (mdh) @ 7.2 g/t Au
including 2 mdh @ 30 g/t Au (FitzGerald and Leonard, 1999).
Follow up drilling of the Bora porphyry mineralisation was
designed to simultaneously target the assumed NE striking
epithermal veins. Because of the incorrect (in hindsight) drilling
directions this yielded sporadic results.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
During March-April 1997 four diamond holes (950 m) were
completed specifically targeting the epithermal veins (FitzGerald
and Leonard, 1999). These were collared in the eastern part of
Toguraci drilling area beneath the Matoa vein, the highest grade
surface exposures discovered at that stage. It is now understood
that it is the base of mineralisation outcropping at Matoa, and
consequently no significant quartz veins were intersected
beneath this outcrop.
Interest in Toguraci waned after these test holes. The
Gosowong Mine development and work on other epithermal
prospects for a time diverted attention and funds. It was
concluded that Toguraci carried no stand-alone deposit, however
its potential to supply supplementary feed for the nearby
Gosowong operation was not discounted (FitzGerald and
Leonard, 1999). This position was fundamental to the subsequent
discoveries as it was to provide encouragement to revisit the area
for an even more detailed investigation.
Prompted by the short mine life of the Gosowong deposit, an
intensive near mine exploration program was instigated during
June 1999. All geological, geochemical and geophysical
information was compiled into a GIS database and prospects in
the Gosowong area were reevaluated. A total cover 1:2500 scale
geological mapping and sampling program was then undertaken
to explore the area within 5 km radius of Gosowong using the
GIS database to focus the work.
At Toguraci, this work commenced outside the previously
mapped area in July 1999. The program included stream
sediment (pan concentrate 80# and BLEG) sampling to
reevaluate the gold anomalies documented from previous work,
and some 25 m spaced ridge and spur composite soil sampling
(Basuki, 2003).
Geological re-mapping and sampling at a more detailed
1:1000 scale was undertaken over Toguraci in April 2000.
Several epithermal style quartz veins were mapped. One of these,
an outcrop sampled in 1994 grading 1.5 m @ 6.4 g/t Au, was
re-sampled. The resample returned a result of 1.5 m @ 89 g/t Au.
363
T H RICHARDS and M D BASUKI DWI PRIYONO
(This discrepancy demonstrates the erratic nature of Gosowong
Goldfield epithermal veins at shoot margins, rather than an
erroneous assay). The vein was named Midas after the
perceived magical touch of the geologist who re-sampled it.
During the re-mapping program the most important observation
was that most of the quartz veins at Toguraci were oriented close
to N not NE as previously interpreted. Several short trenches
were dug to explore along the re-interpreted strike of Midas and
other veins, greatly improving understanding of their geometry.
The Toguraci grid was then reoriented and also extended.
In August 2000 a 50 m-spaced drilling program was
undertaken in the Midas area. The first hole was collared to test
25 m down dip of the high grade Midas discovery outcrop and
intersected 7.4 mdh @ 18 g/t Au from 47.85 mdh. This hole
improved confidence in vein continuity. Midas was then
patterned drilled using 50 m 50 m spacing and scout drilling
was undertaken along other known vein occurrences. Importantly
the quartz exposed immediately to the SW of Midas in the
Damar trench in 1996 (discussed previously this section) was
drill tested and the Damar Shoot discovered.
At Midas a 10 m spaced drilling program was then completed
to increase confidence in the shoots internal grade continuity
around a bonanza hole grading 8.7 mdh @ 330 g/t Au (from
28mdh). Because of the extremely high gold grades it was
decided to complete the Midas resource definition using a 12.5 m
12.5 m spaced drill pattern (Figure 5a).
By early July, the Damar shoot had been drilled on a nominal
50 m 50 m grid along 250 m of strike length and infill drilling to
25 m 25 m was largely completed by late October 2001. Gold
grades and vein widths were found to be relatively consistent
along strike and with depth. This high-grade shoot is confined to a
150 m strike length, did not outcrop, and has not been effectively
closed off at depth at the time of writing (Figure 5b).
In December 2001, Newcrest Mining Limited published an
indicated plus Inferred Resource of 414 000 t @ 27 g/t Au and
42 g/t Ag for 361 000 oz Au and 553 000 oz Ag as a combined
Midas and Damar resource. Damar is the larger deposit with
78 per cent of the announced Indicated plus Inferred in situ gold
ounces (www.newcrest.com.au).
FIG 5a - Midas shoot long section. Gold gram per tonne X true thickness (gm) contours.
364 Adelaide, SA, 19 - 22 September 2004
By March 2002, Gosowong Operations were satisfied that
sufficient supplementary ore had been delineated at Toguraci to
warrant campaigning for a permit to mine in Protected Forest
(see below). Although it was also recognised at this time that the
area had significant potential for additional discoveries, all
exploration ceased at Toguraci and elsewhere in Protected
Forest designated areas of the Contract of Work, until a permit
to mine was granted.
Exploration was re-focused on parts of the tenement
designated as Production Forest by the Forestry Department;
areas where Law 41/99 would not impact on mine development.
Significantly this decision led to the early discovery of the most
spectacular gold shoot located to date in the goldfield, the
Kencana Shoot, in mid-January 2003 and a major drill campaign
ensued.
It never rains but it pours, after a hiatus in exploration at
Toguraci of over 12 months, permission to proceed with mine
permitting was granted by the Indonesian Government in May
2003, just after drilling at the Kencana discovery was escalated.
More rigs were mobilised to allow intensive exploration and
delineation programs to proceed at Toguraci and Kencana
simultaneously. At the time of writing seven man-portable and
two slightly heavier rigs were operating, mostly on these two
prospects.
Development of the Toguraci mine was rapid and the first gold
bar was poured early February 2004.
NON-GEOLOGICAL FACETS OF THE TOGURACI
DISCOVERY
Between 1999 and 2001 the Malukus, including Halmahera
Island was governed under Marshal Law because of a period of
inter-communal violence. Thousands of people are believed to
have died during this conflict. Although the warring parties never
threatened the Gosowong mining operation, on several occasions
exploration was stopped and staff evacuated as a precaution.
During this period exploration activity was restricted to within
about 2 km of the Gosowong Mine infrastructure for safety
reasons.
PACRIM 2004
 DISCOVERY OF THE TOGURACI EPITHERMAL AU-AG DEPOSITS, GOSOWONG GOLDFIELD
FIG 5b - Damar shoot long section. Gold gram per tonne X true thickness (gm) contours.
However the most profound non-technical issue to effect
exploration progress resulted from misunderstandings between
the Indonesian Departments of Mines and Forestry. During 1999
a rash of laws and regulations were rushed through the
Indonesian National Parliament. One of these, Forestry Law
41/1999, carries one clause that expressly forbids open pit
mining in Protected Forest. Because an estimated 80 per cent of
the area along all of the island arcs of Indonesia is classified
Protected Forest and because the Law was deemed to be
retrospective, the impact on the mining industry was immense.
Most Indonesian exploration and mining tenements (termed
Contracts of Work, or CoWs) already issued contain significant
tracts of Protected Forest. Within these areas sizeable exploration
and development funds had been spent using the existing
contracts with the Government of Indonesia. About one half of
the area of the current Newcrest managed PT NHM CoW lays
under Protected Forest (Figure 2), including the Toguraci
deposits.
After numerous consultations it was realised that a broad
group of Indonesian officials recognised the problem this law
had created for mining companies and also recognised that the
sanctity of contract is a key issue with all foreign investors.
Given this, Newcrest and JV partner Aneka Tambang were
hopeful that a favourable resolution of the issue created by Law
41/1999 was achievable. By March 2002 the resource discovered
at Damar and Midas in Toguraci was considered sufficient to
justify awaiting an outcome to the Law 41 impasse. It was
decided however, that no further exploration within the Protected
Forest should proceed until the issue was resolved. Exploration
at Toguraci was suspended even though numerous additional drill
targets remained to be tested.
The search focus turned to the CoW area that lay outside the
constrained Protected Forest; an area considered less prospective
for open pit resources. In November 2002 a strategic decision
was made to begin testing conceptual targets considered too deep
for open pit extraction. A highly ranked conceptual target about
700 m south of the Gosowong Mine was the first to be drilled
and the blind Kencana deposit was intersected (at the right
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
depth but on the wrong structure) in the first hole (DSD 005,
7.25 mdh @ 29 g/t Au from 158.25 mdh, including 1 mdh @ 140
g/t Au). The intersection was made in mid-January 2003. It is
interesting to note that this target would not have been drilled at
that time, and perhaps not for several years, had it not been for
Law 41. This is because work within the Protected Forest would
have proceeded uninterrupted.
CONCLUSIONS
Following are those criteria and perceptions that are considered
most influential in the discovery of the Toguraci deposits and
those that slowed the discovery process and in one case
threatened to stop it.
Toguraci discovery criteria
1. Fundamental to the discovery of the Toguraci deposits was
the recognition that the Gosowong district was a goldfield
and therefore had the potential to carry several economic
deposits.
2. As is generally the case with exploration success, the
Toguraci discoveries would not have been made without
strong senior management and Board support. This is
particularly important when the undertaking is in a nation
temporarily impaired by a retrospective legislation and in a
region deeply traumatised by communal violence.
3. Even though mostly discouraging results from Toguraci
persisted throughout the first years of exploration, the
companys geologists recognised that epithermal vein
exploration was not meant to be easy and endured the
early disappointments.
4. The discovery of Gosowong and the consequent decision to
re-evaluate all the ground in greater detail within a 5 km
radius of the Gosowong mine in the search for mill feed,
rather than for a stand-alone deposit, was critical to the
outcome.
365
T H RICHARDS and M D BASUKI DWI PRIYONO
5. The most effective reconnaissance tools leading to the
discovery were good geological understanding of
epithermal systems by the field team and careful outcrop
and float sampling. Grid based soil sampling did not work
well due to the narrow nature of the vein zones. (It is
believed that sampling would have had to have been closer
spaced across the strike to have been effective.) Stream
sediment sampling located the Bora porphyry copper/gold
mineralisation, leading explorers to the Toguraci vein field.
(However in areas with no porphyry mineralisation,
epithermal veins show a good stream sediment sample
response)
Criteria that set back discovery
1. Law 41/1999 delayed exploration around Toguraci for more
than 12 months and had the potential to have caused the
demise of the entire Gosowong Operations project.
2. Within the Toguraci area the presence of the Bora porphyry
copper-gold mineralisation diverted attention away from
the co-existing epithermal veins for some time.
3. Initial mapping and subsequent drilling mis-directions were
caused by the incorrect interpretation of strike of the major
Toguraci veins.
4. The discovery of the Gosowong deposit diverted staff and
funds from Toguraci. However, this discovery was to later
revitalise the detailed exploration at Toguraci, because of
the need for additional mill feed.
366 Adelaide, SA, 19 - 22 September 2004
ACKNOWLEDGEMENTS
The authors thank the many Newcrest geologists who have
contributed to the discovery and evaluation of the Toguraci veins.
Most of these people remain with Newcrest and continue to
explore and discover at the Gosowong Operations. We also thank
Newcrest Mining Limited for permission to publish this paper.
Dan Wood and Roger Langmead greatly improved the original
drafts of the manuscript and Saman Dami prepared the figures.
REFERENCES
Basuki Dwi Priyono, M D, 2003. The Bora-Toguraci prospect, Gosowong
district, Halmahera, Indonesia, unpublished report for PT
Puncakbaru Jayatama.
FitzGerald, F and Leonard, G, 1999. Report on exploration undertaken,
Central Gosowong District, Halmahera, Indonesia, unpublished
report for PT Puncakbaru Jayatama.
Langmead, R, 1995. The Gosowong epithermal Au-Ag prospect,
Halmahera Indonesia, unpublished report for PT Puncakbaru
Jayatama.
Marjoribanks, R, 1997. Geological interpretation of the Gosowong
region, Halmahera Island, Indonesia, unpublished report for PY Nusa
Halmahera Minerals.
McLeod, A, 2004. Field report on observations made during secondment
to Gosowong Operations, unpublished report to PT Puncakbaru
Jayatama.
Olberg, D J, Rayner, J, Langmead, R P and Coote, J A R, 1999. Geology
of the Gosowong Epithermal Gold Deposit, Halmahera, Indonesia, in
Proceedings Pacrim 99, pp 179-185 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
Vasconcelos, P, 1998. 40Ar/39Ar analysis, Gosowong prospect area,
Halmahera, unpublished report for PT Nusa Halmahera Minerals.
Wilson, D R, 1995. Bora Prospect ground magnetic and resistivity/induced
polarisation surveys, unpublished report for PT Puncakbaru Jayatama.
PACRIM 2004
Discovery Performance of the Western World Gold Industry Over
the Period 1985 - 2003
R C Schodde1
ABSTRACT
This paper summarises the performance of the gold exploration industry
in the Western World since 1970, with particular emphasis on the period
1985 - 2003. It identifies the number and size of major (ie >1 million
ounce) primary gold discoveries made over this period by location and
type of company. Estimates are also made of the likely contribution from
smaller discoveries down to 0.1 million ounces.
It also assesses the overall trend in the percentage of grassroot
discoveries that are converted into mines, and the time taken to get them
into production.
By comparing the discovery data against total exploration expenditures
it is possible to determine the average unit cost of finding a major gold
deposit. It is significant to note that the average unit cost per discovery
and per ounce of resource has tripled over the last 30 years.
An attempt is made to determine the overall economic value generated
by the industry. This is based on a detailed analysis of the rates of return
and the Net Present Values (NPVs) of 109 major gold projects. After
including an adjustment for discoveries not modelled, the NPVs are
compared against the total amount of money spent on exploration to
determine the value-to-cost ratio for the industry.
Based on historical spot gold prices and tax rules, it is found that the
return on investment for gold exploration in the Western World over the
period 1985 - 2003 was around five per cent real after-tax. The returns
achieved in the US and Australia was significantly higher than that for
Canada. The returns also appear to have declined in recent years.
With regard to the relative performance of junior versus major
companies, the paper estimates that the junior sector accounted for 30 -
40 per cent of the total exploration expenditure, and were was associated
with 34 per cent of the total number of major deposits, 27 per cent of the
contained ounces and 25 per cent of the economic value found.
INTRODUCTION
In all business endeavours a fundamental question asked by
management is does this investment make money? In the case
of mineral exploration this is a particularly challenging and
vexed question because of the long lead times, the infrequency of
discovery, and the uncertainty over the final size and likely value
of the prize.
One way to overcome these difficulties is to look at the overall
performance of the industry with the view of using the data as a
benchmark for setting targets for the company.
The current paper builds on studies done by other workers in
this area most notably by Mackenzie (1995), Mackenzie and
Doggett (1992, 1995), Mackenzie Ortiz and Doggett (1997) and
the Metals Economics Group (2000).
To this end, the following paper identifies how much money
was spent on grassroots gold exploration in the Western World
over the last three decades and how many major (defined as
>1 million ounce) gold deposits were found. It also attempts to
determine the value of these discoveries to arrive at an estimate
for the overall return on money spent.
The current analysis excludes any additional gold found at
existing mines, and ignores any byproduct gold credits
associated with base metal discoveries.
1. MAusIMM, Manager Risk Capital Analysis, WMC Resources Ltd,
60 City Road, Southbank Vic 3006.
E-mail: Richard.Schodde@wmc.com
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
With regard to the reported ounces found, these generally refer
to the pre-mined measured and inferred resource for a given
deposit as known in 2003.
For brevity, the terms millions of ounces and thousands of
ounces of gold have been abbreviated to Moz and koz
respectively.
Unless otherwise specified, all numbers referred to in the
paper are in constant June 2002 US dollars.
TRENDS IN EXPLORATION EXPENDITURES
Industry level
Figure 1 shows the trends in exploration expenditures in the
Western World between 1970 and 2003. The data is derived from
the annual surveys compiled by the Metals Economics Group
(1992 to 2003) and from work by Mackenzie and Doggett (1992
and 1995). The split between grassroots exploration, advanced
exploration and feasibility studies, and mine site exploration for
the years prior to 1992 are the authors estimates.
All up, $37.8 billion was spent by private industry on gold
exploration between 1985 and 2003. Approximately $28.6
billion of this was on grassroots and advanced exploration. The
former refers to money spent on finding the deposit outside
existing mining operations and the latter refers to amount spent
on delineating an economic resource and carrying out feasibility
studies up to the point where an investment decision can be
made. Given that the two activities are associated with the same
discovery, the author has combined the two costs together to
arrive at the overall cost of making a grassroots discovery.
Together, these two activities account for three-quarters of all the
money spent on gold exploration.
In practice, the reported resource figure often continues to
grow after the mine has started up. Consequently, some of the
$7.8 billion spent on minesite exploration should also be
assigned back to the original grassroots discovery cost. To keep
the analysis simple, the author has excluded this cost.
Country level
Figure 2 shows where the exploration expenditures were incurred.
Prior to the early 1980s, much of it went into just three countries
Australia, Canada and the United States all of which have a long
history of gold mining. In recent years, there has been a significant
shift in efforts towards grassroots gold exploration in developing
countries within Latin America and Africa.
Company sector level
Of interest is the split in exploration expenditures between major
and junior companies. Unfortunately, only limited data is
available.
An analysis of data from Bouchard (2003), over the period
1985 - 2002, indicated that 36.5 per cent of all exploration
expenditures in Canada was associated with junior companies.
With regard to Australia, a recent study by Maritz (2003)
estimated that between 1994/1995 and 2001/2002, junior
companies spent a total of $A 2125 million on mineral
exploration for all commodities, compared with a total of
$A 6907 million spent by the entire industry. This suggests that
367
R C SCHODDE

Expenditures (US$ Billion in 2002$)

$5 Mine Site
1985-2003
Advanced
Grassroots $17.03
$4 Advanced $11.59 Grassroots
Mine Site $9.16
US$37.78 Billion
$3

$2

$1

$0
1970 1975 1980 1985 1990 1995 2000
FIG 1 - Gold exploration expenditures in the Western World.

Expenditures (US$ Billion in 2002$)

$5 Other
1985-2003 Canada
Australia $4.36 USA
$4 USA $3.41 Australia
Canada $4.95
Other $15.86
$3 US$28.63 Billion

$2

$1

$0
1970 1975 1980 1985 1990 1995 2000
FIG 2 - Grassroots and advanced exploration expenditures by country.

the juniors accounted for 31 per cent of total expenditures in
Australia. However, it should be noted that Maritzs definition of
a junior is broader than the authors. In detail, Maritz defined a
junior as any mining company with a market capitalisation of
less than $A 200 million. On the other hand, Maritzs data
includes minesite exploration which, if excluded, would boost
the share assigned to the junior sector for grassroots exploration.
Finally, a study of the exploration performance of the
international mining industry by the Metals Economics Group
(2000) estimated that, between 1989 - 1998, junior and
intermediate companies accounted for 37 per cent of the total
exploration expenditures in the Western World. These numbers
included minesite exploration. In a more recent study by the
Metals Economics Group (2004), they estimated that, between
1997 and 2003, junior companies spent a total of $3948 million
(in dollars of the day) on all forms of gold exploration. Given
that, over the same period, the entire industry spent $8942
million on gold exploration, implies that the junior companies
accounted for 44 per cent of total expenditures.
Unfortunately, none of the data sources provide a breakdown of
grassroots gold exploration by industry sector over the study
period.
Given the above, it is estimated that the junior sector
accounted for approximately 30 to 40 per cent of the total
industry expenditures on grassroots gold exploration over the
period 1985 - 2003.
NUMBER OF GOLD DEPOSITS DISCOVERED
BETWEEN 1985 - 2003
Over the last decade, WMC has built up a comprehensive database
on geological information on nearly 112 000 mineral occurrences
around the world. This includes discovery data on over 2500
primary gold deposits. This includes 354 grassroots deposits with
a pre-mined resource greater than 100 koz and were found in the
Western World between 1985 and 2003. The cumulative
size-frequency curve for these deposits is plotted in Figure 3.
It should recognised that no geological database is 100 per
cent complete. Notwithstanding this, every effort has been made
to capture as many as possible of the large economic deposits.
Figure 3 shows that the shape of cumulative-frequency curve
rolls-over below 1 Moz. Industry experience is that the size
distribution of mineral deposits follows a lognormal distribution.
Consequently, the overall number of small-sized discoveries was

368 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DISCOVERY PERFORMANCE OF THE WESTERN WORLD GOLD INDUSTRY OVER THE PERIOD 1985 - 2003

Cumulative Number of Deposits

10000 Between 1985-2003
~1000 Gold Deposits > 100koz
were found in the Western World
est 1000 >100koz Cumulative Total = 1127 Moz
1000 Between 1985-2003
354 est 190 >1moz ~190 Gold Deposits > 1moz
were found in the Western World
Cumulative Total = 895 Moz
172
100 identified
estimated data
Western World

10
USA
Canada

1 Australia
0.1 1 10 100
Deposit size (moz Pre-Mined Resource)
FIG 3 - Cumulative number of grassroots gold deposits found between 1985 - 2003.

Number of Discoveries >1moz Expenditures (US$ Billion in 2002$)

25 $5

20 Exploration $4
Known Expenditures
Discoveries (Grassroots + Advanced )
15 $3

10 $2

5 $1

0 $0
Caution :
1970 1975 1980 1985 1990 1995 2000 Incomplete data

Discovery Year
FIG 4 - Number of major discoveries per year versus exploration expenditures.

estimated by extrapolating the curve below the 1 Moz threshold.
On this basis, it is estimated that approximately 1000 primary
gold deposits (greater than 100 koz) were found in the Western
World over the study period. These contain 1127 Moz of gold.
Approximately 190 of these deposits were larger than 1 Moz.
Figure 3 also includes data on the number of deposits found in
Australia, the United States and Canada. It is estimated that 200,
150 and 100 gold deposits (>100 koz) were respectively found in
these three countries. They respectively contain 179, 206 and
61 Moz of gold.
The above numbers are consistent with studies done by other
workers. In detail, Geoscience Australia (2002) identified 147 gold
discoveries, rediscoveries and renewals in Australia since the
1960s. Hogan et al (2002) estimated that around 172 Moz of gold
was found in Australia between 1984/1985 and 2001/2002.
Cranstone (2002) identified that 103 Moz of gold was found in
Canada between 1985 and 1990 though 60 Moz of this was
associated with porphyry and VMS base metal deposits.
The frequency of major gold discoveries per year is shown in
Figure 4. On average, between 1985 and 2003, around ten
deposits containing more than 1 Moz were found each year. Half
of these deposits contained more than 3 Moz of gold.
COST PER DISCOVERY
All up, between 1985 and 2003 approximately 190 gold
discoveries larger than 1 Moz were found in the Western World
this includes 172 identified deposits (see Figure 4) and 18
unknown deposits (as extrapolated from the cumulative
frequency curve). Dividing this by the $28.6 billion spent on
exploration implies an average finding cost of around $150
million per major discovery.
Figure 5 shows that the average cost per discovery has
increased significantly over time especially in the last five
years. The latter may due to the time delay in reporting and
delineating a major discovery, resulting in under-estimation in
the true number of discoveries. For this reason, the results for the
most recent years of the survey period should be treated with
caution.
With regard to the period 1985 - 2000, the average cost per
known major discovery was three-times higher than that for
1970 - 1984 ($160 million versus $61 million). Not only did the
average cost per discovery go up, but the weighted average size
of each discovery made shrank from 7.2 to 5.0 Moz.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 369

R C SCHODDE

Cost per Major Discovery (US$m in 2002$)

$400
Weighted Average Cost
1970 - 84 : US$61m for 7.2 moz deposit
1985 - 00 : US$160m for 5.0 moz deposit
$300
Caution :
Incomplete data

$200

$100

$0
1970 1975 1980 1985 1990 1995 2000
Discovery Year
FIG 5 - Average cost per major discovery.

Gold Discovered (moz pa)

250
Unspecified
200 0.1-1moz
1-3moz
150 >3moz

100

50 Caution :
Incomplete data

0
1970 1975 1980 1985 1990 1995 2000
Discovery Year
FIG 6 - Estimated amount of gold discovered.

Cost (US$/oz in 2002$)

$100
Weighted Average Cost Caution :
Incomplete data
1970 - 84 : US$7/oz
$80 1985 - 00 : US$25/oz

$60

$40

$20

$0
1970 1975 1980 1985 1990 1995 2000
Discovery Year
FIG 7 - Average cost per ounce of gold discovered.

370 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DISCOVERY PERFORMANCE OF THE WESTERN WORLD GOLD INDUSTRY OVER THE PERIOD 1985 - 2003

These results are consistent with studies by other workers. For
example, Mackenzie and Doggett (1992) determined that the
average historic cost of discovery of a gold deposit in Australia
in 1969 - 1986 was $15 million in constant 1990 dollars,
compared with average 1969 - 1988 costs of $24 million in
Brazil and $96 million in Canada. After adjusting for inflation,
these figures equate to around $21, $34 and $135 million in
constant 2002 dollars for Australia, Brazil and Canada
respectively. It should be noted that the Mackenzie and Doggett
data included several deposits smaller than 1 Moz in size.
Figure 6 highlights the trend in the amount of gold discovered
per year. All up, between 1985 and 2003 a total of 1127 Moz of
gold is estimated to have been found in the Western World.
Around 895 Moz of this is contained in deposits larger than
1 Moz. The average unit discovery cost for the period
1985 - 2000 is estimated to be around $25 per ounce (Figure 7).
This is nearly four times the $7 per ounce cost calculated for the
period 1970 - 1984.
CONVERSION RATES
From a business perspective, identifying the cost of discovery is
only part of the story. To gain a complete picture of the industrys
overall performance one also needs to identify their value. A
useful metric for this is the percentage of these discoveries that
were ultimately converted into mines.
Figures 8 and 9 detail the status of 276 major gold discoveries
made since 1970. It shows that most of the recent discoveries are
still at the exploration or feasibility stage. This is not surprising
given the long lead times involved in delineating the resource,
evaluating the economics of the project and securing the
necessary permits and finance to build the mine.
Figure 9 indicates that around 70 per cent of all major
discoveries are eventually converted into mines. A further five to
ten per cent of projects were stalled at the decision-to-build
stage. This was due to a wide variety of reasons including
environmental and social (Crown Jewel in USA), legal (Las
Cristinas in Venezuela), political (Gold Ridge in the Solomon
Islands) and economic issues (Metates in Mexico).

Number of Projects
80 Exploration/Undeveloped
Stalled Project
60 59 Feasibility
60
Under Construction
45 44 Operating Mine
40 Closed Mine

20
20 15
8

0
1970-74

1975-79

1980-84

1985-89

1990-94

1995-99

2000-03

FIG 8 - Number and status of major gold projects.

Percentage
100% Exploration/Undeveloped

Stalled Project
80% Feasibility

Under Construction
60%
Operating Mine

Closed Mine
40%

20%

0%
1970-74

1980-84

1985-89

2000-03
1975-79

1990-94

1995-99

FIG 9 - Percentage breakdown of major projects by status.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 371

R C SCHODDE
There is a general perception is that it now takes longer to start
up a mine than in the past, especially in developed countries.
Analysis indicates that this is not the case. Figure 10 plots the
discovery and mine start-up dates for 214 known grassroots
deposits found in the period 1970 - 2003. It shows no change in
the length of the delay over the last 30 years. Over this period it
took 6.3 years for half of the mines to start production.
Surprisingly, the delay was shorter in Australia, Canada and the
United States than in other countries (5.4 versus 8.3 years).
Notwithstanding this, some mines took more than 20 years to
develop.
Startup Year
Delay . 15 10 5 0 Years
2010
2000
1990
Average
Delay Time
1980
Australia + Canada + USA 5.4 Yrs
8.3 Yrs
Other Countries
1970
1970 1980 1990 2000 2010
Discovery Year
FIG 10 - Discovery and start-up dates for >100 koz gold projects.
While, on average, it may not be taking any longer to start-up a
gold mine, it is fair to say that the effort required in securing the
necessary permits and approvals is much higher than in the past.
VALUE OF THE DISCOVERIES
To determine the overall value created by exploration, detailed
financial models were developed for many of the known
economic discoveries. To keep the task manageable, the
modelling was limited to discoveries made after 1985 that
contain more than 1 Moz of gold.
Wherever possible, the author used the actual reported or
forecast production schedules, capital and operating costs for
each mine. The primary source of data was Brook Hunt (2002).
The information was fed into a spreadsheet, which incorporated
gold prices, local tax rules, exchange and inflation rates to come
up with an annual cashflow data for each project.
The analysis covered all expenditures and production data for
the project between 1985 and 2025. For a given project, the cash
flows were discounted back to the year that a decision was made
to build the mine. For example, if a mine operated for ten years
from 2000 and the decision-to-build was made in 1998, then all
the cashflows were discounted back to 1998. This equates to the
NPV value for the project.
If, over time, the projects cash costs exceeded its revenues
(because the gold price chosen was lower than the reported
costs), the financial model was designed to shutdown the mine
early to conserve cash.
The cash flows were discounted at seven per cent real
after-tax. This reflects the cost of capital for a typical gold mine
in Australia, but makes no allowance for any risk premium for
operating in difficult countries.
372 Adelaide, SA, 19 - 22 September 2004
The two main variables considered were gold prices and tax
rules.
Over the study period, the price of gold has fluctuated from a
high (in real terms) of $765/oz in December 1987 to a low of
$271/oz in February 2001. Consequently, the timing on when a
project was developed has a huge impact on the value created.
Tax is also an important variable. As an example, in Australia,
prior to 1990, gold producers did not pay corporate income tax.
Between 1991 and 1999 the tax rate was 36 per cent, and is now
currently set at 30 per cent. Furthermore, in 2001 the State
Government of Western Australia introduced a 2.5 per cent
royalty.
Three general cases were modelled. The first (called the Spot
Price Case) involved using the actual historical gold price and
tax rules prevailing at the time the mine operated. For any sales
post-2004, it is assumed the future price remains fixed at
$US 400/oz in real terms. This case sets a lower bound on the
projects value.
In practice, many producers took advantage of financial
instruments to sell forward production. This often resulted in a
higher effective price for the gold produced. Unfortunately, it is
very difficult to identify the actual price achieved for all the
projects modelled. As a proxy for hedging, a second case
(Forward Price Case) was constructed which involved
selling-forward all of the gold at the price (in constant real terms)
at the time the decision was made to build the mine. The case
also assumed no future changes in tax rules. This case effectively
sets an upper bound on the likely value of the project as most
producers didnt hedge 100 per cent of production.
A third case (Real Price Case) was constructed which involved
using a constant real price over the life of the mine. In this case,
the current (2004) tax rules were used. The intent of this case is
to determine the value of finding and developing a given deposit
today using todays gold prices and investment rules. The price
was notionally set at $US 400/oz in constant 2002 dollars to
reflect the price of gold prevailing at the time this paper was
written.
It should be cautioned that the Real Price Case ignores the
issue of changes in costs over time. For example, if the mine was
found and developed today, it would benefit from using the latest
systems and equipment. Another key issue is that many of the
mines operated during times of high gold prices. If the same
mine was developed today it might use a different cut-off grade
and hence may have a completely different production schedule
and cost structure.
All up, 109 gold projects were modelled. This represented
57 per cent of the estimated total 190 major discoveries made
during this period. They also contained 74 per cent of the
estimated ounces found. On this basis, the modelled results
should give a fair representation of the true performance of the
overall industry.
The modelled results for the Spot Gold Case are given in
Figures 11 and 12.
As can be seen in Figure 11, many of the projects had Internal
Rates of Return (IRRs) well in excess of 20 per cent real
after-tax. Consequently, it is not surprising that most of them
were developed into mines.
The estimated Net Present Values (NPVs) for the various
projects are given in Figure 12. As a generalisation, the
larger-sized deposits were more valuable though there are
obvious exceptions to this rule. Over the study period, the most
valuable discovery was Yanacocha in Peru which contains
more than 50 Moz of gold and has an estimated NPV @ seven
per cent discount rate of $1881 million. The least valuable
project was Mt Todd in Australia, which had major technical
problems with treating hard low-grade ore. This project cost
$290 M to build but only produced 336 koz of gold before
closing down in 2000.
PACRIM 2004
 DISCOVERY PERFORMANCE OF THE WESTERN WORLD GOLD INDUSTRY OVER THE PERIOD 1985 - 2003

IRR at Decision to Build

Raleigh Plutonic Yanacocha
100% Pajingo Redeemer
Yilgarn Star
Case 1 : Historical
Thunderbox
80% SPOT gold price
Chatree
Eskay Ck
60% Granny Smith
Morila

Pierina
40% Las Cristinas Pascua

Tarkwa Donlin Creek

20%
Cadia Cerro Casale
Cripple Creek Sun
0%
1 10 100
-20%
Deposit size (moz Pre-Mined Resource)
FIG 11 - Internal Rate of Return for major gold deposits found between 1985 - 2003.

NPV at Decision to Build (US$m @ 7% discount rate in 2002$) Yanacocha

$1881m
$1000
Case 1 : Historical Pascua
SPOT gold price $1337m
$800

$600 Donlin Creek

Pierina
Tarkwa
$400 Morila Granny Smith Las Cristinas
Phoenix

Pajingo
$200
Cripple Creek Cerro Casale
$0 Cadia Sun
1 Refugio
Syama
10 100
-$200 Mt Todd
($251m) Deposit size (moz Pre-Mined Resource)
FIG 12 - Net Present Value for major gold deposits found between 1975 - 2003.

NPV per Ounce (US$/oz in 2002$)

$120
Pajingo Weighted Average NPV = US$19/oz
(assign zero value to unprofitable projects)

$80 Perama
Sepon
Hill
Raleigh El Penon Pierina
Esquel Morila
Chimney Ck
Pascua Yanacocha
$40 $29 Granny Smith
Weighted
$19 Las Cristinas
Average
$13
Cerro Casale
$0 Cadia Sun
1 10 Cripple Creek 100

-$40
Deposit size (moz Pre-Mined Resource)
FIG 13 - NPV per ounce of resource for major gold deposits found between 1985 - 2003.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 373

R C SCHODDE

In the Spot Price Case, the combined value of the 109 deposits
modelled was $12.35 billion. It should be noted that this figure
excludes the value the 26 projects with negative NPVs. This was
done on the basis that with better knowledge these projects
should not have been developed and hence the loss of value is
due to management and operational issues rather than the fault of
exploration.
The reported NPVs are sensitive to the price case used. As a
generalisation, the Forward Price Case generally gave the
highest values. The only exception is for projects developed in
the late 1990s when gold prices were low.
A good crosscheck on the reasonableness of the valuations can
be found by comparing the calculated NPV per ounce of gold
against the sales price of known gold projects. Figure 13 shows
that there is a wide range of NPV values, with the highest being
the high-grade Pajingo deposit in Australia (at $117/oz), which
started production in the late 1980s during a period of gold prices
and zero corporate tax. Under the Spot Price Case, the weighted
average value of all the economic projects modelled was $19 per
ounce. The equivalent values under the Forward Price and
Constant Price cases were $25 and $21 per ounce respectively.
For comparison, data from Schodde (2003) identified that the
weighted average sales price for undeveloped gold deposits sold
in North America and Australia over the period 1993 - 2003 was
$21 per ounce in nominal dollars.
WHO FOUND THE DEPOSITS?
Within the industry there is vigorous debate about what type of
companies are better at exploration principally junior versus
major companies. Based on the current study, the answer
depends on which metric is used to measure success namely
the number of deposits discovered, the total ounces found or the
value created.
For purposes of this study, the author has separated the major
companies into two groups Major Producers (defined as
companies producing >300 koz of gold per annum plus
Diversified Companies with >$500 M pa in other sales revenues)
and Moderate Producers (100 to 300 koz pa of gold). The Junior
Companies were defined as the sum of small producers
(<100 koz pa), junior explorers (which have no production) and
individual prospectors. This approach differs from other workers,
such as Maritz (2003), with use market capitalisation as means
for defining the various groups.
Table 1 and Figure 14 show which types of organisations that
discovered the 190 major gold deposits found between 1985 and
2003. The general trends in discovery rates over time can be
found in Figure 15.
It should be noted that there was incomplete information on 23
of these deposits. They were assigned to the known categories on
a pro-rata basis. Using this approach, it was estimated at 44 per
cent of the major discoveries were made by Major Producers,
with an additional 11 per cent found by Moderate Producers. By
comparison, the Junior Companies accounted for 34 per cent of
the discoveries. A further five per cent were found by joint
ventures between the two main groups.
At face value, the relative performance of the junior sector is at
odds with previous studies by other workers. These include a
study by CRU International, as reported by Maponga and
Maxwell (2000), where it was estimated that between 1970 and
1997, junior exploration companies made 51 per cent of
significant gold and copper discoveries in Australia. A more
recent study by Geoscience Australia, as reported by Hogan et al
(2002), identified that junior companies accounted for 66 per
cent of all gold discoveries, rediscoveries and renewals in
Australia since the 1960s.
It is argued that the difference is due to the assessed date of
discovery, location and size of discovery.
Figure 15 shows the percentage breakdown of discoveries by
date. As can be seen, in the 1970s and 1980s Junior Companies
made a significant proportion of the major discoveries but this
has diminished in more recent years.
With regard to size, Table 1 indicates that Junior Companies
found more of the 1 - 3 Moz orebodies than the Major
Producers. A recent study by Schodde (2003) of the exploration
performance of the Australian Gold Industry identified that
junior companies found half of the identified discoveries in the
0.1 to 1 Moz size range.
A breakdown of the 190 major discoveries indicates that 36 of
these were in Australia. Of these, 13 were found by Junior
Companies plus a further three as joint ventures between major
and junior companies. This equates to 36 to 44 per cent of all the
major discoveries in Australia. The implication is that Junior
Companies are very active in Australia.

TABLE 1
Major gold deposits (>1 Moz) found in the Western World between 1985 - 2003, by company.

Discovery Sector
Major Moderate Junior Shared (major Other (oil cos Government Not specified Unknown and Total
producers producers companies + juniors) and industl) not in database
Number of major (>1 Moz) gold deposits found
>5 Moz 25 6 13 - 1 1 - - 46
3 - 5 Moz 20 5 9 1 1 2 2 - 41
1 - 3 Moz 30 7 34 6 2 4 3 18 103
Total 75 18 56 7 4 7 5 18 190
Pro-rated total 84 20 65 9 4 8 - - 190
(%) (44%) (11%) (34%) (5%) (2%) (4%) - - (100%)
Amount of gold found in major gold deposits (Moz)
>5 Moz 389 47 122 - 6 7 - - 571
3 - 5 Moz 64 19 39 4 3 7 7 - 144
1 - 3 Moz 49 13 58 11 2 6 5 34 180
Total 503 79 219 16 11 20 12 34 895
Pro-rated total 520 84 237 19 12 23 - - 895
(%) (58%) (9%) (27%) (2%) (1%) (3%) - - (100%)

374 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DISCOVERY PERFORMANCE OF THE WESTERN WORLD GOLD INDUSTRY OVER THE PERIOD 1985 - 2003

Number of Major Discoveries

Unknown
80
Govt
65 Other
62 61
60 Shared
Prospector
47
Junior
40 Small
Moderate
20 Major
20 15 Diversified
8

0
1970-74

1975-79

1980-84

1985-89

1990-94

1995-99

2000-03
FIG 14 - Number of major discoveries by company type.

Percentage of Major Discoveries

Unknown
100%
Govt
Other
80%
Shared
Prospector
60% Junior
Small
40% Moderate
Major
Diversified
20%

0%
1970-74

1975-79

1980-84

1985-89

1990-94

1995-99

2000-03

FIG 15 - Percentage breakdown of major discoveries by company type.

With regard to the amount of gold found, Table 1 indicates that
Major Producers found 58 per cent of the ounces discovered in
major deposits between 1985 and 2003. The Junior Companies
accounted for 27 per cent, with a further two per cent associated
with joint ventures between junior and major companies. This is
consistent with the observation that Major Producers tend to
focus on exploring for giant gold deposits, while the junior sector
is more active at developing smaller projects.
With regard to the value created, the Major Producers
dominated the scene. Figure 16 ranks the top 30 gold discoveries
by value (as per the historical Spot Price Case). Major Producers
made 15 of these discoveries versus eight by Junior Companies.
More significantly, in the top ten discoveries by value, the
Majors made seven.
As stated before, in the Spot Price Case, the combined value
of the 109 deposits modelled was $12.35 billion with an
average unit value of around $19 per ounce. With regard to
discoveries made by the Major Producers, 46 deposits were
modelled containing 393 Moz of gold. This group also found a
further 38 deposits containing 127 Moz that were not modelled.
Assuming that the notional value of the additional discoveries is
the same as the average modelled value of $19/oz, it is possible
to estimate the total value found by the various sectors. On this
basis, the total value of all 190 major deposits found between
1985 and 2003 is estimated to be around $16.61 billion. The
Major Producers accounted for $9.81 billion or 59 per cent of
the overall value. The Junior Companies accounted for 25 per
cent with an additional two per cent associated with joint
ventures between junior and major companies. A further ten per
cent is associated with Moderate Producers.
A similar percentage breakdown in value between the various
sectors is achieved under the other two price scenarios (see Table
2).
It should be cautioned that the above analysis might
underestimate the importance of Junior Companies in creating
wealth as it excluded discoveries smaller than 1 Moz, which is
the size range than many juniors are actively working on.
WHERE WAS THE GOLD FOUND?
Referring back to Figure 3, it is estimated that, between 1985 and
2003, approximately 1127 Moz of gold was found through
grassroots exploration the Western World on deposits greater
than 100 koz. This was made up of 179 Moz found in Australia,
206 Moz in the United States, 61 Moz in Canada and 681 Moz in
other countries.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 375

R C SCHODDE

NPV at Decision to Build (US$m @ 7% discount rate in 2002$)

$2000
$1800 Case 1 : Historical Major Producer + Diversified
$1600 SPOT gold price Moderate Producer
$1400 Small Producer + Junior Explorer + Prospector
$1200 Other
$1000
$800
$600
$400
$200
$0

TA G O
DO PA HA

CO O
UT A

SU -JU O
Y RO
E

ER
CA N
CR A

GR RIS ILA

MA PA AL

C S
ES LA IX
A

IE
ES AY
PH LD

PE L
DG IC

IM T O A
SE B ITH

(O S

IS EE
DO E)

DE K
G U PIE K

NY AS

SA AM
PL W

KE AF
U

A
RT

NO
EL UE

CH BA IO L
NA RIN

EE

LL

RE RE E
EE

VE EN

RI ON
MC XID
OC

PO

EM
W
NL SC

NE PIL
NA

KA DE

RY JIN
RK

EW

NR ND
PO RIS
C OR
AN TIN
SM

L A AH

D
NO

Q
CR
O

D
AC

E
M
S
IN

Y
N

N
YA

S
LA
LA

NI
FIG 16 - Top 30 gold discoveries between 1985 - 2003.

TABLE 2
Value of major gold deposits (>1 Moz) found in the Western World between 1985 - 2003, by company.

Discovery sector
Major Moderate Junior Shared (major + Other (oil cos Government Total
producers producers companies juniors) and industl)
Number of Modelled 46 10 40 5 2 6 109
deposits Not modelled 38 10 25 4 2 2 81
Pro-rated total 84 20 65 9 4 8 190
Amount of Modelled 393 53 183 10 9 17 665
gold (Moz) Not modelled 127 31 54 9 3 6 230
Total 520 84 237 19 12 23 895
Estimated value of major discoveries ($US billion in constant 2002$)
Value ($b) SPOT price case
Modelled 7.46 1.15 3.13 0.16 0.26 0.19 12.34
Not modelled 2.36 0.57 1.01 0.16 0.05 0.11 4.27
Pro-rated total 9.81 1.72 4.14 0.32 0.31 0.30 16.61
(%) (59%) (10%) (25%) (2%) (2%) (2%) (100%)
Value ($b) FORWARD price case
Modelled 9.23 1.99 4.49 0.41 0.16 0.24 16.51
Not modelled 3.15 0.77 1.35 0.22 0.07 0.15 5.71
Pro-rated total 12.38 2.76 5.84 0.63 0.23 0.39 22.22
(%) (56%) (12%) (26%) (3%) (1%) (2%) (100%)
Value ($b) REAL price case ($US 400/oz)
Modelled 8.05 1.51 3.48 0.23 0.26 0.20 13.74
Not-modelled 2.62 0.64 1.12 0.18 0.06 0.12 4.75
Pro-rated total 10.67 2.15 4.61 0.41 0.32 0.33 18.48
(%) (58%) (12%) (25%) (2%) (2%) (2%) (100%)

Details of the estimated value of these discoveries for the
various locations under the three different price cases can be
found in Table 3. These figures are based on the 109 modelled
projects (which contain 665 Moz) plus estimated values for the
other 462 Moz of discoveries not modelled. The value of these
other deposits was notionally calculated using the average
modelled discovery value for the world. This approach ignores
the issue of possible variations in value between countries
associated with differences in local costs, tax rules and country
risk issues. Furthermore, the value may also be overly optimistic,
as the non-modelled deposits tend to be less economic (which is
often the reason why there is no cost data available for them).
Notwithstanding the above qualifications, as shown in Table 3,
it is estimated that, based on the Spot Price Case, the total value
created by the Western World exploration industry over the
period 1985 - 2003 was around $20.92 billion. This is based on a
seven per cent real after-tax discount rate. Approximately $3.40
billion of this was found in Australia, $3.48 billion in the United
States, $1.26 billion in Canada and $12.79 billion in other
countries. On a percentage basis, 61 per cent of the total value
was found outside the traditional gold producing countries of
Australia, Canada and the United States.

376 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DISCOVERY PERFORMANCE OF THE WESTERN WORLD GOLD INDUSTRY OVER THE PERIOD 1985 - 2003

TABLE 3
Estimated number of gold deposits (>100 koz) found in the Western World between 1985 - 2003, by country.

Location of discovery
Australia USA Canada Other countries Total Western World
Number of Modelled >1 Moz 29 16 7 57 109
deposits Not modelled >1 Moz 7 13 5 56 81
Not modelled <1 Moz 164 121 88 437 810
Total 200 150 100 550 1000
Amount of gold Modelled >1 Moz 119 113 18 415 665
(Moz) Not modelled >1 Moz 14 58 20 137 230
Not modelled <1 Moz 46 34 23 129 232
Total 179 206 61 681 1127
Exploration expenditures 4.36 3.41 5.00 15.86 28.63
(grassroots + advanced) ($b)
Estimated value of major discoveries ($US billion in constant 2002$)
Value ($b) SPOT price case
Modelled >1 Moz 2.28 1.76 0.46 7.86 12.34
Not modelled >1 Moz 0.26 1.08 0.38 2.54 4.27
Not modelled <1 Moz 0.86 0.64 0.43 2.40 4.32
Total 3.40 3.48 1.26 12.79 20.92
(%) (16%) (17%) (6%) (61%) (100%)
Value ($b) FORWARD price case
Modelled >1 Moz 5.08 2.54 0.55 8.35 16.51
Not modelled >1 Moz 0.35 1.45 0.50 3.40 5.71
Not modelled <1 Moz 1.15 0.85 0.57 3.21 5.77
Total 6.58 4.84 1.62 14.96 27.99
(%) (24%) (17%) (6%) (53%) (100%)
Value ($b) REAL price case ($US 400/oz)
Modelled >1 Moz 2.64 1.69 0.49 8.92 13.73
Not modelled >1 Moz 0.29 1.21 0.42 2.83 4.75
Not modelled <1 Moz 0.95 0.71 0.47 2.67 4.80
Total 3.88 3.60 1.38 14.42 23.28
(%) (17%) (15%) (6%) (62%) (100%)

HOW MUCH VALUE WAS CREATED?
The ultimate objective of exploration is to create value. One
measure of wealth creation is to look at the ratio between the
value of what was discovered and the cost of finding it. However,
while simple to measure, this approach ignores the issues of
timing (ie when the money was spent versus when the value was
created), option-value (principally the opportunity to make
additional money if the gold price goes up) and taxation
(companies can get tax credits for money spent on exploration).
Even so, it gives an approximate measure of the industrys ability
to generate wealth.
Industry level
Based on the historical Spot Price Case, Table 4 shows that over
the period 1985 to 2003 the Western World Industry created an
estimated $20.92 billion of value from gold exploration at a
seven per cent discount rate. Over the same period it spent
$28.63 billion on grassroots and advanced exploration. This
implies a value-to-cost ratio of 0.73. In other words, the industry
generated 73 cents of after-tax value per (pre-tax) dollar spent
with a net loss of 27 cents. Alternatively, if the discount rate was
lowered to 4.8 per cent the industry would have broken-even.
Figure 17 shows the trend in value creation versus exploration
expenditure over time. It is surprising to note that the two periods
of greatest value creation (1986 - 1987 and 1993 - 1994) actually
preceded the peaks in exploration expenditure (1988 and 1997).
This goes against the conventional wisdom that there is a lag of
several years between expenditure and discovery.
Notwithstanding the proviso that the recent discovery record is
incomplete, Figure 17 clearly shows that there has been a marked
decline in industry performance from 1998 onwards. In addition
to the reduced amount of ounces found in recent years, the
industry also suffered from lower spot gold prices which affected
the unit NPV value of a given project.
If the company sold-forward all of the projects production at
the time the decision was made to build the mine (Forward Price
Case), the value created would have been $27.99 billion
resulting in a value to cost ratio of 0.98 and a breakeven discount
rate of 6.8 per cent.
In practice, the value created by the industry lies somewhere
between these two cases.
An interesting scenario to consider is the likely value of
finding and developing the same portfolio of projects today, at
todays gold price of $US 400/oz. This is captured in the Real
Price Case, which shows the industry would have created $23.28
billion of value resulting in a value-to-cost ratio of 0.81 and a
breakeven discount rate of 5.5 per cent.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 377

R C SCHODDE

TABLE 4
Estimated value of gold deposits (>100 koz) found in the Western World between 1985 - 2003, by country.

Location of discovery
Australia USA Canada Other countries Total Western World
Estimated value created from grassroots exploration @ 7 % discount rate ($US billion in constant 2002$)
SPOT price case 3.40 3.48 1.26 12.79 20.92
FORWARD price case 6.58 4.84 1.62 14.96 27.99
REAL ($US 400/oz) price case 3.88 3.60 1.38 14.42 23.28
Exploration expenditures ($US billion in constant 2002$)
Grassroots + advanced 4.36 3.41 5.00 15.86 28.63
Value to cost ratio
SPOT price case 0.78 1.02 0.25 0.81 0.73
FORWARD price case 1.51 1.42 0.32 0.94 0.98
REAL ($US 400/oz) price case 0.89 1.05 0.28 0.91 0.81
Discount rate required to breakeven
SPOT price case 5.2% 7.2% negative 5.5% 4.8%
FORWARD price case 10.3% 9.5% negative 6.4% 6.8%
REAL ($US 400/oz) price case 6.2% 7.5% negative 6.3% 5.5%

US$ Billion
$4
Case 1 : Historical
SPOT gold price
Exploration
$3 Expenditures
(Grassroots + Advanced )
Estimated NPV
of Discoveries
$2

$1

No NPV data
prior to 1985
$0
Caution :
1980 1985 1990 1995 2000 Incomplete data

Discovery Year
FIG 17 - Estimated NPV created versus exploration expenditures per year in the Western World.

Some of Australias most profitable mines were found in the

Country level
Under the Spot Price Case, on a country-by-country basis,
United States generated a value-to-cost ratio of 1.02, versus 0.78
for Australia and only 0.25 for Canada. The corresponding
breakeven discount rates were 7.2 per cent, 5.2 per cent and
5.5 per cent for USA, Australia and other countries respectively,
and a negative return for Canada. The surprisingly poor
performance for Canada is a combination of a relatively low
discovery rate coupled with a very high level of expenditure on
exploration. In detail, during the period 1985 - 1990, tax
incentives associated with flow-through financing, stimulated
$3.5 billion of spending on gold exploration in Canada. This was
equal to 30 per cent of all the entire western worlds expenditures
at the time. However, over this period only seven deposits larger
than 1 Moz were found in Canada.
Under the Forward Price Case, the value-to-cost ratio for
Australia rose to 1.51, which is higher than the 1.42 for USA,
and 0.98 for the Western World. This variance is due to
differences in timing of projects between the various countries.
early part of the study period when gold prices (and hence
forward prices) were very high and when no corporate taxes
were charged.
Company sector level
With regard to relative performance of different sectors of the
industry, Tables 1 and 2 indicate that Major Producers and
Moderate Producers collectively accounted for around 55 - 65 per
cent of total expenditures, 55 per cent of the major deposits found
by number, 67 per cent of the total ounces found and 68 - 70 per
cent of the value created. In comparison, Junior Companies
(optimistically including all of the value of the joint ventures with
major companies) accounted for 30 - 40 per cent of expenditures,
39 per cent of the total number found, 29 per cent of the ounces
and around 27 - 29 per cent of the value.
In terms of value created per dollar spent, it appears that the
Junior Companies did not perform as well as the Major
Producers.

378 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 DISCOVERY PERFORMANCE OF THE WESTERN WORLD GOLD INDUSTRY OVER THE PERIOD 1985 - 2003
This observation matches the results of a similar study by
Metals Economics Group (2000), which concluded that the
majors have been more effective at discovering significant new
orebodies than the juniors and intermediates. This study was
based on an analysis of gold and base metal discoveries made
between 1989 and 1998 with an in situ metal value greater than
$US 1 billion. In detail they found that the majors were
associated with 80 per cent of the discoveries by in situ value and
73 per cent of the discoveries by number yet only accounted
for 63 per cent of the total exploration expenditures.
It should be noted that the above data was based on discoveries
larger than 1 Moz. As mentioned before, the junior sector is quite
active in finding and developing deposits in the smaller size
range. Notwithstanding this, factoring in smaller deposits is
unlikely to affect the general observations of the present analysis.
This is because deposits less than 1 Moz in size made up less
than one-quarter of the ounces discovered and associated value
created (see Table 3).
DISCUSSION
The paper estimates that over the period 1985 - 2003 the Western
World gold exploration industry generated an average rate of
return of around five to seven per cent. This is of the same order
as the average cost of capital for the mining industry. This
implies that investing in exploration is, on average, a break-even
proposition.
It should be emphasised that the above returns are industry
averages only. Individual companies and projects performed
better. Also, these returns dont fully capture the option-value
associated with investing in gold. In particular, some people view
gold as a defensive play in their overall investment portfolio.
On a more general note, it is interesting to compare the returns
generated from gold exploration with other high-risk investments
such as the biotech industry.
A detailed study by Grabowski, Vernon and DiMasi (2002) for
the Pharmaceutical Research and Manufacturers of America
looked at the performance of the US biotech industry between
1990 - 1994. They found that only one out of 5000 to 10 000
compounds screened is ultimately approved as a new medicine;
that it takes an average ten to 15 years to bring a new drug to
market; and that seven out of ten drugs brought to market never
earn enough revenue to cover the average out-of-pocket cost of
drug development. They also determined that the average cost of
developing a new drug was $US 802 million pre-tax or around
$500 million after-tax. In comparison, the average NPV generated
was only $US 520 million resulting in a value-to-cost ratio of
0.65. This suggests that, on average, investing in the biotech
industry is also a breakeven proposition.
In their paper, Grabowski et al concluded that, the task is
getting harder as R&D costs are rising, success rates falling
and the time-to-market lengthening.
From the above it appears that, in spite of its glamour, biotech
companies face challenges equal to if not harder than that faced
by the minerals exploration industry.
CONCLUSIONS
Over the study period 1985 - 2003, the Western World gold
industry found an estimated 1127 Moz of gold in grassroots
discoveries. Over this period a total of $28.63 billion was spent
on grassroots and advanced exploration resulting in an average
finding cost of $25 per ounce of resource.
This gold was contained in 1000 deposits greater than 100 koz,
190 of which were in major deposits larger than 1 Moz. This
equates to an average cost of $150 million per major discovery
(containing, on average, 5.0 Moz).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Compared to the period 1970 - 1984, the average unit
discovery cost per ounce and per major discovery rose by a
factor of four and three respectively.
Over time, it appears that around 70 per cent of all major
discoveries were eventually developed as mines. For those
projects that became mines, half of them were developed within
6.3 years. Contrary to popular perception, the time taken to put
the project into production has remained unchanged over the last
three decades. Also, on average it took less time to develop a
grassroots project in Australia, USA and Canada than other
countries (5.4 versus 8.3 years).
Detailed financial modelling of 109 of the major deposits
found that many of these discoveries were very profitable, with
IRRs well in excess of 20 per cent. In general, the larger deposits
were more valuable with the most significant discovery made
being Yanacocha in Peru.
After including an adjustment for discoveries not modelled, it
was found that using historical spot gold prices, the industry
generated on average 73 cents worth of after-tax value for every
pre-tax dollar spent on exploration. This is a lower-bound
estimate as many producers hedged their production at higher
prices. As an upper bound, assuming that all gold production was
sold-forward at the decision-to-build stage, the average return
rose to 98 cents per dollar spent. These equate to a rate of return
of 4.8 to 6.8 per cent for the industry.
If the same deposits were found today, but developed using the
same production profile and costs as before, with all output sold
at a constant $US 400/oz, the average return would be 81 cents
per dollar spent.
Based on historical spot gold prices, companies exploring in
the United States and Australia performed significantly than
those in Canada generating 102, 78 and 25 cents of value
respectively per dollar spent.
With respect to major versus junior companies, it was found
that the latter performed better in terms of the number of deposits
found, but the former found more ounces and created more
value. It is estimated that, over the study period, Major
Producers and Moderate Producers collectively accounted for
around 55 per cent of the major deposits found by number,
67 per cent of the total ounces found and 68 - 70 per cent of the
value created. In comparison, Junior Companies (including joint
ventures with major companies) only accounted for 39 per cent
of the total number found, 29 per cent of the ounces and around
27 - 29 per cent of the value. However the junior companies are
estimated to have incurred 30 - 40 per cent of the total
exploration expenditures.
Notwithstanding the fact that investing in gold exploration on
average appears to be a break-even proposition, the industry
provides a similar return as other high-risk investments such as
biotech.
Finally, it should be recognised that the current paper only
looked at average returns for the industry. Individual companies
do much better or worse than this. The challenge is to find ways
to be in the former category. The hope of this paper is that it
provides some useful yardsticks for measuring a companys
performance along its path to success.
ACKNOWLEDGEMENTS
The author would like to thank WMC Resources Ltd for
permission to publish this paper.
It should be noted that the views expressed in this paper are
those of the author and do not necessarily reflect the current and
future business strategies for WMC Resources Ltd.
REFERENCES
Bouchard, G, 2003. Mineral Exploration, Deposit Appraisal and Mine
Complex Development Activity in Canada, Natural Resources of
Canada report, Ottawa, p 3.20.
379
R C SCHODDE
Brook Hunt and Associates, 2002. Gold Mines Cost Study, Surrey
England. To fill in gaps for the years 1985-90, the analysis drew on
earlier studies from the same source.
Cranstone, D, 2002. A history of mining and mineral exploration in
Canada and the outlook for the future, Natural Resources Canada
report, Ottawa, p 25.
Geoscience Australia, 2002. Data supplied for inclusion in the Hogan et al
(2002) report, p 82.
Grabowski, H, Vernon, J and DiMasi, J, 2002. Returns on Research and
Development for 1990s New Drug Introductions,
Pharmacoeconomics, 20, Dec supplement 3, 11-29.
Hogan, L, Harman, J, Maritz, A, Thorpe, S, Simms, A, Berry, P and
Copeland, A, 2002. Mineral Exploration in Australia: Trends,
Economic Impacts and Policy Issues, ABARE eReport 02.1,
Canberra, December, p44.
Mackenzie, B W, 1995. The mineral wealth of nations, in Proceedings
PACRIM Congress 1995, pp 351-358 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
Mackenzie, B W and Doggett, M D, 1992. Worldwide Trends in Gold
Exploration, Centre for Resource Studies working Paper 49, Queens
University, Ontario, January.
380 Adelaide, SA, 19 - 22 September 2004
Mackenzie, B W and Doggett, M D, 1995. Economics of Mineral
Exploration in Australia, unpublished report for Western Mining
Corporation, Adelaide. September.
Mackenzie, B W, Ortiz, F J and Doggett, M, 1997. Economic potential of
mineral exploration in Chile: Evidence from the historical record,
Centre for Resource Studies, Queens University, Ontario, April.
Maponga, O and Maxwell, P, 2000. The internationalisation of the
Australian mineral industry in the 1990s, Resources Policy,
26(4):199-210.
Maritz, A, 2003. Tax Incentive Options for Junior Exploration Companies,
ABARE eReport 03.4, for the Department of Industry, Tourism and
Resources, Canberra, March, p 6.
Metals Economic Group, 2000. Reassessing the roles of exploration and
acquisition in the 21st century mining company, Halifax, July.
Metals Economic Group, annual surveys 1992 to 2003. Corporate
exploration strategies: a worldwide analysis, Halifax.
Metals Economic Group, 2004. World Exploration Trends. Special
Report for the PDAC 2004 International Convention, Halifax, July.
Schodde, R C, 2003. Quantitative Analysis of the Risks and Rewards in
Australian Gold Exploration, presented at Australian Institute of
Geoscientists Conference, Perth, November.
PACRIM 2004
The Hillgrove Gold-Antimony-Tungsten District, NSW, Australia
C K Switzer1, P M Ashley2, B Hooper3 and B Roach3

ABSTRACT with current resources in the order of 340 000 ounces (Straits
website, 2004). Historical antimony production is estimated at in
Mineral occurrences of the Hillgrove Au-Sb-W district are hosted in late
Palaeozoic polydeformed, hornfelsed metasediments and excess of 50 000 metric tonnes. Over 2000 t of scheelite
Permo-Carboniferous granitoids of the New England Orogen. In excess concentrates have been produced from the field. The Hillgrove
of 204 individual occurrences have been identified to date with the mine ceased production in 2002 and went into receivership.
mineralisation developed as strike extensive (>20 km of known veining) Mineral occurrences of Au-Sb-W are hosted in late Palaeozoic
and potentially depth extensive steeply dipping fissures. These are polydeformed, hornfelsed metasedimentary rocks and
contained within an elongate area measuring some 9 km by 6 km in Permo-Carboniferous granitoids of the New England Orogen
surface dimensions. Recorded gold production of 720 000 ounces along (Boyle, 1990; Gilligan et al, 1992). In excess of 204 individual
with current resources in the order of 340 000 ounces highlight the
occurrences have been identified to date with the mineralisation
significance of this district. Historical antimony production is estimated
at in excess of 50 000 metric tonnes. Significant tungsten of over 2000 t developed as strike extensive (>20 km of known veining) and
in the form of scheelite has also been produced from the field. potentially depth extensive steeply dipping fissures. Mineralisation
Mineralisation is developed in veins, vein breccias, sheeted veins, is focussed in a north-west striking belt between the Chandler and
network stockworks and as alteration sulfide haloes to the main structures. Hillgrove faults (Figure 1).
The vast majority of fissures are subvertical and vary in widths of up to 20
m in places. Paragenetic studies have previously indicated that the earliest
mineralising event was a scheelite-bearing phase of quartz veining.
Subsequent phases of arsenopyrite-pyrite-quartz-carbonate veining were
accompanied by gold and minor base metal sulfides. Alteration is typically
sericite-ankerite-quartz. Overprinting stibnite-quartz veining with
gold-electrum, aurostibite and arsenopyrite form an important subsequent
phase. Veining can be inferred from historical records to extend for vertical
depths of over 1 km.
Preliminary structural studies highlight the potential transpressional
nature of the orogenic event associated with the Hillgrove mineralisation.
Significant north-south oriented shear and fracture zones occur in the
deeper exposed regions of field. Large north-east oriented faults with both
ductile and brittle characteristics also crosscut the district. The vast
majority of mineralised veining has a distinct north-west oriented strike.
Ore shoots are typically subvertical with internal movement indicators
suggesting a component of left lateral movement associated with the
latest stibnite-quartz-gold vein and breccia stage. Fibre lineations and
slickensides in the plane of the vein are horizontal in the most intense
vein zones, and appear to steepen to plunges of >20 where the
mineralisation is erratic suggesting an increased component of strain.
Syn-mineralisation lamprophyre dykes have an age range of
247 - 255 Ma (late Permian-early Triassic) and have temporal and
geochemical affinities to the voluminous Moonbi Suite of high-K
granitoids. Fluid inclusion studies have indicated homogenisation
temperatures in gangue quartz in the range 100 - 250C and that fluids
were of low salinity. Metal zonation within the field can be inferred on the FIG 1 - Location and geology of the Hillgrove region.
basis of past production to be from Au-As at depth to Sb-Au-As at
shallower levels, with minor scheelite occurring throughout the production
interval. Straits Resources purchased the deposit in 2004 and now
The Hillgrove gold-antimony-tungsten occurrences have strong controls 100 per cent of the mining leases and surrounding
affinities to many other orogenic gold-antimony deposits elsewhere, exploration tenements. An active exploration program is
particularly in New Zealand, except that the ore mineral assemblage is underway to define further resources to re-start the mining
telescoped, with earlier deeper level mineralisation overprinted at the operation. Previous exploration has only involved very minor
same structural level by later, shallower mineralisation. drilling from either underground or on surface and the potential
of the field is considered untapped by modern methods.
INTRODUCTION Exploration is focussing on defining the nature, width and grade
of the gold halo mineralisation. The previous operation focussed
The Hillgrove Mineral Field lies in north-eastern NSW, Australia more on antimony production, as stibnite concentrates, with gold
about 20 km east of Armidale (Boyle, 1990; Gilligan et al, production tending to be ancillary.
1992). The field covers nine by six kilometres and is dissected by
a deep gorge system of up to 500 m depth. The field has been
mined since the 1870s and produced over 720 000 ounces of gold GEOLOGY AND STRUCTURE
Mineralised vein and breccia systems at Hillgrove are hosted in
1. MAusIMM, Switzer Geological Services, 1 Maynard Drive, Bunya biotite-grade metamorphosed sedimentary rocks of the late
Qld 4055. Palaeozoic Girrakool Beds (originally shale, siltstone, argillite,
greywacke), biotite monzogranite (S-type) of the ~300 Ma
2. MAusIMM, Earth Sciences, University of New England, Armidale Hillgrove Adamellite and granodioritic-dioritic rocks of the early
NSW 2351.
Permian Bakers Creek Diorite Complex. The structures and
3. Straits Resources, PO Box 1641, West Perth WA 6872. mineralisation post-date, and are unrelated to any of the host

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 381

C K SWITZER et al

rocks. Syn-mineralisation lamprophyre dykes that are both cut by

mineralisation as well as having intruded mineralised structures,
are dated at 247 - 255 Ma (Ashley et al, 1994), thus bracketing
the potential age of mineralisation to approximately the
Permian-Triassic boundary. The dykes are closely related
spatially to mineralised structures, are up to a few metres wide
and include minette and vogesite types. Geochemically, the
lamprophyres are related to the high-K I-type granitoids of the
Permo-Triassic Moonbi Plutonic Suite (Ashley et al, 1994).
Nearly all of the mineralised structures of the Hillgrove region
lie between two major east-northeast striking regional structures,
the Hillgrove and Chandler Faults (Figure 1). These structures are
largely ductile and mylonitic in character, cutting the granitoids
and metasedimentary rocks. Metamorphic grade changes across
the Chandler Fault imply significant asymmetric uplift along the
northern side of the fault, occurring between 266 - 256 Ma
(Landenberger et al, 1995; Ashley and Craw, 2004). Structural
studies highlight the potential transpressional nature of the
orogenic event associated with the Hillgrove mineralisation. FIG 2 - Mineralised structure hosted in dilatational jog, with
Significant north-south oriented shear and fracture zones are quartz-stibnite-gold vein infill and adjacent vein breccia zone.
exposed in the deeper gorges of the region and may have a Width of image is approximately 1 m.
controlling role in the subsequent development of the mineralised
veining that is hosted in brittle structures (Ashley and Craw,
2004). The majority of mineralised veining has a north-west
oriented strike, with dips commonly 70 to vertical. Mineralised MINERALISATION
structures commonly pinch and swell, according to the presence of Mineralisation is developed in veins, vein breccias, sheeted
local dilatational sites (Figure 2), leading to large variation in veins, network stockworks and as alteration selvages of
widths of mineralised veins and breccias (from <1 cm to several disseminated and veinlet arsenopyrite and pyrite adjacent to the
metres). Ore shoots are typically subvertical with internal main structures. Although mineralised structures are commonly
movement indicators suggesting a component of left lateral <1 m wide, zones of stockworking and brecciation, plus their
movement associated with the latest stibnite-quartz-gold vein accompanying sulfidic halo zone may attain widths of up to 20 m
breccia stage. Fibre lineations and slickensides in the plane of the in places. Paragenetic studies (Boyle, 1990; Ashley and Craw,
vein are horizontal in the most intense vein zones, and appear to 2004) have previously indicated that the earliest mineralising
steepen to moderate plunges of >20 degrees where the event was a scheelite-bearing phase of quartz veining.
mineralisation is erratic suggesting an increased component of Subsequent phases of arsenopyrite-pyrite-quartz-carbonate
strain. Veining can be inferred from historical records to extend for veining were accompanied by gold and minor base metal
vertical depths approaching 1 km (Figure 3). sulfides. Alteration is typically sericite-ankerite-quartz and this

FIG 3 - Schematic cross-section from SW-NE across the Bakers Creek gorge at Hillgrove.
The main structures are steeply dipping and extend to depths of several hundred metres to 1 km.

382 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE HILLGROVE GOLD-ANTIMONY-TUNGSTEN DISTRICT, NSW, AUSTRALIA
FIG 4 - Three-dimensional image of underground workings and drillholes at Hillgrove, colour-coded with gold.
accompanies the disseminated and veinlet arsenopyrite and
pyrite zones. Overprinting stibnite-quartz veining with
gold-electrum, aurostibite and arsenopyrite form an important
subsequent phase. The sulfidic haloes about the mineralised
structures vary from being narrow and tight, to up to 20 m wide.
In these zones, it has been shown that gold grades (with little or
no accompanying Sb) can be significant, with gold hosted
invisibly in the sulfides. Arsenopyrite is the main host to
invisible gold, although a smaller proportion is also hosted in
pyrite that tends to be of arsenical composition (Ashley et al,
2000). The disseminated gold halo about mineralised structures
is being sampled and drilled in detail to define the grade. Mineral
concentrates derived from the sulfidic halo material have proven
to yield good gold recoveries from pressure oxidation treatment,
followed by conventional cyanidation.
The presence of the sulfidic halo about mineralised structures
and the other changes to wallrock mineralogy have led to the
development of geochemical alteration haloes that can extend for
up to tens of metres. The alteration-mineralisation process has led
to addition of Au, Sb, As, S, CO2, K and Rb, with depletion of Na
and Sr (Ashley and Craw, 2004). Fluid inclusion studies have
indicated homogenisation temperatures in gangue quartz in the
range 100 - 250C and that fluids were of low salinity (Comsti
and Taylor, 1984). Metal zonation within the field can be inferred
on the basis of past production to be from Au-As at depth to
Sb-Au-As at shallower levels, with minor scheelite occurring
throughout the production interval. Structural, alteration and
mineralisation characteristics of Hillgrove accord with many other
orogenic gold deposits, although Hillgrove is unusual in
potentially having formed progressively during orogenic uplift
leading to a telescoped array of vein systems with overprinting of
earlier mineralisation by later (eg W by As-Au by Sb-Au).
A detailed three-dimensional model of the previous mining and
sampling has been developed that is assisting in understanding of
the various mineralised systems. The veins cross cut lithology but
can vary markedly in their nature, vein styles, alteration, width and
grades of Au, Sb, As and W (Figure 4).
CONCLUSIONS
The Hillgrove gold-antimony-tungsten occurrences have strong
affinities to other orogenic gold-antimony deposits elsewhere,
particularly New Zealand, except that the ore mineral
assemblage is telescoped, with earlier deeper level mineralisation
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
overprinted at the same structural level by later, shallower
mineralisation. The mineralised vein systems have a strong
structural control and there is considerable potential for:
1. extension at depth;
2. extension along strike of known structures; and
3. discovery of new blind systems at depth or under regolith
cover.
There is also considerable opportunity to develop significant
gold production from the arsenopyrite-pyrite halo about
mineralised structures and to investigate production of antimony
(stibnite) concentrates again to take advantage of currently
buoyant world antimony prices. Straits Resources will for the
first time attempt to unlock the big picture of the Hillgrove
mineralised system and its potential to be a multi-million ounce
gold district.
REFERENCES
Ashley, P M, Cook, N D J, Hill, R L and Kent, A J R, 1994. Shoshonitic
lamprophyre dykes and their relation to mesothermal Au-Sb veins at
Hillgrove, New South Wales, Australia, Lithos, 32:249-272.
Ashley, P M and Craw, D, 2004. Structural controls on hydrothermal
alteration and gold-antimony mineralization in the Hillgrove area,
NSW, Australia, Mineralium Deposita, 39:223-239.
Ashley, P M, Creagh, C J and Ryan, C G, 2000. Invisible gold in ore and
mineral concentrates from the Hillgrove gold-antimony deposits,
NSW, Australia, Mineralium Deposita, 35:285-301.
Boyle, G O, 1990. Hillgrove antimony-gold deposits, in Geology of the
Mineral Deposits of Australia and Papua New Guinea (Ed: F E
Hughes), pp 1425-1427 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
Comsti, E C and Taylor, G R, 1984. Implications of fluid inclusion data
on the origin of the Hillgrove gold-antimony deposits, NSW, Bull
Proc Australas Inst Min Metal, 289:195-203.
Gilligan, L B, Brownlow, J W, Cameron, R G and Henley, H F, 1992.
Metallogenic study and mineral deposit data sheets Dorrigo-Coffs
Harbour 1:250 000 metallogenic map, Geological Survey of New
South Wales, Sydney, 509 p.
Landenberger, B, Farrell, T R, Offler, R, Collins, W J and Whitford, D J,
1995. Tectonic implications of Rb-Sr biotite ages for the Hillgrove
Plutonic Suite, New England Fold Belt, NSW, Australia,
Precambrian Research, 71:251-263.
Straits Resources Limited, 2004. http://www.straits.com.au
383
The Discovery and Early History of the Mt Leyshon Gold Deposit,
North Queensland
G S Teale1 and J E Lynch2
ABSTRACT
The Mt Leyshon gold deposit produced in excess of three million ounces
of gold and considerable silver over a period of approximately 130 years.
Most of this production occurred between 1987 and 2001. Mining
commenced in 1872, however it was highly likely that Chinese miners
were not only removing alluvial gold from the area but were also mining
very rich reef gold. Records for early alluvial and hard rock mining do
not exist, however gold production was probably considerable.
In the early 1980s Marathon Petroleum (Australia) Ltd (MPAL) joint
ventured into the Mt Leyshon area which had been held as a mining lease
by Noranda Australia Ltd since 1968. Noranda at that time was
concentrating its exploration effort on the Balcooma base metal prospect
areas to the north of Charters Towers. In early 1982 MPALs parent
company was taken over by US Steel. This led to an opportunity for
MPALs Australian management to purchase all tenements and licences
(except for coal, oil and gas) which were duly purchased for $US 1
million. At this time MPAL had a remarkable array of gold tenements
(many of which were subsequently mined by other companies) and Mt
Leyshon was considered to be the highlight. Early seed capital ($A 3.1
million) was raised with backers being the Bank of America, the NSW
Government Superannuation Scheme, Westpac, private individuals and a
Queensland grocery chain by the name of Jack the Slasher.
The Pan Australian Mining Ltd prospectus was registered for the issue
of 28 million shares of 25 cents each and a premium of 15 cents to raise
$A 11.2 million in late 1984. The highlight of the prospectus was the
low-grade reserve of 16.5 million tonnes @ 1.4 g/t Au at Mt Leyshon. In
house it was estimated that the deposit contained in excess of six million
tonnes @ 2 g/t of oxidised ore which was amenable to heap leaching. It
was also known at this time that less than one quarter of the deposit had
been delineated. Drilling was continuing with some exceptional
intersections reported.
In mid 1986 Pan Australian Mining Ltd purchased Noranda Australia
Ltds 50 per cent share in Mt Leyshon for $A 5.75 million. This would
later rate as one of the best buys in decades. In its first 12 months of
operation the deposit produced approximately twice the amount of gold
that had been produced in the previous 100 years. The mine was
ultimately taken over and controlled by Normandy Poseidon in 1989.
INTRODUCTION
Mt Leyshon first produced gold in 1871, although Chinese miners
prior to this were more than likely mining alluvial, eluvial and
some high-grade, oxidised and enriched gold veins. Until 1916
total production from the area was approximately 46 700 oz of
bullion (85 per cent gold) from approximately 200 000 t of ore
(Orr, 1995). These figures however are most likely grossly
underestimated. The gold mineralisation is contained within the
Mt Leyshon Breccia Complex (MLBC; or the Mt Leyshon
intrusive and breccia complex of Orr, 1995) of which the Mt
Leyshon Breccia (Orr, 1995) is the major host to gold
mineralisation. The geology of the MLBC has been discussed by
Morrison et al (1987), Paull et al (1990) and Wormald et al (1991)
and the ore mineralogy is discussed in a later section of this paper.
The Mt Leyshon area (Figures 1 and 2a) had been held as a
mining lease by Noranda Australia Ltd (Noranda) since 1968.
MPAL joint ventured into the property in 1980. The property had
1. GAusIMM, Teale and Associates Pty Ltd, PO Box 740, North
Adelaide SA 5006.
2. FAusIMM, Lynch Mining Pty Ltd, PO Box 524, Fortitude Valley,
Qld 4006.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
been initially targeted by MPALS senior gold geologist (Mr J
Schupp) as a breccia property worthy of further exploration.
MPAL demonstrated the disseminated low-grade nature of the
gold mineralisation in an initial drill program in 1981. Noranda
subsequently resumed management in May 1982 when the
potential for high-grade, heap leachable ore was recognised by
mining studies undertaken by MPAL. PAM after its formation in
November 1982 pursued the exploration, evaluation and
metallurgical testwork, along with Noranda, until a decision to
commence a feasibility study in March 1986. PAM purchased
Norandas 50 per cent share in May 1986 and commenced
construction of a much modified development plan in July 1986
using Minproc Engineers. The plant was completed in November
and the first gold poured in February 1987. PAMs share price
lifted from $0.25 in mid 1985 to $5.00 in May 1987, capitalising
the company at $400 million!
THE EARLY HISTORY OF THE MT LEYSHON
OREBODY, 1870 - 1980
Mt Leyshon had tantalised prospectors and miners alike from the
time of its discovery in 1871. The discovery predated the major
gold finds at Charters Towers by Mossman, Fraser and Clarke in
1872 who may have visited the Mt Leyshon area prior to heading
to Charters Towers (Orr, 1995). Coarse gold and nuggets were
taken from creeks and gullies around the mountain (by both
Chinese and later workers) with nuggets up to 143 oz even being
located in the 1920s (Hooper, 2002).
The overall low grades of ore hindered mining in the 1800s
and early 1900s and only higher grade veins were mined with
mining generally stopping at the water table. Detailed discussion
of mining in the early years is presented in Hooper (2002) and
also discussed by Orr (1995).
The area had been viewed as containing a potential copper
resource and had also been explored as a porphyry molybdenum
target. Limited drilling programs, for copper, were undertaken by
Mount Isa Mines and Noranda in the late 1960s and 1970s but no
economic resource was discovered (Orr, 1995). The work did
however confirm the presence of widespread but low-grade gold
mineralisation.
THE DISCOVERY OF THE MT LEYSHON
OREBODY
MPAL had been aware of the potential for breccia hosted gold
mineralisation and an initial drilling program was undertaken to
evaluate Mt Leyshon. The takeover of MPALs parent company
Marathon Oil by US Steel in 1982 enabled the then management
of MPAL (Mr Robert Bryan, General Manager and Mr John
Lynch, Exploration Manager) to recognise an opportunity and
bought all MPALs mineral tenements and licences for
$US 1 million. Early seed capital ($A 3.1 million) was raised
from a number of companies including Bank of America, NSW
Government Insurance Office (GIO), Westpac, and private
investors.
PAM was floated in late 1984 and raised $A 11.2 million. The
highlight of the prospectus was obviously Mt Leyshon with
resources of 16.5 million tonnes @ 1.4 g/t Au. The company at
this time recognised that less than one quarter of the deposit had
385
G S TEALE and J E LYNCH

FIG 1 - Location and regional geological setting, Mount Leyshon gold deposit. Note the concentration of gold occurrences in the NE trending
corridors, which also contain many of the Permo-Carboniferous subvolcanic complexes. (Paull et al, 1990).

been outlined by drilling and included in this reserve was a
resource of six million tonnes @ 2 g/t Au which was amenable to
heap leaching. Based on this initial resource, Mt Leyshon mine
became one of the earliest heap leach operations in Australia (see
Figures 2e and 2f).
The deposit, in its first year of operation, produced
approximately twice the amount of gold that had been produced
in the previous 100 years and paid back its initial capital within
three months of commencement of mining. The mine went from
strength to strength with broad zones of mineralisation being
intersected in drilling which highlighted the bulk tonnage
potential of the orebody (eg MLD90, 94 m @ 3.8 g/t Au;
MLD95, 166 m @ 2.1 g/t Au; MLD99, 94 m @ 4.4 g/t Au;
MLD103, 218 m @ 3.3 g/t Au). The Mt Leyshon deposit had
been discovered having been turned from a possible porphyry
copper/molybdenum prospect to a world-class gold orebody. Mt
Leyshon produced in excess of three million ounces of gold.
Pyrite is the most common sulfide and is present as an
alteration phase, along with sericite, in the early felsic
porphyries, in numerous varieties of breccia clast and in many of
the altered intrusive/extrusive rock-types. It is present in all
styles of mineralisation. Analysed pyrite contained trace amounts
of Zn and Cu but no As, Ni or Co. Gold in solid solution in pyrite
was not detected. Early formed pyrite (either in veins, matrix or
disseminations) contains abundant hematite and Fe-chlorite
inclusions with many of the later veins (especially gold-bearing
variants) having a selvedge of pyrite (which contains hematite
inclusions) which abuts intensely chloritised zones within the
breccia.

THE MT LEYSHON OREBODY

ORE MINERALOGY AND CHEMISTRY
Gold mineralisation within the MLBC (Figure 2g) is associated
with dominant pyrite, chalcopyrite, galena and sphalerite with
numerous lesser sulfide/sulfosalt phases. These occur with a
gangue of predominantly chlorite, carbonates, micas and quartz
in replacement, cavity-fill, disseminated and vein style
mineralisation. The latter appears to contain higher gold grades
which are associated with a distinctive suite of Bi-Pb-
Ag-Cu-bearing minerals. Highly oxidised veins of this type may
have been mined by Chinese workers during the early eluvial and FIG 2a - Aerial view of Mount Leyshon prior to commencement of
alluvial mining of the area. mining (1983).

386 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 THE DISCOVERY AND EARLY HISTORY OF THE MT LEYSHON GOLD DEPOSIT, NORTH QUEENSLAND

FIG 2b - Aerial view of operations after commencement of CIP

operations (1990). FIG 2f - Irrigated leach pads (1987).

FIG 2c - Early access roads, above the abandoned White Face

Mine workings and tailing (1983).

FIG 2g - High-grade ore (NQ core) from the Mount Leyshon

Breccia (27.5 ppm Au).
FIG 2d - Pre mine stripping of ore and waste (1986).

FIG 2e - Loading of first leach pads at Mount Leyshon (1986).

Mount Leyshon was one of the earliest heap leach operations FIG 2h - Crackle breccia developed adjacent to the Mount Leyshon
in Australia. Breccia (hand for scale).

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 387

G S TEALE and J E LYNCH
FIG 3a - Complex intergrowth of galena and lindstromite
(Pb3Cu3Bi7S15) from DDH-MLD5-95.4 m. The lindstromite is the
light coloured phase. (Mag 600; reflected light, nicols partially
crossed; 1 cm = 0.025 mm).
FIG 3b - Wispy gold cutting across an aikinite grain. Larger gold
grain at right of photo contains approximately ten per cent Ag and
is 0.2 mm in length (Mag 600; reflected light; 1 cm = 0.025 mm).
Sphalerite has an iron content of from approximately two per
cent to 12 per cent with higher iron sphalerite possibly present.
Its colour in transmitted light ranges from deep blood red to a
pale honey yellow. It is a ubiquitous but not overly abundant
phase in gold-bearing veins where it occasionally includes gold
in chalcopyrite-rich domains. It is abundant in pyrite-poor (and
gold-poor) galena-sphalerite (minor chalcopyrite) veins as well
as galena-rich matrix replacement material. Manganese contents
in the sphalerite range up to 0.5 per cent and cadmium contents
are in the 0.15 per cent to 0.3 per cent range. Some gold-bearing
veins contain iron-rich sphalerite as inclusions in pyrite with
iron-poor sphalerite co-existing with (lower temperature)
chalcopyrite, gold and bismuth-bearing sulfosalts. Galena is
present in vein, matrix, replacement and cavity-fill
mineralisation styles and is occasionally observed, along with
sphalerite, intergrown with calcite in sericite altered feldspar
phenocrysts of the early porphyries. Its composition is variable,
being almost pure PbS in sphalerite-galena veins and matrix
replacement material through to a complex PbS solid solution
(PbSSS) when associated with Bi and Au-bearing veins and
cavity-fills. The PbSSS can contain up to 5.5 per cent Bi, 2.1 per
cent Ag and minor Cu. Gold is not present in solid solution in
either galena or sphalerite however wisps of gold in some
myrmekite-like galena-aikinite intergrowths (Figure 3b) may
represent unmixing. Microprobe homogenous intergrowths of
matildite-galena are present in early, higher temperature
chalcopyrite-molybdenite-gold Bi-bearing sulfosalt veins.
388 Adelaide, SA, 19 - 22 September 2004
Chalcopyrite occurs with all mineralisation styles and when
associated with Bi-bearing phases often contains inclusions of
Ag-rich gold. It contains up to 1.7 atom per cent Zn in solid
solution and has an excess of Fe atoms over Cu atoms. Up to
700 ppm silver has been recorded in chalcopyrite.
Bismuthinite-aikinite series (Bi2S3-CuPbBiS3) minerals are a
major associate of gold mineralisation at Mt Leyshon. In addition
to bismuthinite and aikinite the phases hammerite (Pb2Cu2Bi4S9),
lindstromite (Pb3Cu3Bi7S15), friedrichite (Pb5Cu5Bi7S18) and
pekoite (PbCuBi11S18) have been recorded. In association with
galena these phases can occur as spectacular and complex
intergrowths (eg see Figure 3a) with separate intergrowth phases
often containing a third, finer grained exsolution phase.
Silver-rich wittichenite was noted as an intergrowth phase with
friedrichite and galena in one sample. The various phases contain
negligible Sb (up to 0.1 per cent) and Ag (up to 900 ppm). The
above tend to be confined to chalcopyrite-rich veins or
cavity-fills with sphalerite and galena occurring as minor phases.
The latter, when present, is often complexly intergrown with the
numerous Bi2S3-CuPbBiS3 phases. The sphalerite, when present,
often contains inclusions of gold and is Fe-poor and choked with
chalcopyrite exsolution/inclusion bodies. Within chalcopyrite the
phases occur as large lenticular, skeletal grains (generally
associated with coarse gold), as small flame-like bodies, as
rounded inclusions and as intergrowths with chalcopyrite. They
anneal fractured pyrite, are included in pyrite, and form
lamellar-like bodies along sphalerite cleavage planes. Cosalite
and associated phases may be present and wittichenite (often in
association with covellite) generally rims and replaces the
various phases, via a Bi-sulfosalt-chalcopyrite reaction.
As with the bismuthinite-aikinite series minerals the other
Bi-bearing minerals or groups occurring at Mt Leyshon are also
intimately associated with gold. Benjaminite
((Ag,Cu)3(Bi,Pb)7S12) to gustavite (PbAgBi3S6) series, emplectite
(CuBiS2) to wittichenite (Cu3BiS3) minerals, ?pavonite
(AgBi3S5), berryite (Pb2(Cu,Ag)3Bi5S11)-lillianite
(Pb3Bi2S6)-cosalite (Pb2Bi2S5) minerals are present and many
contain significant silver. Cosalite and associated phases are
usually observed only when coarse gold grains are present. They
are markedly anisotropic, probably due to the presence of silver
and copper in the lattice. They occur in chalcopyrite and are
invariably rimmed by (Ag-rich) wittichenite. Myrmekite-like
intergrowths of cosalite and wittichenite have been observed and
analysed and this intergrowth is very rarely reported in the
literature. The Mt Leyshon cosalites contain up to 5.2 per cent
Cu (possibly an intergrowth present) but generally around 2.5 per
cent Cu and have Ag contents of from 0.4 per cent to four per
cent. A plot of the cosalites indicates a drift away from the
theoretical formula (Pb2Bi2S5) to such compositions as
Cu1Ag0.16Pb5.8Bi4.9S14. The possibility exists that this is a new
sulfidic phase or a microprobe homogenous growth.
Silver contents of the various phases mentioned above are
variable with cosalite (up to four per cent), a pavonite-like
mineral (up to eight per cent), gustavite (up to 7.5 per cent) and
wittichenite (up to three per cent) all containing significant silver.
The benjaminite, gustavite, ?pavonite minerals appear to occur
only in chalcopyrite-molybdenite-gold (-minor cosalite) veinlets
which may represent a higher temperature, early formed
gold-bearing paragenesis. This paragenesis may equate to the
Stage 1 laminated quart-molybdenite stockworks described by
Paull et al (1990).
Gold is present as predominantly free gold (including
electrum) throughout the deposit with only the (?) pavonite
containing detectable gold (1200 ppm). Most gold is observed in
veins, veinlets or sulfide-rich cavity-fills, in an intimate
relationship with bismuth-bearing minerals (Teale, 1987). The
gold grains predominantly fall within the one to 600 micron
range, with a marked concentration from one to 20 microns. A
significant number of grains (primary) reaching 1 mm to 1.5 mm
have been observed in drill core with the largest (primary) grain
PACRIM 2004
 THE DISCOVERY AND EARLY HISTORY OF THE MT LEYSHON GOLD DEPOSIT, NORTH QUEENSLAND
recorded being 1.1 cm in length. The silver content of gold
ranges from approximately three per cent to 31 per cent with
silver contents increasing in the later parageneses (Teale, 1987).
Gold inclusions in pyrite from early pyrite-rich veins usually
contains from three per cent to seven per cent Ag. Early formed,
disseminated pyrite (associated with phyllic alteration) is almost
always free of gold inclusions, except in the early porphyries.
Here gold grains (five to ten micron) can be located within the
core areas of pyrite well away from any mineralised vein
structures. This pyrite can also contain inclusions of sphalerite
and chalcopyrite and the gold grains contain inclusions of
pyrrhotite. The disseminated pyrite is associated with
sericite-calcite-anatase fluorite (replacing biotite) and
calcite-sulfides (replacing feldspar phenocrysts). In the latter,
galena-sphalerite-chalcopyrite are often intimately intergrown
with carbonate and are considered to have developed during the
main (phyllic) phase of porphyry alteration. Miarolitic cavities
are common in this sequence indicating high initial volatile
contents. The presence of apatite as a gangue phase in the later
vein mineralisation crosscutting the porphyry is unusual and
has not been noted elsewhere in the complex. The gold observed
in some disseminated pyrite in early porphyries is possibly some
of the earliest formed gold in the MLBC.
More silver-rich gold grains from later veins, cavity fills and
matrix replacements can exhibit zoning (to more silver-rich rims)
and can contain up to 0.7 per cent Cu. No other elements, for
example mercury or tellurium were detected in the gold. As
mentioned there is a tendency for gold to become more
silver-rich in the late parageneses with the most silver-rich gold
(electrum) observed healing fractured pyrite in bismuth-rich vein
assemblages. Gold associated with bismuth sulfides and
sulfosalts, either as inclusions or at grain boundaries accounts for
approximately 84 vol per cent of the total gold observed (Teale,
1987).
PAN AUSTRALIAN MINING LTD
SUCCESS THROUGH PERSISTENCE
The Mt Leyshon orebody was developed by PAM despite a series
of events that almost halted the process. These included:
1. MPAL almost discarding the project in April 1982 as being
too low grade.
2. Noranda sought the right of first refusal when MPAL
transferred its interest to PAM. This was refused by the then
Treasurer, Mr Paul Keating.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
3. Scepticism within the investment community that:
Mt Leyshon was too low grade,
heap leaching would not work in the tropics,
the management of PAM was too inexperienced to
develop the mine, and
PAM could raise the capital required to fund development.
4. The initial reserve calculations had extensive tonnage but
were uneconomic at the time. Management was able to
demonstrate its belief that within the low-grade reserve
there existed a higher grade, heap leachable resource.
5. The acquisition of Norandas 50 per cent share was made at
a time of depressed gold prices which enabled PAM to
purchase 100 per cent of the deposit.
In summary the development of the extensive Mt Leyshon
gold deposit by a fledgling Australian gold exploration company
was the result of technical understanding of a complex
geological situation, a corporate culture that encouraged
geological, mining and metallurgical ideas and the lateral
thinking to develop mining and financial models that enabled the
project to be financed.
REFERENCES
Hooper, C, 2002. Mt Leyshon, 120 p (Imageprint: Townsville).
Morrison, G W, Teale, G S and Hodkinson, I, 1987. Geology and gold
mineralisation at Mount Leyshon, north Queensland, in Proceedings
Pacific Rim Conference 1987, pp 777-780 (The Australasian Institute
of Mining and Metallurgy: Melbourne).
Orr, T H, 1995. The Mount Leyshon gold mine: geology and
mineralisation, in Mineral Deposits of Northeast Queensland; Geology
and Geochemistry, 17th IGES, May, Townsville, Queensland.
Paull, P G, Hodkinson, I, Morrison, G W and Teale, G S and 1990. The
Mt Leyshon gold deposit, in The Geology of the Mineral Deposits of
Australia and Papua New Guinea (Ed: F E Hughes), pp 1471-1481
(The Australasian Institute of Mining and Metallurgy: Melbourne).
Teale, G S, 1987. Mineragraphic and metallurgical aspects of the Mt
Leyshon orebody, unpublished company report.
Wormald, P J, Orr, T H, Hodkinson, I P, 1991. The Mt Leyshon gold
mine (NE Queensland), an intrusive breccia and igneous complex, in
Proceedings World Gold Conference 1991, pp 223-232 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
389
A Search for Gold in Indonesian New Guinea
M R Thirnbeck1
ABSTRACT
The Indonesian province of Papua is a remote jungle-clad region covering
the western half of the mountainous island of New Guinea. Despite being
relatively under-explored to this day, Papua is home to the greatest
mineral success story of the 20th century. The Grasberg/Ertsberg minerals
district atop the central dividing range contains the worlds largest reserve
of both copper and gold (Freeport-McMoRan Copper and Gold, 2002).
With large mines at Ok Tedi, Porgera and Lihir, and many
well-delineated resources such as Frieda River, Wafi and Hidden Valley,
neighbouring Papua New Guinea, the eastern half of New Guinea has
enjoyed sustained exploration campaigns for gold and copper since the
mid 1960s. The combined Grasberg and PNG exploration successes can
be measured in terms of ounces of gold produced over the last decade
versus available landmass, confirming the island of New Guinea and
surrounding islands as a world-class destination for gold explorers.
Yet despite this obvious regional prospectivity and the profitability of
the Grasberg operation, a sustained exploration effort in the Papua
province in Indonesia has not been achieved. Periodic political
constraints, poor infrastructure, high costs, operational challenges, social
and environmental concerns and lack of immediate exploration success
often impact exploration investment decisions. Accordingly a lesser
number of mineral occurrences have been documented in Papua
compared to those seen along tectonic strike in geologically similar
Papua New Guinea.
In seeking to remedy this geological disparity, Mutiara Resources
Corporation in 1994 embarked on acquiring large tracts of mostly
unexplored areas in Papua province to the north of the central range. A
total of 19 blocks consisting of 12 projects were acquired, covering an
area of 82 736 square kilometres. Joint Ventures were formed over
11 projects and generally exploration activities were funded by the major
partner. Between 1994 and 1998 more than $US 15 million were spent
conducting regional reconnaissance exploration primarily for gold and
copper over most of these areas.
Exploration involved maintaining numerous base-camps to aid
helicopter-supported regional stream sampling of approximately 6500
sites. Interpretation of Landsat imagery assisted in constructing regional
geological maps and identified areas warranting closer inspection.
Airborne magnetic surveys, totalling more than 85 000 line km, were
flown over selected areas totalling 41 400 square km or 50 per cent of the
combined project area.
Fallout from the Busang fraud of 1997, coupled with economic and
political turmoil that beset Indonesia shortly thereafter, curtailed
grass-roots exploration activities throughout much of Indonesia.
Increasing land-holding costs and drastically reduced exploration budgets
led to selected project terminations, major relinquishments and the
withdrawal of all original Joint Venture partners by 1999.
Many gold stream anomalies generated during the 1994 - 1998 period
were never investigated. Hard-rock gold occurrences discovered at
Kwatisore, Wasewa, Supiori, Kuiri, Biber and Waena were all
relinquished prior to complete evaluation. Only the 108 600 hectare
Idenburg project, less than 1.5 per cent of the original area, has been
retained into late 2004. Exploration at Idenburg has identified lode type
gold mineralisation of mesothermal origin.
This paper presents an exploration case history, describing the
exploration work undertaken, and the principal findings of this decade-long
campaign.
INTRODUCTION
The island of New Guinea beckons exploration and discovery. At
hotels and airport lounges throughout the Indonesian province of
1. MAusIMM, President Director, PT Iriana Mining Services, Gedung
Graha Simatupang, Tower I, Block D, 8th Floor, Jl TB Simatupang
Kav 38, Jakarta 12540, Indonesia. E-mail: thirnbeck@cbn.net.id
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Papua (formerly Irian Jaya) you often bump into geologists,
botanists, zoologists or anthropologists willing to talk about their
experiences in this frontier land. However for the first time since
exposure of the Busang fraud in 1997 we are starting to see
increasing numbers of geologists sniffing around in Papua
always a reliable indicator that any exploration upswing has
teeth. Like many high-cost frontier areas, Papua province only
sees upsurges in mineral exploration activity during the peak of
any exploration investment cycle.
Neighbouring Papua New Guinea has enjoyed more consistent
exploration efforts since the mid 1960s, partly fueled by regular
exploration and mine development successes. Sustained
geological investigations culminated in major copper and/or gold
mine developments at Bougainville, Porgera, Ok Tedi, Lihir and
Misima along with the delineation of significant deposits at
Frieda River, Wafi and Hidden Valley.
Exploration successes on the Indonesian half of the island of
New Guinea have been more sporadic but nevertheless
momentous. PT Freeport Indonesia commenced production of
copper and gold at Ertsberg in 1972, some 36 years after the
mountain of ore was first described by Dutch explorers. A few
nickel laterite deposits were also delineated around Jayapura and
on Gag and Waigeo islands in the early 1970s. But there was
little else to talk about until Freeports 1988 discovery of the
world-class Grasberg porphyry (Potter, 1996) just 3 km
northwest of the Ertsberg mine. The development of the
Grasberg mine coincided with the evolution of the Indonesian
Contract of Work tenure system and over two decades of relative
political and fiscal stability under President Soeharto. The
impressive ability to ramp copper and gold production at
Grasberg ensured that by the mid 1990s all major global gold
and copper miners had raised their in-house knowledge of the
Papua province and its mineral investment opportunities. A
notable success story from the Grasberg-triggered 1990s
exploration upsurge in Papua was the discovery and delineation
of the more than ten million ounce Wabu Ridge gold deposit by
PT Freeport Indonesia (OConnor et al, 1999).
In the early 1990s Mutiara Resources Corporation (MRC) was
engaged in the timber industry with logging concessions inland
from the Papua north coast feeding a plywood factory on Biak
Island. A chance meeting in a Nabire hotel between a timber
tycoon and a freelance geologist quickly converted opportunity
into reality as MRC embarked on a wave of mineral project area
applications throughout Indonesia, with a primary focus in Papua
province.
By 1996 MRC had assembled project areas covering more
than 80 000 square kilometres within northern Papua and had
facilitated the entry of major mining companies into the
Indonesian mining sector via Joint Venture agreements. MRC
took its Papuan mining project interests to the Canadian market
and in late 1996 listed on the Toronto and Vancouver Stock
Exchanges as Iriana Resources Corporation (IRC). Exposure of
the Busang scam in early 1997 badly affected the venture and the
ensuing inability to raise finance for an Indonesian gold play
listed on Canadian exchanges ensured only limited piecemeal
exploration was possible in the following years until IRC
suspended trading on the Toronto Stock Exchange in April 2002.
The remaining projects were re-organised back into an MRC
affiliate in lieu of the MRC-provided shareholder loan that
sustained IRC from early 1999.
391
M R THIRNBECK

REGIONAL GEOLOGY
The island of New Guinea is the product of interaction between
the northward moving Australian continental plate and oceanic
plates related to the general southwest drift of the Pacific Plate
(Figure 1). Understanding of the evolution of Australias
northern and eastern margin has been hampered by incomplete
regional mapping in north central New Guinea, however new
isotopic and age data support evidence for an Early to Late
Triassic volcanic arc in northern New Guinea (Crowhurst et al,
2004). Generally, rocks with oceanic affinities predominate in
the north of the island and incorporate more Australian
continental material southwards towards the Central Range. Dow
et al (1988) describe rock types of the Pacific Plate category to
include upfaulted mantle material of presumed Mesozoic age,
Jurassic oceanic crust and Palaeogene island arc volcanics and
subordinate oceanic crust. These Pacific Plate rocks north of the
Central Range are mostly covered by an uplifted and folded,
mid-Tertiary to Quaternary flysch sequence. All of MRCs
projects were north of the Central dividing mountain range
which spines Papua.
THE PROJECTS
The project area blocks (Table 1) are shown in Figure 2 and are
described generally from west to east.

FIG 1 - Structural elements of Papua province, Indonesia.

TABLE 1
Project details and JV partners.
Project Area (km2) COW company name Major partner Status
Ransiki 9550 PT Barrick Mutiara Ransiki Barrick Gold (1996 - 1997) COW terminated in 1997
Kwatisore 3254 COW not formed Aurora Gold (1994 - 1996), IRC (1996 - 1997) COW application returned in 1997
Supiori 563 COW not formed IRC (1998 - 2000) COW application returned in 2000
Waropen 10 003 COW not formed Barrick Gold (1995 - 1997) COW application returned in 1997
Zeiwa 418 COW not formed Barrick Gold (1996 - 1997), IRC (1997 - 1999) COW application returned in 1999
Iriana 16 109 PT Iriana Mutiara Mining Battle Mountain (1994 - 1999), Applied for termination in 2004
IRC (1999 - 2002)
Van Daalen 5177 PT Iriana Mutiara Van Daalen Barrick Gold (1995 - 1997), IRC (1997 - 1999) COW terminated in 1999
Van Daalen North 302 COW not formed Barrick Gold (1996 - 1997), IRC (1997 - 1999) COW application returned in 1999
Mamberamo 8434 COW not formed Battle Mountain (1995 - 1997) COW application returned in 1997
Sentani 146 PT Iriana Sentani Western Mining Corp (1995 - 1997), COW terminated in 2003
IRC (1997 - 2002)
Senggeh 14 330 PT Iriana Senggeh Morrison Knudsen (1994 - 1995) Cyprus Amax COW terminated in 2003
(1995 - 1998), IRC (1998 - 2002)
Idenburg 14 450 PT Iriana Mutiara Idenburg Barrick Gold (1994 - 1997), IRC (1997 - 2002) Year II of Exploration Period

392 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


FIG 2 - Location of MRC/IRC project areas in Papua province, Indonesia.
RANSIKI PROJECT
The Ransiki project covered 955 000 hectares of the coastal strip
along the western shores of Cendrawasih Bay, and evolved into a
sixth generation contract of work signed on 28 April 1997 by PT
Barrick Mutiara Ransiki. Barrick Gold Corporation held an
85 per cent interest in the project and under the terms of the JV
funded and conducted all exploration. Barrick withdrew from the
Ransiki JV in 1997 and the project was terminated shortly
thereafter.
Photogeological interpretation from Landsat imagery assisted
in identifying four lithotectonic domains based on regional
differences in structural style and litho-unit content, from north
to south, the Arfak, Kemum and Birds Neck Domains and the
Lengguru Fold Belt (Barrick Mutiara Ransiki, 1997). The Arfak
Domain is dominated by island arc volcanics at the southern
margin of the Pacific Plate and is separated from the Kemum
domain by the Ransiki Fault System. The Kemum Domain is an
elevated block of metamorphic basement of the northern margin
of the Australian Plate. The Birds Neck Domain and Lengguru
Fold Belt are folded and thrusted Mesozoic and Tertiary cover
sequences of the Australian Plate and are differentiated from
each other on the basis of structural style. This region represents
a product of escape tectonics resulting from the collision
between the Australian plate and the remnants of volcanic belts
carried by the Pacific plate (Pubellier and Ego, 2002).
Exploration completed by the JV consisted of a single program
of boat and vehicle supported drainage sampling from 84 sites
and reconnaissance geological mapping. At each drainage
sample site -40# BLEG, -60# stream sediment and panned
concentrate samples were collected. Stream sediment samples
were assayed for gold, silver, arsenic, copper, lead, zinc,
molybdenum, bismuth, antimony, mercury, nickel, cobalt and
chromium. Sampling was concentrated along the Ransiki and
Seiy rivers and along all rivers draining into Cendrawasih Bay.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
A SEARCH FOR GOLD IN INDONESIAN NEW GUINEA
Twelve stream sediment samples and five BLEG samples
returned slightly gold anomalous results. No panned concentrates
and none of the rock samples gave anomalous (>0.1 ppm Au)
gold assays. No further follow-up investigations were undertaken
before the project termination in 1997 due to lack of immediate
exploration success, coupled with the inability to attract a new
investor after the withdrawal of Barrick Gold from the JV.
Boulders containing alunite were recorded in a creek with
anomalous copper (124 ppm) and gold (0.18 ppm) stream
sediment geochemistry draining volcanic rocks of the Arfak
Domain. The reported occurrence of alunite-bearing tuffaceous
floats at 13412.47E and 123.91S have yet to be investigated.
KWATISORE PROJECT
The Kwatisore project evolved into a sixth generation COW
application covering approximately 325 400 hectares on the
southwestern shore of Cendrawasih Bay, centred 80 km west of
Nabire. A northwest trending package of Palaeozoic and Mesozoic
sedimentary, volcanic and metamorphic rocks cover most of the
project area and are thrust southwesterly upon Tertiary rocks of
the Lengguru Fold Belt (Robinson et al, 1990) exposed along the
southwestern project boundary. A large pluton of the Triassic
Kwatisore Granite occurs in the southeastern part of the project
area. Structure is dominated by the Wandamen fault zone, which
trends north-northwest through the centre of the block.
MRC first scouted the area in 1992 (Keleher, 1993a) and
discovered gold mineralisation in narrow quartz sulfide veins
exposed in a road cut at 13451.8E and 321.34S along the main
logging access road into the area. A two-metre horizontal channel
rock chip assayed 4.83 g/t Au, 41 g/t Ag and 0.16 per cent Pb. A
nearby interval returned eight metres at 2.40 g/t Au, 46 g/t Ag and
0.52 per cent Pb. The veins were steeply dipping, mostly less than
50 cm thick and hosted in graphitic phyllites near the western
contact of the Kwatisore Granite.
393
M R THIRNBECK
A JV was formed with Aurora Gold Ltd who sole funded and
conducted reconnaissance investigations in late 1995 as part of
conditions to earn 85 per cent of the proposed project company
to be formed upon signing the COW. Exploration was helicopter
supported and involved collection of 213 BLEG, 204 stream
sediments and 98 rock chip samples. The BLEG samples were
analysed for gold, silver and copper. Stream sediment samples
were analysed for gold, silver, arsenic, antimony, mercury,
copper, lead, zinc, bismuth, molybdenum, barium and tungsten.
No anomalous results of significance were encountered during
the initial reconnaissance phase and Aurora withdrew from the
project in 1996. MRCs publicly listed vehicle IRC elected not to
proceed with forming a COW project and returned the
application area back to the Government in late 1997.
SUPIORI PROJECT
The Supiori project was an application for an eighth generation
COW covering 56 320 hectares over Supiori Island to the west of
Biak. IRC applied for the project area after inspecting an alluvial
gold rush that sprung up in late 1998. Hundreds of artisanal gold
miners swamped a number of southerly draining creeks to the
west of Korido where basement volcanic and metamorphic rocks
were first recorded by Visser and Hermes (1962). Most activity
was centred near Maradori village where gold nuggets up to two
centimetres were frequently being won.
Pieters et al (1979) recommended follow-up investigations of
copper mineralisation seen in metamorphic rocks on the
southwest coast of Supiori island, where a rock chip with streaky
pyrite and pyritic quartz bands with specks of bornite assayed
0.87 per cent copper.
IRC collected 37 stream samples and 21 rock chips during
field visits from 1998 to 2000. There were no anomalous results
in the northern half of the island. A regional gold-copper
anomaly draining metamorphic rocks was located immediately
northwest of Korido in the area where the alluvial gold rush
occurred in 1998 - 1999. Best -150# stream sediment results
were 3.12 ppm Au and 159 ppm Cu in the 10 5 km anomaly,
centred on 14030E and 130S. Gold was pannable in many
creeks in the metamorphic terrain with a highest panned
concentrate result of 235 ppm Au. Volcanic massive sulfide style
mineralisation hosted in sheared spilite was sampled in creeks
near Maradori with the three rock chips collected returning
values of 1.05 g/t Au and 1.88 per cent Cu; 0.59 g/t Au and
2.35 per cent Cu and 2.21 g/t Au and 0.54 per cent Cu. An
intensely altered spilite with native copper in fractures assayed
27.2 g/t Au and 954 ppm Cu. Altered dacite float were seen in
creeks in the southeast of Supiori island.
Despite this encouraging regional anomaly near Korido, IRC
elected not to proceed into full COW status and returned the area
back to the Government in late 2000. A large protected reserve
park in the central part of Supiori Island influenced the decision
not to proceed with further investigations.
WAROPEN AND ZEIWA PROJECTS
A JV was formed between an MRC affiliate (PT Mutiara Iriana
Utama) and Barrick Gold Corporation with the latter sole
funding all expenditures through commencement of commercial
production for an 85 per cent shareholding in the future COW
company. MRC first scouted the Wapoga River basin in 1993
(Keleher, 1993b). Reconnaissance exploration was undertaken by
PT Mutiara Iriana Utama in late 1995, over two blocks covering
1 000 300 hectares on the southeastern shore of Cendrawasih
Bay, along the Wapoga coastline east of Nabire. Exploration in
this initial phase collected BLEG, -60# stream sediment and
panned concentrates from 94 sites. Best geology was seen in
Paleogene volcanic rocks in the southeast of the project area
within the headwaters of the Katehawa River. Gold was panned
in almost all creeks in the headwaters of the Katehawa river
394 Adelaide, SA, 19 - 22 September 2004
where mineralised and altered volcanic float were noted,
including a quartz vein float rock chip assaying 7.5 g/t Au.
The project evolved into two seventh generation COW
applications; the Waropen project covering 403 560 hectares in
three blocks and the Zeiwa project of one block covering 41 780
hectares. The Zeiwa project area was adjacent to the southern
boundary of the Waropen project area and covered parts of the
Katehawa River headwaters.
Exploration completed by the JV included production of a
1:250 000 lithostructural summary map interpreted mainly from
Landsat TM imagery. Five litho-tectonic domains were
interpreted for this western part of the North Coast Basin of
Papua; Neogene marine clastic sediments of the Wapoga
Domain, the Waropen Fault System, the Chaotic Rock Domain,
clastic-dominated assemblages of the Rouffaer Mountain
Domain and the Paleogene Volcanic Arc (Barrick Indonesia (F)
Ltd and Multi Gemilang Abadi, PT, 1997). The main structural
feature is the major E-W oriented Rombak Fault Zone which
defines the boundary between the Rouffaer Mountain and
Chaotic Rock/Waropen Fault System Domain.
Helicopter-supported drainage sampling was undertaken in the
two COW application areas in early 1997. BLEG, -60# stream
sediment and panned concentrate samples were collected from
292 sites covering most of the southern two Waropen blocks and
the Zeiwa block. BLEG samples were analysed for Au, Ag and
Cu, whilst stream sediment samples were assayed for Au, Ag,
Cu, Pb, Zn, Mo, Bi and As.
More than a dozen multi-element stream geochemical
anomalies, scattered throughout the northern clastic-dominated
assemblages were attributed to interbedded conglomerate
sources with notable ultramafic detritus.
Within the Zeiwa COW application block, weak porphyry
copper mineralisation was identified at the Wasewa prospect in
the headwaters of the Katehawa River where potassic altered
diorite porphyries intrude strongly hornfelsed sediments. Copper
mineralisation occurs as veining in overprinting zones of phyllic
alteration within the diorite porphyries. The prospect area,
covering approximately 10 km2 is defined by five adjacent creeks
displaying anomalous drainage geochemistry, centred on
13632.8E and 259.8S. Values up to 340 ppm Cu, 21 ppm Mo
and 0.21 ppm Au were returned from stream sediment sampling.
The maximum BLEG gold value was 145 ppb Au. Gold was
panned in four of the five creeks.
Field observations indicated the mineralisation vectoring
stronger to the south into a COW block held by PT Irja Eastern
Minerals Corporation, a subsidiary of Freeport-McMoran Copper
and Gold. Within this block, surface sampling of a 200 m by
500 m mineralised intrusive at Katehawa East associated with a
magnetic high returned anomalous values ranging between
1.0 - 4.5 per cent Cu and 0.75 - 4.0 g/t Au (Freeport-McMoRan
Copper and Gold, 1997). Porphyry style mineralisation was also
reported in the Haiura area, some 35 km southwest of Wasewa,
where grades from surface samples range between 1.0 - 1.25 per
cent Cu and 0.75 - 2.5 g/t Au and a 1997 drill intersection at
Haiura returned 228 m @ 2.08 g/t Au.
Barrick withdrew from the JV in 1997 and the Waropen COW
application was terminated shortly thereafter. IRC continued to
hold the Zeiwa COW application block, however the inability to
attract new investors at the time led to the eventual termination of
the Zeiwa project in 1999.
IRIANA PROJECT
The Iriana project comprised 1 610 890 hectares divided into
four blocks covering large tracts of north central Papua, evolving
into a sixth generation COW signed on 28 April 1997 by PT
Iriana Mutiara Mining. Exploration commenced in August 1994
with Battle Mountain Gold holding an 80 per cent shareholding
in PT Iriana Mutiara Mining by sole funding all expenditures
through completion of a bankable feasibility study.
PACRIM 2004

Blocks 1 and 2 covered rocks of Pacific Plate geology. Block 1
(Cendrawana) was centred over the Gauttier Range, which owed
its high relief to a WNW-trending, thrust-faulted block of Tertiary,
island arc volcanics and intrusives of basic to intermediate
composition (Iriana Mutiara Mining, 1997). Miocene-Pleistocene
flysch sediments of the Makats Formation and Mamberamo Group
crop out on the flanks of the Gauttier and several smaller ranges.
The Mamberamo Thrust belt dominates structure in the region and
is marked by shale diapirs and chaotically faulted blocks of
sedimentary rocks, with occasional fault wedges of volcanic rocks.
Block 2 (Koada) is underlain by similar poorly consolidated flysch
sedimentary rocks. Separate inliers of unaltered intermediate
volcanics and phyllite were encountered in the Sungai Nadabui
area within the Koada block.
Blocks 3 and 4 cover the southern boundary zone of Pacific
Plate rocks, including upfaulted Pacific Plate mantle material of
the presumed Mesozoic Irian Jaya Ophiolite Belt and Tertiary
island arc volcanics and intrusives and tectonically transitional
Tertiary metamorphic sequences.
Interpretation of acquired Landsat TM imagery and available
air photos assisted in constructing regional drainage base maps
and outlining broad geology (Iriana Mutiara Mining, 1997).
A fixed wing aeromagnetic survey was flown for MRC in
December 1995 over the Cendrawana block and the northern half
of Block 3, encompassing 18 522 line km. The survey line
spacing was 500 m flown N-S with tie lines flown perpendicular
to the survey lines at a spacing of 5000 m.
Helicopter-supported exploration completed by PT Mutiara
Iriana Minerals (an MRC affiliate) in the 1994 - 1997 period
included regional drainage geochemical sampling targeting
porphyry copper-gold, epithermal and sediment-hosted gold and
gold-base metal skarn deposit styles. A total of 2672 sites were
sampled on first pass investigations. At each sample site, stream
float and outcrop were recorded and selected rock specimens
sampled for laboratory analysis for gold, silver, copper, lead,
zinc, and arsenic and occasionally nickel, cobalt, bismuth and
molybdenum. Stream sediment were sieved -80 mesh in the field
and the -200# fraction was further sieved in the laboratory and
analysed for gold, silver, copper, lead, zinc, nickel and arsenic
and occasionally cobalt, bismuth and molybdenum. Panned
concentrates were always assayed for gold and in early stages for
antimony, mercury and arsenic, however analyses for the latter
three elements were discontinued when anomalous values were
interpreted to reflect pyrite content in the heavy mineral
concentrate.
The regional reconnaissance drainage-sampling program
identified nickel laterite mineralisation flanking the Siduarsi
Range and numerous gold anomalies many of which were
attributed to auriferous conglomerate lenses within sediments of
the Mamberamo formation. However gold in stream sediment
anomalies seen draining volcanic, metamorphic and ultramafic
terranes were accorded greater significance and follow-up
investigations identified hard rock gold prospects at Biber and
Kuiri.
The Biber prospect is located on the northern slopes of the
Gauttier range in the centre of the Cendrawana block at
13853.0E and 231.3S and is hosted by intermediate to mafic
Auwewa Volcanics, intruded by gabbroic and dioritic intrusives
(Iriana Mutiara Mining, 1997). The initial anomalous BLEG
sample collected in the mouth of Sungai Jomen, 14 km
downstream of Biber assayed 61 ppb Au. Follow-up stream
sampling returned anomalous values along the length of Sungai
Jomen with peak values of 580 ppb Au in BLEG, 582 ppb Au in
-200# stream sediment and 5569 g Au in HMC, seen in the
headwater reaches in the immediate prospect area. Prospect
investigations included, 60 line km grid establishment, collection
of 1134 ridge/spur and grid soil samples, detailed mapping and
rock chip sampling (685 mostly channel samples) and 52 line km
of ground magnetics. Soil sampling outlined a 1200 m by 600 m
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
A SEARCH FOR GOLD IN INDONESIAN NEW GUINEA
gold anomaly (average 48 ppb Au) partially coinciding with an
area of moderate to strong propylitic alteration developed along a
WNW-trending fault system. Mineralisation is strongest in
fault/breccias containing quartz, pyrite and epidote. Best rock
chip results were from a 0.5 m wide quartz-pyrite shear zone
assaying 15.6 g/t Au and 3.05 g/t Au from a quartz veined,
propylitically altered mafic volcanic rock. However
mineralisation seen at surface is limited to narrow shears and
fractures along a fault parallel to the thrusts that bound the
accreted belt of Auwewa Volcanics and further investigations
such as drilling were not considered.
The Kuiri prospect (13617.4E and 39.4S) was discovered
by PT Mutiara Iriana Minerals with an anomalous BLEG sample
of 86 ppb Au, coupled with a value of 1053 g in the HMC
sample from the regional sample site some 4 km downstream of
the prospect. The prospect is located adjacent to a prominent
NE-trending fault, which displaces the southern margin of the
Irian Jaya Ophiolite Belt at the contact with Derewo
Metamorphic rocks. Grid soils, 21 line km of ground magnetics,
detailed mapping and sampling led to the hand-digging of 14
trenches to better expose mineralisation in the steep forested
terrain. The prospect is partially defined by a 1000 m by 400 m
northeast trending soil gold anomaly defined by the +30 ppb Au
contour. Gold mineralisation is associated with
pyrite-arsenopyrite bearing, quartz veined, potassically and
phyllically altered, porphyritic intermediate intrusive rocks and
volcanics. A 2 - 3 m thick pyritic quartz vein at a
metamorphic-volcanic contact returned 6.5 m @ 3.63 g/t Au and
37.7 g/t Ag from an oblique intersection. Trench 13 exposed a
narrow (<1 m), northwest dipping shear zone with minor splays,
hosted by metavolcanics. Sampling of this clay-silica pyrite shear
zone included high-grade channel results of 1.0 m @ 129 g/t Au
and 0.4 m @ 98.3 g/t Au. Mineralisation is restricted to the
brecciated vein structure with no mineralisation, and limited
alteration extending into the host metavolcanics.
PT Mineserve International (a subsidiary of Freeport
McMoran) was contracted by PT Iriana Mutiara Mining to
complete a review of the project in 1998. An infill
reconnaissance helicopter hoist-sampling program was
undertaken in Blocks 3 and 4 with a total of 142 sites sampled
during the campaign. Follow-up investigations of selected
gold-copper anomalies failed to locate significant indications of
mineralisation and both blocks were relinquished in late 1998.
The only area retained into 1999 was a block of 16 470 hectares
covering nickel-cobalt laterite mineralisation at Siduarsi
(Thirnbeck, 2001). Battle Mountain gave notice withdrawing
from the project in 1999 and application to terminate the project
was lodged in early 2004, following the lack of interest to further
evaluate the Siduarsi nickel-cobalt laterite prospect.
VAN DAALEN PROJECT
The Van Daalen project comprised two blocks, one a sixth
generation COW signed 28 April 1997 by PT Barrick Mutiara
Van Daalen covering 517 700 hectares and an adjoining seventh
generation COW application covering 30 220 hectares to the
immediate northeast of the COW block. Both blocks were
located in Paniai Regency in the headwaters of the Van Daalen
and Rouffaer Rivers within a structurally complex zone of the
Irian Jaya fold and thrust belt. Barrick Gold Corporation held
85 per cent of the COW company shares and funded exploration
under the terms of a JV with MRC.
Four litho-tectonic domains were identified from Landsat
imagery (Barrick Mutiara Van Daalen, 1997). From south to
north, Mesozoic shelf sequences of the Kembelangan Group
from the Australian Plate were separated from schist/phyllite
units of the Derewo Metamorphic Belt by the prominent regional
Derewo Fault Zone. The Rouffaer Fault Zone and Gauttier Offset
separated Derewo Metamorphics with ultramafics and intrusives
395
M R THIRNBECK
of the Irian Jaya Ophiolite Belt (Monnier et al, 2000). Further to
the north, Neogene volcanics and clastics of the North Coast
Basin were seen in lowland areas of the COW project.
A helicopter-borne magnetic-radiometric survey was flown for
MRC at 400 m spacing over both blocks, a total of 16 466 line
km. The most prominent feature in both the magnetics and
radiometrics was the southern boundary of the Ophiolite Belt.
Forty geophysical-lithostructural targets were identified for
ground investigations.
Helicopter supported field investigations had collected broadly
spaced drainage samples over approximately 60 per cent of the
project area when exploration activities were suspended in mid
1997. BLEG, -80# stream sediment, panned concentrate and rock
chip samples were collected at 204 sampling sites that allowed
direct helicopter access. Stream sediments were analysed for
gold, silver, copper, lead, zinc, molybdenum, bismuth and
arsenic. Half the geophysical-lithostructural targets were not
sampled and 16 geochemical anomalies were identified for
further investigations that were not carried out. Barrick Gold
withdrew from the JV in late 1997, with the Barrick share of the
project reverting to IRC. Western Mining Corporation checked
the porphyry copper potential under a 1998 option but elected
not to proceed further. Despite many unchecked drainage
anomalies, the high exploration and land holding costs were
major factors in IRCs decision to terminate the COW project in
early 1999 when competition for limited funds was intense.
MAMBERAMO PROJECT
The Mamberamo project was a sixth generation COW
application covering 843 400 hectares in north central Papua,
centred on the middle to lower reaches of the Mamberamo River.
The exploration effort was governed by a memorandum of
understanding between MRC and Battle Mountain Gold. Battle
Mountain would sole fund expenditures through to completion of
a bankable feasibility study and hold 85 per cent shareholding
once a COW company was formed.
Geology comprised disrupted sedimentary rocks of the
Mamberamo Basin as evidenced by numerous mud volcanoes
(Williams and Amiruddin, 1984). Helicopter supported
reconnaissance sampling of 930 drainage sites was completed by
PT Mutiara Iriana Minerals in 1996. Gold was panned in several
drainages and three gold in stream sediment anomalies were
identified. However the lack of any supporting arsenic or base
metal anomalism supported field observations that gold was shed
from conglomerate lenses within the thick sedimentary sequence
of the Mamberamo Basin. No further investigations took place
and the project was terminated in 1997.
SENTANI PROJECT
MRC formed a JV with Western Mining Corporation to evaluate
nickel laterite deposits in the Sentani area, near the provincial
capital Jayapura. However Western Mining Corporation
withdrew as joint COW applicants prior to establishment of the
COW company and the project was continued with 95 per cent
ownership by MRCs publicly listed vehicle IRC. The Sentani
project, covering 14 640 hectares, was a seventh generation
COW signed 19 February 1998 by PT Iriana Sentani. The main
project focus was evaluation of nickel laterite deposits fringing
Tanahmerah Bay (Thirnbeck, 2001).
Geology is dominated by the Cyclops massif, an ophiolite
sequence with component residual mantle peridotites, cumulate
gabbros, dolerites, normal mid-oceanic ridge basalts and minor
amounts of boninitic lavas (Monnier et al, 1999).
In late 1998 colluvial gold was discovered at Waena within the
Sentani project boundary and an ensuing gold rush developed.
Gossanous boulder subcrop characterised by chalcedonic quartz
vein stockwork, occur in grass fields over a 300 m by 1000 m
396 Adelaide, SA, 19 - 22 September 2004
zone centred along a inferred east-west trending north dipping
thrust fault. Limited rock chip sampling returned a best result of
1 m @ 1.71 g/t Au from a chalcedonic vein stockworked gossan.
Gold mineralisation at Waena (14037.7E, 234.4S) occurs
in hydrothermal breccias with strong silicic, carbonate and pyrite
alteration hosted in serpentinised ultramafic igneous rocks,
possibly an olivine bearing pyroxenite. The alteration mineralogy
is dominated by amorphous silica and chalcedony. Secondary
fluid inclusions from late stage quartz overgrowing opaline silica
cement were measured to homogenize in the range from 138 to
150 Celsius (Iriana Sentani, 1998). Furthermore multiple
episodes of hydrothermal brecciation and fluidisation of wallrock
indicate a high-level epithermal system at Waena.
After failing to identify additional tonnage potential in and
around the known nickel laterite deposits, the Sentani project
was terminated in 2003.
SENGGEH PROJECT
MRC applied for the Senggeh project area in northeastern Papua
initially with Morrison Knudsen Corporation in 1994. Cyprus
Gold Australia replaced Morrison Knudsen in 1995 and as fully
funding 85 per cent partners progressed the project through to a
sixth generation COW signed on 28 April 1997 by PT Cyprus
Amax Iriana.
A fixed wing air magnetic survey for 20 000 line km was
completed for MRC over the entire project area apart from a
17 km wide security strip adjacent to the international border
with Papua New Guinea. North-south flight lines at 800 m
spacing were closed down to 400 m in resultant areas of interest.
Interpretation defined two priority targets with magnetic features
compatible with the Grasberg/Porgera target model.
Drainage base maps were prepared from Landsat TM imagery
interpretation and available airphotos. A helicopter, boat and
vehicle supported reconnaissance program by PT Cyprus Amax
Iriana collected drainage samples from 1071 sites in the second
half of 1997 (Cyprus Amax Iriana, 1998). IRC funded
exploration over the eastern third of the project area under an
option arrangement to earn up to a 46 per cent interest in the
eastern third of the block. Observations confirmed regional
geology as dominated by shallow dipping sediments of the
Mamberamo basin. Faulted inliers of phyllites, schists and
ophiolite rocks were encountered, particularly in the southeastern
quadrant of the project area. Lack of helicopter landing sites
precluded direct inspection of the magnetic anomaly targets.
At each site a BLEG and -150# stream sediment sample was
collected. BLEG samples were analysed for gold, copper and
silver, whilst the stream sediment samples were assayed for gold,
silver, copper, lead, zinc, arsenic, molybdenum, nickel, antimony
and bismuth.
Also, where possible, magnetic concentrates were collected
using a six square inch magnet plate to collect any magnetic
fraction in the stream (Cyprus Amax Iriana, 1998). This is a rapid
technique requiring a low sample weight, relying on precipitated
gold and pathfinder gold being associated with limonitic particles
which include maghaemite. The latter is readily extractable by
magnet directly from the streambed. All up magcon samples were
able to be collected from 72 per cent of sites. The magcon samples
were analysed for gold, silver, copper, lead, zinc, arsenic, iron,
molybdenum, nickel, antimony and bismuth.
Numerous drainage gold anomalies were generated during the
first pass reconnaissance sampling. However Cyprus withdrew
from the JV in October 1998 before any follow-up sampling was
undertaken thus marking the end of regional field investigations.
IRC progressed the project by immediately reducing the COW
area from 1 433 000 hectares to 63 090 hectares, retaining two
blocks covering the best regional gold anomalies at Nambla and
the Southwest Target area.
PACRIM 2004

Anomalous zones dropped prior to any follow-up
investigations during this major relinquishment included gold
stream anomalies at Lereh, Senggi, Ampas, Arso and Wembi, a
nickel laterite near Ubrub and a weakly copper mineralised
diorite intrusive also near Ubrub. The gold anomalous areas
appeared to drain sedimentary rocks containing lenses of
conglomerate that in some cases were found to be gold-bearing.
Whilst gold is detected in panning in these areas (small scale
alluvial mining continues at Senggi) there was no mineralised
float nor was there any supporting anomalous geochemistry in
other elements in the stream sediments.
In general the magcon results mirrored the stream sediment
results apart from detecting some low level antimony anomalies
ranging from 3 - 6 ppm Sb that were not detected in conventional
stream sediment geochemistry. These low-level antimony
anomalies were also not followed up.
The nickel geochemistry in stream sediment samples clearly
revealed a provenance break aligned along a broad
north-northwest trending valley in the sedimentary formations of
the Mamberamo basin. The northern half of the project area
consistently returned background nickel in stream sediment
values ranging from 50 to 120 ppm Ni suggesting sediments
derived from rocks of oceanic affinity. In stark contrast drainages
within sedimentary formations derived from presumed
continental rocks as seen in the southern half of the project area
were always below 50 ppm Ni.
The Southwest Target area was retained in a block covering 31
220 hectares where pronounced airmag features (13954.4E,
325.6S) infer shallow buried intrusives localised at the
intersection of arc normal and arc parallel structures. These
airmagnetic anomalies have target signatures comparable to
Grasberg and Porgera. The airmag anomaly overlapped a
250 sq km area shedding anomalous gold. Thirteen sites recorded
gold in BLEG values greater than 50 ppb Au, with a maximum
gold in BLEG value of 1400 ppb Au and maximum stream
sediment (-150#) value of 2450 ppb Au both from the same site.
These anomalies could only be reached with helicopter support
and were unable to be further investigated due to financial
constraints.
The other retained block at Nambla covered the northern
flanks of the Idenburg Inlier terrane. Reconnaissance
investigations recorded anomalous stream gold geochemistry
with a maximum BLEG value of 125 ppb Au and a gossan
breccia float assaying 79.2 g/t Au and 124 g/t Ag. Limited
follow-up investigations in late 1998 recognised extensions of
the mineralised Mafi River thrust zone seen in the adjoining
Idenburg project area to the south. Brecciated and silicified
ultramafic sampled from the thrust zone (14038.0E, 336.8S),
assayed 0.89 g/t Au, 484 g/t Ag, 2.91 per cent Pb and 11 per cent
Zn. Outcropping massive magnetite skarn (14036.1E, 337.1S)
in Tekai Creek assayed 0.16 g/t Au and contained minor amounts
of the rare earth mineral, allanite.
The Senggeh project was eventually terminated in early 2003
to preserve funds for other exploration projects exhibiting more
promise.
IDENBURG PROJECT
Approval to explore the Idenburg project area was received by
joint applicants MRC and Barrick Gold in early 1995. Exploration
activities commenced in August 1995, operating under a Joint
Venture whereby Barrick held 85 per cent of the COW company
by sole funding all expenditures through commencement of
commercial production. A sixth generation COW was signed by
PT Barrick Mutiara Idenburg on 28 April 1997 covering 1 445 000
hectares of east central Papua. The eastern boundary of the project
abutted the border with Papua New Guinea and extended
westwards over a 260 km distance covering the northern flanks
and foothills of the central range of eastern Papua.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
A SEARCH FOR GOLD IN INDONESIAN NEW GUINEA
Interpretation of satellite imagery identified six major
litho-tectonic domains (Barrick Mutiara Idenburg, 1998). From
south to north these are Mesozoic shelf slope sequences of the
Australian Plate, schist and phyllite of the Derewo
Metamorphics, interleaved metamorphic and melange units and
intrusives of the Irian Jaya Ophiolite Belt, Neogene volcanics
and sediments of the North Coast Basin and Palaeozoic
metamorphic and intrusive rocks of the Idenburg Inlier.
A fixed wing magnetic survey was completed for MRC over
the entire area for 30 595 line km, flown north-south at 400 m
line spacing with tie lines flown east-west at 4000 m line
spacing. One hundred and eleven geophysical-lithostructural
targets were identified for follow-up.
The 1995 - 1996 helicopter-supported reconnaissance
exploration program comprised broadly spaced BLEG, -60#
stream sediment, panned concentrate and rock sampling from
845 sites (Barrick Mutiara Idenburg, 1998). BLEG samples were
assayed for gold and silver. Stream sediment samples were
assayed for gold, silver, copper, lead, zinc, molybdenum,
bismuth, arsenic, antimony, mercury, nickel, cobalt and
chromium. Rock samples were analysed for the same suite of
elements as stream sediments plus potassium. The geochemical
sampling generated 116 single and multi-element anomalies to
follow-up with over half located in the Idenburg Inlier where the
geochemical anomalies were often coincident with
geophysical-lithostructural targets. More than half the panned
concentrate samples from the Idenburg Inlier contained visible
gold. The JV collected a total of 1443 rock chips in the Idenburg
COW. Eighteen of these recorded values greater than 1 g/t Au,
with 17 of these regional rock chips from the Idenburg Inlier.
Barrick Gold planned to continue funding activities at
Idenburg but withdrew from the project in late 1997 as part of a
countrywide pullout. Barricks share of the project reverted to
IRC and the COW company was renamed PT Iriana Mutiara
Idenburg. By mid 1998 vast tracts of the Idenburg project area
had been relinquished, eventually retaining only 108 600
hectares in a single block covering the prospective Idenburg
Inlier terrane. Fortunately an all weather road provides access to
the reduced project area, 120 km directly south of the coastal
provincial capital Jayapura.
Funding constraints ensured only piecemeal follow-up
investigations during the 1998 - 2003 period when it was
virtually impossible for a junior exploration company listed on a
Canadian Stock Exchange to attract funds to develop a grassroots
gold exploration play in Indonesia. Nevertheless prospecting of
near road areas of the Idenburg Inlier in 1998, identified
numerous gold mineralised outcrops of mesothermal origin, the
most notable where gold bearing gossan boulders marked out a
5 km segment of the trace of a northwest trending thrust structure
characterised by dismembered ophiolite slices in the Mafi River
valley. Initial rock chipping of 16 scorodite and fuchsite bearing
gossans in a 500 m by 150 m segment in the hanging wall of the
thrust zone (14039.0E, 338.4S) returned gold assay results
ranging from 5.23 g/t Au to 33.4 g/t Au, with a best channel chip
of 7 m averaging 26.7 g/t Au. Extensions to the Mafi occurrence
were suspected 500 m to the southeast where follow-up
prospecting of a 737 g/t Au gossan float in a small creek was
sourced to a narrow 10 cm pyrite-quartz-chalcopyrite vein that
assayed 1018 g/t Au. Gold in this high-grade vein sample occurs
in irregular 20 m long grains infilling fractures between pyrite
and quartz grains that occur later than the chalcopyrite.
Other areas of interest identified in 1998 within 20 km of Mafi
included Bermol where mylonitised quartz sulfide vein material
assayed 13.9 g/t Au; a 20 cm mesothermal quartz sulfide vein at
Tekai assaying 81 g/t Au; a 50 cm boulder vein at Afley assaying
28.6 g/t Au and at Nova where fingernail sized gold nuggets
were readily pannable.
397
M R THIRNBECK
More detailed investigations from late 2000 to early 2002
funded by project optionholders advanced Mafi and Bermol to
prospect status. Diamond drilling in the second half of 2000 at
Mafi intersected mesothermal quartz sulfide vein gold
mineralisation hosted in magnesite-silica altered serpentinite,
with a best interval of 12.6 m at 8.0 g/t Au, 108 g/t Ag and
2.69 per cent Pb from 6.0 to 18.6 m depth in hole 14. The
distinctive magnesite-silica altered host rock has been mapped
along the Mafi thrust structure in excess of 10 km. The drill
intersected gold mineralisation dips shallowly southwest at
approximately 15 within the hanging wall of the thrust structure
and remains open downdip under alluvium cover. Regional
footwall rocks at Mafi are almandine bearing amphibolite
metamorphic facies, with the precursor rocks most likely being
an intermediate igneous rock, possibly a tuff.
Meanwhile at Bermol, some 10 km south of Mafi, host rocks to
mineralisation include unaltered regional metamorphic rocks of
upper greenschist facies. Detailed investigations included 14 line
km of grid soil sampling and mapping and channel rock chip
sampling of mineralised lode exposures. Structure is dominated by
a north trending, shallow west dipping thrust that is interpreted to
be southern extensions of the thrust structure seen to the north at
Mafi. Prospect geology includes fine-grained weakly foliated
diorite, feldspar porphyry, mafic dykes and schists. Best channel
chip results were 14 m at 6.91 g/t Au and 4 m at 19.4 g/t Au from
vertical samples 150 m apart where the gently dipping
quartz-sulfide lode vein outcropped through a hilltop. The
quartz-sulfide veins at Bermol contain early pyrite followed by
arsenopyrite, with chalcopyrite and later electrum deposited in
fractures. Preliminary bottle roll tests indicate that the gold can be
cyanide-leached at a recovery rate of about 80 per cent for the
fresh unoxidised samples with estimated 15 per cent sulfides,
mostly arsenopyrite and pyrite with minor chalcopyrite. Drilling is
the next planned phase of exploration at Bermol.
Additional hunting-party prospecting in 2002 and 2003
identified mineralised gold structures hosted in diorite at Kali Sua
Sinta where the discovery channel sample across a mineralised
structure with mesothermal quartz veining assayed 3 m at 65.5 g/t
Au. A grab rock chip of fresh quartz pyrite vein within this
mineralised exposure assayed 348 g/t Au. Mineralised quartz
sulfide scree material are seen scattered on surrounding slopes and
as float in the main river both upstream and downstream adjacent
to the discovery exposure suggesting additional veins are present
in the Kali Sua Sinta district. More detailed surface investigations
by PT Iriana Mutiara Idenburg are continuing through 2004 to
define drill targets in this area some 11 km west of Bermol.
CONCLUSIONS
The discovery of the world-class Grasberg porphyry copper-gold
deposit in 1988 triggered an upsurge of mineral exploration
activity throughout Papua province that peaked in 1997. Between
1994 and 1996, MRC attracted some major international mining
companies to Papua and formed Joint Ventures to explore large
tracts of under-explored country to the north of the central range.
In general, first pass regional reconnaissance exploration was
completed over most project areas. However planned follow up
investigations were thwarted by the withdrawal of major funding
partners in the 1997 - 1999 period in response to the perceived
deteriorating investment climate attributed to fallout from the
Busang scandal of 1997, coupled with economic and political
turmoil that plagued Indonesia into 1998.
Following the withdrawal of all the initial JV partners by 1999,
MRC/IRC continued exploration in selected projects as funding
permitted. Gold mineralisation styles identified include VMS on
Supiori Island, epithermal breccias at Waena, skarn in Senggeh
project, porphyry-related at Wasewa, vein breccia shears at Kuiri
and numerous mesothermal lode occurrences within the Idenburg
project.
398 Adelaide, SA, 19 - 22 September 2004
After the buoyant optimism of the mid 1990s, this ambitious
gold search of frontier country in Indonesian New Guinea has a
decade later failed to meet initial expectations. Despite immense
setbacks and hurdles, particularly in attracting further funding
following exposure of the Busang fraud in 1997 and the
1997 - 1999 mass withdrawal of JV partners, the venture has
identified and retained a promising gold exploration project at
Idenburg. Further exploration, in particular diamond drilling, is
required at Idenburg to ascertain the significance and
development potential of the widespread lode type gold
mineralisation of mesothermal origin.
REFERENCES
Barrick Indonesia (F) Ltd and Multi Gemilang Abadi, PT, 1997. Report
on SIPP activities for the period June 1996 to May 1997 for Waropen
seventh generation COW applications, Irian Jaya, unpublished report
to the Indonesian Department of Energy and Mineral Resources.
Barrick Mutiara Idenburg, PT, 1998. Report on SIPP activities for the
period May 1995 to May 1997 for Idenburg COW Irian Jaya,
unpublished report to the Indonesian Department of Energy and
Mineral Resources.
Barrick Mutiara Ransiki, PT, 1997. Contract of Work Termination
Report, unpublished report to the Indonesian Department of Energy
and Mineral Resources.
Barrick Mutiara Van Daalen, PT, 1997. Report on SIPP activities for the
period May 1995 to May 1997 for Van Daalen COW and Van Daalen
seventh generation COW application Irian Jaya, unpublished report
to the Indonesian Department of Energy and Mineral Resources.
Crowhurst, P V, Maas, R, Hill, K C, Foster, D A and Fanning, C M, 2004.
Isotopic constraints on crustal architecture and Permo-Triassic
tectonics in New Guinea: possible links with eastern Australia,
Australian Journal of Earth Sciences, 51(1):107-124.
Cyprus Amax Iriana, PT, 1998. Senggeh COW partial relinquishment
report, unpublished report to the Indonesian Department of Energy
and Mineral Resources.
Dow, D B, Robinson, G P, Hartono, U and Ratman, N, 1988. Geology of
Irian Jaya, Preliminary Geological Report, Indonesia-Australia
Geological Mapping Project. Geological Research and Development
Centre, Indonesia and Bureau of Mineral Resources, Australia.
Freeport-McMoRan Copper and Gold Inc, 1997. Freeport-McMoRan
Copper and Gold updates exploration and expansion activities, News
Release to New York Stock Exchange, 12 May.
Freeport-McMoRan Copper and Gold Inc, 2002. Annual Report to
Shareholders.
Iriana Mutiara Mining, PT, 1997. SIPP exploration report for period
2 August 1994 to 28 April 1997, unpublished report to the Indonesian
Department of Energy and Mineral Resources.
Iriana Sentani, PT, 1998. Sentani COW fourth quarterly report, 1998,
unpublished report to the Indonesian Department of Energy and
Mineral Resources.
Keleher, C D, 1993a. Report on findings from precious and base metals
exploration work conducted in the Kwatisore and Danau Jamur SKIP
prospect area, unpublished PT, Mutiara Iriana Minerals report,
August.
Keleher, C D, 1993b. Report on findings from precious and base metals
exploration work conducted in the Wapoga and Koada SKIP prospect
area, unpublished PT, Mutiara Iriana Minerals report, August.
Monnier, C, Girardeau, J, Pubellier, M and Permana, H, 2000. The central
ophiolite belt of Irian Jaya (Indonesia): petrological and geochemical
evidence for a back-arc basin origin, Earth and Planetary Sciences,
331:691-699.
Monnier, C, Girardeau, J, Pubellier, M, Polve, M, Permana, H and
Bellon, H, 1999. Petrology and geochemistry of the Cyclops
ophiolites (Irian Jaya, East Indonesia): consequences for the
Cenozoic evolution of the north Australian margin, Mineralogy and
Petrology, 65:1-28.
OConnor, G V, Sunyoto, W and Soebari, L, 1999. The discovery of the
Wabu Ridge gold skarn, Irian Jaya, Indonesia in Proceedings
PACRIM 99 Congress (Ed: G Weber), pp 549-557 (The Australasian
Institute of Mining and Metallurgy: Melbourne).
Pieters, P E, Ryburn, R J and Trail, D S, 1979. Geological reconnaissance
in Irian Jaya, 1976 and 1977, Bureau of Mineral Resources,
Australia, Record 1979/19.
PACRIM 2004

Potter, D R, 1996. What makes the Grasberg anomalous, implications for
future exploration, in Proceedings Porphyry Related Copper and
Gold Deposits of the Asia Pacific Region Conference, pp 10.1-10.13
(Australian Mineral Foundation: Adelaide).
Pubellier, M and Ego, F, 2002. Anatomy of an escape tectonic zone:
Western Irian Jaya (Indonesia), Tectonics, 21(4I):1-15.
Robinson, G P, Ryburn, R J, Harahap, B H, Tobing, S L, Bladon, G M
and Pieters, P E, 1990. Geology of the Kaimana Sheet area, Irian
Jaya, Explanatory note and geological map, Indonesia Sheet 3012
(1:250 000 scale) (Geological Research and Development Centre:
Bandung).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
A SEARCH FOR GOLD IN INDONESIAN NEW GUINEA
Thirnbeck, M R, 2001. The Sentani and Siduarsi nickel-cobalt laterite
deposits, northeast Irian Jaya, Indonesia, in Proceedings PNG
Geology, Exploration and Mining Conference, pp 245-254 (Ed:
G Hancock) (The Australasian Institute of Mining and Metallurgy:
Melbourne).
Visser, W A and Hermes, J J, 1962. Geological results of the exploration
for oil in Netherlands New Guinea, Koninklijk Nederlands
Geologisch Mijnbouwkundig Genootschap Verhandelingen
Geologische Serie 20.
Williams, P R and Amiruddin, 1984. Diapirism and deformation east of
the Mamberamo River, Northern Irian Jaya. Geological Research and
Development Centre, Bandung, Bulletin 10, pp 10-20.
399
Nagarze Gold Porphyry, Tibet PRC
D S Tyrwhitt1

ABSTRACT

Nagarze Gold Porphyry is located about 120 km south west of Lhasa adjacent to the main highway
between Lhasa and Nepal, with Lake Yamdrok Yumco at 4500 m elevation lying immediately to the
east of the project area. Orchid Resources China, a wholly owned subsidiary of Orchid Capital of
Perth, Western Australia, has a co-operative joint venture covering over 200 km2 exploration licences
with the China Tibet Geological Institute (CTIGS) (30 per cent) giving Orchid the right to earn a
70 per cent equity through a five year exploration phase, followed by two years of feasibility costing
respectively $US 7 million for exploration and $US 10 million for feasibility. The area of exclusive
interest covers 2000 km2 south of the major Brahmaputra River. Should the CTIGS elect not to
contribute, Orchid can earn up to 80 per cent interest by sole funding.
Over the past decade the Tibet Bureau of Geology with input from the Qinghai Bureau of Geology
and a small diamond drilling program by the Tibet Copper Company, has outlined a central porphyry
complex at an elevation of 5500 m which comprises quartz feldspar porphyry and dioritic porphyry
intrusions with strong pyritic alteration, and in the contact zone abundant quartz pyrite vein
stockworks. Soil and rock chip geochemistry has shown a strong geochemical anomaly of between
0.5 g/t to over 0.7 g/t gold measuring 700 m NW-SE 400 m NE-SW which is broadly coincident
with the mineralised porphyry. A reconnaissance pit excavated on the northern contact area gave
5.59 g/t Au in mineralised porphyry. The deposit has not yet been surveyed by any geophysical
method nor has its mineralisation been drilled.
Over the coming two months, Orchid Resources in a contract with the China Tibet Institute of
Geology will be conducting a 200 200 m pitting and trenching program over the porphyry and will
be trenching and pitting a number of satellite targets where narrow dykes of porphyry intrude Triassic
age argillite and interbedded sandstone.
Several of these dykes contain highly anomalous gold values ranging from 5 g/t to a maximum of
12 g/t Au over one metre sample intervals. One of these targets is being tested 40 m below surface by
a shallow adit, driven along the trend of the dyke.
These intrusive porphyry rocks have not yet been accurately dated but are believed to be of
Himalayan age between ten and 30 million years.

1. FAusIMM, Consulting Geologist, DS Tyrwhitt and Associates Pty Ltd, PO Box 422, South Yarra Vic 3141.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 401

Mining in Fiji
B Wesson1
ABSTRACT
Mining in Fiji, a journey of discovery
The nation of Fiji is located the South Pacific on the Rim of Fire some
four hours flying time from the eastern sea board of Australia. It is a tourist
destination situated on the main route from Australia to the Americas. Fiji
is a developing country that has had a few setbacks but the country has
able, happy people that are very resilient. Most importantly, the country is a
safe destination for tourists and businesspeople as the local people are
friendly, have good ethics, and most speak English. Fiji is a very willing
nation, hungry for knowledge and advancement.
The Fiji geological setting on the Rim of Fire, that runs from New
Zealand across the Pacific through Fiji and on to Papua New Guinea, has
many large deposits of gold and copper. Fiji is an exciting destination for
mineral explorers with discoveries of gold, silver, manganese, bauxite,
iron and copper. Gold was first discovered in 1921 and the first two mines
opened in 1935; one of which was Emperor Gold Mining Company
(EGM) the other being Mt Kasi.
EGM mines gold on the margin of the Tavua caldera, some 85 km
north of the Nadi international airport. EGM has operated continuously
for close to 70 years, producing approximately 6.9 million ounces of gold
and has an underground reserve of 847 600 ounces with a total mineral
resource of 12 335 600 tonnes at 10 g/t (as given in the Emperor Mines
Limited Annual report 2003). The mineralisation in Fiji is very exciting
and it is common to discover spectacular intersections and outcrops,
which are typical for this type of epithermal narrow vein mineralisation,
but be aware that the orebodies are variable and not easy to understand.
Over the last ten years, EGM has continuously mined approximately
120 000 ounces of gold per annum from underground sources at a grade
between 5 - 8 g/t. Mining has mostly taken place on previously known
orebodies following them on strike and down plunge. Over the ten years,
the company has undergone a transformation. Resources have been
1. MAusIMM, 50 Margaret Street, Sydney NSW 2000.
E-mail: bwesson@emperor.com.au
FIG 1 - Fijis location in the South Pacific in relation to Australia and the Americas.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
increased, mine infrastructure rebuilt and improved to reduce
mine-operating costs. Commodity prices have not moved with the cost of
goods and services forcing companies to reinvent the way they do
business, companies must embrace change and innovation to increase
margins to wear the increase in the cost of production as well as volatility
in the gold price and exchange rates. EGM have achieved a more robust
operation through change as detailed in this paper.
LOCATION AND COUNTRY
The republic of Fiji is well positioned in the South Pacific,
located 2800 nautical miles southwest of the Hawaiian islands
and 1750 nautical miles north east of Sydney Australia as
indicated in Figure 1. Fiji is easily accessible; four hours flying
time from Sydney. Outlying small islands of Fiji are close to the
International Date Line and the time zone is two hours ahead of
Sydney.
The nation of Fiji consists of many islands with two larger
main islands, Viti Levu and Vanua Levu. There are over 250
islands that cover a land area of 18 270 sq km supporting a
population of approximately 826 000 that has a population
growth rate of 2.1 per cent per annum.
Figure 2 shows the main island of Viti Levu and Fijis capital
city Suva in the South East of the country. Suva has the largest
population of 167 975 in and around the area. The next largest
city is Lautoka with 43 273 residents (Fiji government
population statistics, 1996). Emperor Gold Mine is situated
85 km by road to the northwest of the town of Nadi. The main
international airport is situated in Nadi. The western division of
Fiji (Singatoka to Raki Raki) hosts the majority of tourist resorts,
sugar cane industry, forestry and Emperor gold mine. The
population is well educated with an adult literacy rate of 87 per
cent and most people speak English, the average life expectancy
is 70 years of age. The nation gained its independence from
Britain in October 1970.
403
B WESSON

Emperor Gold Mine

Mt Kasi
curtailed gold
mine

Capital of Fiji

FIG 2 - Map indicating the larger islands and the location of major towns and rivers.

Major setbacks in the economy were the 1987 coup when
Sitiveni Rabuka seized power and the more recent May 2000
attempted coup by George Speight. Since 2000 the countrys risk
credit rating has started to recover. In Table 1, the third column
provides Moodys comparative ratings as at June 2004. The risk
premiums are indicative numbers considering world economies
at the time.
The Fiji country credit rating remains lower than expected as
the country has recovered from the coup. However, the Labor
opposition party is still in dispute with the government over their
rights under the constitution, which legally casts some
uncertainty over the government and is the reason for the lower
investment grade risk.
Fiji is a great place to do business, having a good rule of law,
based on the British system, no language barriers and a
reasonable availability of competitively priced workers. The Fiji
Trade and Investment Bureau offer very attractive financial
incentives to entrepreneurs who wish to setup in Fiji and engage
in the export of commodities. The four largest net export-earning
sectors as a percentage of total net earnings are tourism at 55 per
cent, garments at 13 per cent, sugar at three per cent and gold at
three per cent. The export of gold attracts revenue of
approximately $A 61 million per annum from the Emperor Gold
mine in Vatukoula.
MINING IN FIJI
History
Emperor Mines Limited was founded 19 July 1935 in Melbourne
by the following Directors: W H Smith, Chairman; T R Victor;
Sir M Hedstrom; P F Cody; J Wren and E G Theodore,
Managing Director, after taking control of the shares in Emperor
Gold Mining Company Limited Fiji EGM from TSP
investments Singapore. At the time, EGM held 100 acres of
mining leases and 708 acres of prospecting leases located 10 km
from Tavua. The existence of Gold in Fiji was first reported in
1868 and payable quantities of gold discovered in 1921. The
EGM discovery was made by Bill Borthwick who found payable
quantities of gold in the Lololevu creek in 1932 and production
commenced in 1935 at Vatukoula. Since this time, 6.9 million
ounces of gold have been processed and there remains a stated
JORC categorised underground reserve of 847 600 ounces with a
total mineral resource of 12 335 600 tonnes at 10 g/t (as given in
the Emperor Mines Limited Annual Report, 2003).

TABLE 1
Comparative investment ratings for various countries.

Ranking for investment Country Moodys credit rating Risk premium percentage Credit grade
1 Australia Aaa 0 Top of investment grade
2 Malaysia Baa1 2.03 % Middle of investment grade
3 Thailand Baa1 2.25 % Middle to low investment grade
South Africa Baa2 4.03 % Lower investment grade
4 Philippines Ba2 7% Top of speculative grade
5 Fiji Ba2 6% Top of speculative grade
6 Papua New Guinea B1 9% Middle Speculative
7 Indonesia B2 11.25 % Middle Speculative

404 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


Geology and description of operation
Regional tectonic setting
Prior to 10 Ma, Fiji formed part of an elongated Vitiaz Arc which
included the Vanuatu, Fiji and Lau segments as a concave
southwesterly arc with an east-bordering trench and westwards
dipping Benioff zone (Corbett and Nethery, 2001). At about
10 Ma (the age is uncertain) the Vanuatu segment became
detached from the Fiji segment and commenced a sinistral
clockwise rotation, opening the North Fiji Basin as an extensional
back arc basin. The sinistral Fiji Transform developed between the
two arc segments as they diverged. The Fiji segment commenced a
90 anti-clockwise rotation with respect to the Lau segment, as
can be seen in Figure 3.
At around 5 Ma the Lau and Fiji segments detached from the
frontal arc and commenced an anticlockwise rotation around a
pole situated in north New Zealand, and the Lau Basin, an
extensional back-arc basin, developed westwards from the newly
formed Tonga Arc. Thus, in the period of development of the
Tavua volcano and mineral deposition from 5.5 Ma to 3.7 Ma,
the optimum sinistral slip orientation was NE, dextral slip was
oriented ESE, and extension structures oriented NNW. The
system is strongly influenced by extension in the North Fiji
Basin to the west, and the Lau Basin to the east.
There are many large mineralised zones and mines on the
Pacific Ring of Fire as given in Figure 3, such as Ok Tedi,
Porgera, Lihir and Emperor Gold Mine all on the Ring of Fire.
Fiji mining tenements prospecting in Fiji
Fiji is a highly prospective area hosting a wide range of minerals
that have been identified and in many cases mined on a small
scale. Presently there are 38 active prospecting leases and four
mining leases in Fiji mostly exploring for gold and copper. Of
the four mining leases, EGM owns three and Mt Kasi one.
FIG 3 - Pacific Ring of Fire indicating the major mineral deposits along the seduction zone (Source: The Fiji Airborne
Geophysical Survey Project, Fiji Mineral Resource Department and the Australian Geological Survey Organisation).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
MINING IN FIJI
Mines have come into operation over the years but apart from
EGM, most have had short mine lives, closing in most cases for
economic reasons. Mt Kasi was one of the more recent mines
that was opened by Pacific Island Gold and produced just over
half a million ounces of gold over three years before insolvency.
The underlying reason for the majority of these companies
failing has been the lack of capital to identify resources and
install infrastructure to allow the resource to be economically
mined and processed.
This occurs because of the nature of the mineralisation, that
hosts spectacular grades but is highly variable in grade,
continuity, gold recovery, width and dip, making them extremely
difficult to drill out into a JORC reserve category for
bank-ability. The only method of fully appreciating the
mineralisation is to diamond drill the deposit on a wide grid
using core orientation and as soon as there is sufficient
information to give the explorer some confidence of the
mineralisation costeaning and decline development is needed to
expose the mineralisation and supply sufficient information for
mine design and feasibility study.
Emperors Tuvatu advanced gold exploration project
The Tuvatu project (SPL 1283 and 1296) is located in the Sabeto
valley some 24 km north west of Nadi international airport.
Mineralisation is found in auriferous veins and flatmake
structures associated with late stage epithermal mineralisation.
Flatmakes are the name used by Emperor to describe shallow
dipping orebodies. This project is the most advanced exploration
project in Fiji as there has been a substantial amount of surface
exploratory drilling, 51 484 m of diamond core, 9265 m of RC
and a further 13 407 m of diamond underground drilling. To fully
understand the orebodies, a decline of 1541 metres was
developed to intersect all the main areas of mineralisation and
create a platform for underground drilling. Underground drilling
was desirable, as exploration had previously been curtailed in for
405
B WESSON
Tuvatu Gold Prospect
Nittetsu Namosi Copper Prospect
FIG 4 - Fiji current tenement holdings and significant areas (Source: Fiji mineral resources department).
FIG 5 - Modelled Tuvatu orebodies showing orientation on surface
landscape (Source: A Vigar and Associates in 1998 modelling for
feasibility study).
five months a year due to the high rainfall and steep terrain.
Other advantages were the production of monzonite from the
decline used to armour the roads. As drilling moved underground
there was no need to establish roads up the steep terrain or
supply water to drill rigs 400 vertical metres above the valley
floor river. Aesthetic and environmental rehabilitation work on
406 Adelaide, SA, 19 - 22 September 2004
Mt Kasi Gold Mine
Emperor Gold Vatukoula
the surface in the flight path of international aircrafts landing at
the Nadi international airport was no longer necessary. A total of
1541 metres of decline, strike and rise development has been
carried out. Included in this development is an 800 m exploration
decline developed to a depth of 240 m below surface as shown in
pink below. The decline traverses the main lodes.
Exploration has delineated some 1.6 million tonnes at a grade
of 8.5 g/t gold and a feasibility study was carried out indicating
that the project is viable but required a larger resource base to
support the initial mine construction cost.
Large base metal prospect
Emosi copper prospect is a large porphyry-type copper prospect
discovered in 1930, situated 30 km NW of Suva. This prospect
hosts low-grade copper gold with approximately 1.1 billion
tonnes at 0.4 per cent Cu and 0.13 g/t Au. This prospect is
presently being explored by Nittetsu.
Land access
The procedures to follow for exploration in Fiji are simple. A
company must commit to a work plan and show that they have
the competence and finance to carry out the plan. The Fiji
Mineral Resources Department MRD would then issue a
Special Prospecting Lease SPL to the prospector on paying the
lease fees. Prospectors must be prepared to negotiate with
landowners and the Native Lands Trust Board to determine the
PACRIM 2004

level of compensation to landowners for any disturbance of their
land. As is the case in many countries it is the most important
task to create harmony and participation with landowners during
the exploration phase, as the landowners and surrounding village
dwellers can assist the explorer. When a mining title is required,
a compensation agreement must be concluded before a mining
licence will be issued. These negotiations can be time-consuming
to build trust and at times can be frustrating but in the majority of
cases explorers gain access.
Emperor Gold Mine in the Tavua caldera
Geology of the Tuvua caldera
It is postulated that there were five eruptions of the Tavua
volcano the first some 43 Ma years ago. The initial eruption was
large but subsequent eruptions were small and non-explosive
with magna flowing freely from the chamber. The empty
chamber then collapsed forming a large 78 km2 depression with a
26 km2 area within the caldera margin. The rocks of the Tavua
volcano were derived from potassium-rich magma of shoshonitic
association with an evolution from absarokite parent magma to
shoshonite, banakite (trachyandesite) and monzonite derivatives.
When the collapse occurred, the forces created tearing of the area
on the margin of the caldera, which resulted in bedding fractures
that radiate down into the margin of the caldera.
The gold deposition occurred when ground water mixed with
the hot alkaline magma, causing hydrothermal fluids to feed
through faults and into the fractures of the collapsed caldera.
Gold deposition occurred when the hydrothermal fluids cooled in
the cracks (selvages) and deposition took place on the walls of
the cracks. Gold mineralisation is located on the outer edges on
the mineralised veins.
Usually bonanza grades in the Pacific Rim type epithermal
systems are found when roscoelite is present. Roscoelite
(K(V,Al,Mg)2 AlSi3O10(OH)2) is found in gangue material of low
temperature early stage epithermal gold, silver and tellurium
deposits as found at Emperor, Porgera Zone VII and Mt Kare.
FIG 6 - Diagram of the mineralised bedding fractures (Corbett and Nethery, 1999).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
MINING IN FIJI
THE METAMORPHOSIS OF INFRASTRUCTURE
AT EMPEROR OVER TEN YEARS
Figure 7 is a view of the Smith Wallace shaft complex that is the
main rock hoisting shaft servicing the R1 and Matanagata
sections through 19 and 15 levels. The shaft is 705 m deep and
hosts the mines main pumping system transferring an average of
9 Ml to surface. Ore is hoisted up the Smith shaft directly into
the crusher to the fine ore bin through two mills and split through
the flotation, refractory ore to roasting and free milling ore to
CIP.
Ore is mined from four areas, namely Matanagata, R1, Main
decline and Philip shaft. Mining takes place on the margin of the
Tavua caldera distributed along the radius.
Smith Shaft is the main shaft with seven compartments and
hosts the main pump stations on ten and 16 levels. The shaft is
705 m deep, equipped with two winders, one for services and the
other for haulage of ore and waste. This shaft services the
Matanagata and R1 mining sections through 19 and 15 levels
respectively. The Matanagata section mines flatmakes that dip
between 20 and 40 degrees into the caldera margin. Matanagata
orebodies have been mined for the past ten years using typical
South African long wall methods employing hand held drilling,
electric scraper winches and timber support. The Matanagata
orebodies have been continuous on strike to the northeast and on
dip are believed to pinch out at the caldera margin below 23 level
approaching 800 m below surface. The Matanagata orebody has
been one of the most profitable production areas since 1992 and
has produced at grades in excess of the resource grade.
The R1 orebodies were found many years ago through churn
hole drilling inside the caldera. Emperor delineated these
resources in the early to mid 1990s and developed into the area.
The orebodies were more complex, with steep and flat structures
that dip in a variety of directions. Grades can be spectacular but
the average mined grades are 6 to 8 g/t due to dilution and the
complexity of mining. Many mining methods are utilised due to
the varying nature of the orebodies. Mine methods are mechanised
cut and fill, manual shrink stoping and long wall methods.
407
B WESSON
Power station upgrade 10.4 MW and refrigeration plant
Handpicking No 1 and No 2 winder upgrades
Photo 1, Emperor Gold Mine Vatukoula in 2000
New Ball mill
Smidth Mill upgrade
30 metre in ground thickener for tailing retreatment
FIG 7 - Emperor Gold Mine, Vatukoula in 2000.
At Philip shaft a 5.5 m diameter shaft was sunk by Western
Mining in 1987 to mine the Prince William Flatmake. Western
Mining operated Philip shaft over a three-year period and
extracted 176 980 ounces of gold from the shaft at an average
grade of 31.1 g/t Au. The Prince William orebody at Philip shaft
outcropped on surface but did not carry any grade, however
Western Mining delineated high-grade gold mineralisation from
approximately 200 m below surface and mined this orebody over
a dip of 400 metres. Areas of the Prince William were not mined
below 16 level (568 m below surface) due to hot water. Emperor,
in the Phase 2 plan, is targeting these areas and the deeper
extremities of the Prince orebodies previously part of the
Vatukoula deeps project.
The main decline centrally located near Smith shaft was
developed from the Emperor open cut to mine the shatter shear,
which was a large stockwork that on average produced a mined
grade of approximately 5 g/t. As bulk mechanised mining could
take place with low access costs through a large decline to
surface, this orebody was the major source of ore from the
decline until 1996. The decline has also exposed old areas that
have high grade in remnant pillars, virgin blocks of ground and
broken ore left behind. The main decline is a nominal 4 4 m
decline that spirals down to 12 level with ore and waste access to
Smith shaft on eight level.
Upgrade of metallurgical plant
Hand picking
Hand picking of ore was practiced in many mines after the turn
of the century, as the production and process capacity was
limited, and labour was considered a very inexpensive
commodity. In 1994 tests were carried out by installing extra
hand picking bays along an inclined conveyor belt-carrying
product with a top size of 100 and bottom size of 50 mm. This
was a good social exercise as it found work for the local woman
that under the Fiji Mines act cannot work on a mine. With special
dispensation, tests were carried out and it was found that one
hand picker could pick a tonne an hour of waste off the belt.
408 Adelaide, SA, 19 - 22 September 2004
Crusher upgrade to 3 stage 1 Mt/annum
Increased leach capacity by 27%
As any waste picked before second stage crushing reduced the
metallurgical costs by $US 11 per tonne and a hand pickers
hourly wage was $US 1.40 Emperor increased mill capacity or
process retention time through hand picking. In the ideal world,
it would be economic to prevent the waste from entering the ore.
In practice, the veins are narrow and stopping heights are on
average 1.2 m with some sections maintaining 0.9 m and finding
difficult to work the stopes efficiently.
The project cost $US 425 000 to replace one of the main
conveyor gantries and build the handpicking station. Combined
with this initiative was the management of risk as the existing
gantries structures were heavily corroded and not
self-supporting. Detailed engineering was complex, as the
replacement and modification needed to be implemented without
loss of production. The maximum delay in processing of ore was
five days, as ore passes would fill up and impact the production
cycle. In the 60 days prior to the shutdown the structural steel
was erected around the existing operating plant with a huge
emphasis on safety. The project was commissioned in early 1995
on time in budget with no accidents or incidents.
The design was for 30 pickers and assumed a picking rate of 15
per cent of the total ore throughput of the plant. On
commissioning, it was found that depending on percentage of
coarse fraction of ore from underground the picking station could
achieve a picked waste rate of 20 per cent at an average discharged
waste grade of 0.2 g/tonne. The project resulted in a large
reduction in plant process costs, increased capacity and leaching
times and created employment opportunities for approximately 60
workers. Presently 49 pickers are employed working various shifts
to coincide with shaft hoisting of ore. The hand pickers remove
over 15 per cent of waste rock from the belt.
Comminution
The grinding circuit initially consisted of two parallel circuits,
one known as Dads army the other the Smidth mill. Dads army
was installed in 1944 and consisted of four 1.8 m diameter mills
driven by 134 kW synchronous motors configured with a rod
PACRIM 2004

mill feeding two grate discharge ball mills and a final product
ball mill. The 750 kW, 3.5 m overflow FL Smidth ball mill was
installed by Western mining in the late 1980s. Dads army
production capacity was low at 18 tonnes per hour and the four
mills were onerous to operate achieving a poor, variable grind,
typically 55 per cent 75 m that influenced flotation efficiency
and gold recovery. After 50 years of operation Dads army were
very unreliable with frequent metal bearing failures and the
shells were severely eroded.
In 1993 mine planning in conjunction with geology
determined that the optimum size of underground mining
operation that supported the resource base was approximately
800 000 tonnes per annum. The existing mills could achieve
55 tonnes per hour, 467 000 tonnes per annum (97 per cent
availability) plus 15 per cent extra for slimes that bypass the mill
circuit in de-gritting.
The discussion to install a new mill
To process 800 000 tonnes per annum the grinding section
required a mill upgrade to replace the old plant capable of
grinding 43 tonnes per hour (97 per cent availability and less
15 per cent slimes). Testwork was carried out to determine the
nature of the ore and after modelling the existing Smidth mill,
the following was evident:
The -25 mm ore feed size to the Smidth mill caused
inefficiency in power consumption, liner wear, scatting. The
single cyclone and variable speed pumps caused large
variations in grind and mass flow, negatively affecting the
flotation efficiency causing the loss of gold to flotation tail.
Sulfide refractory ore in the tail bypasses the roasting process.
Variation in feed size was a function of the circular fine ore
bin with a central filling point from the crushers. The fine ore
remains in the middle of the bin and the course ore roles
away to the outer area of the bin leaving the fines in the
centre. Due to this phenomenon, it was important to schedule
the method of drawing the various chutes to achieve a
reasonable feed distribution to the mills.
The Smidth mill was fitted with wave liners, which are only
suitable for a fine feed product not the -25 mm feed at the
time. The limited charge lift caused poor grinding efficiency
and severe scatting. The mill unit costs were high for power
and labour to rehandle the mill scats.
SAG and AG mill testwork carried out concluded that the ore
was amenable to both. SAG and AG milling was not pursued
as the ore mined comes from many different orebodies with
no blending facilities and the other factor was the cost of
electrical power at $A 0.15 per kWh. Another detractor was
the risk of having one mill and the need to wash and screen
the ore before handpicking discussed above.
The target grind was determined to be P75 75 m, far finer
than the P55 75 m being achieved at the time.
The modelling of the Smidth mill and the examination of
other milling applications indicated that a 3.7 m diameter and
4.9 m long would suit the application with an input power of
ideally 1200 kW but a minimum of 1000 kW.
Motivation for replacement of Dads army mills:
a new mill was 13 kW per tonne more efficient than the old
mills equating to $US 210 000 per annum in power cost;
the old mills required over $US 120 000 of repairs that would
cause large delays;
the overall operating costs per tonne milled for the existing
Dads army was $US 1.82/t, modelling indicated that a new
larger mill would reduce the operating costs per tonne
approximately 51 per cent to $US 0.88/t; and
the new mill allowed the mine to increase production.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
MINING IN FIJI
The mill project
The gold price was low at the time and capital expenditure was
$A 1.4 million for the project. A suitable mill was located in
Sweden at a closed iron ore mine. Morgardshammer manufactured
the mill in the early 1960s and refurbished the mill replacing the
shell just before the iron ore mine closed. The mill was a good low
risk buy; however, the drive train was too small at 860 kW. The
mill was purchased and landed in Fiji on site for $A 400 000 and a
used 1000 kW drive train was located in Australia and the mill
was installed and commissioned in five months.
The mill was sited alongside the roaster away on bedrock
some 100 m from the existing Smidth mill. This was necessary to
curb foundation costs and project risk as the area around the
Smidth mill was all backfill and unsuitable without pilling for a
ball mill. This has given the plant a strange look with two mills
some distance from one another. If the mills were located along
side one another the existing operator would have managed both
mills. By locating the new ball mill away from the existing mill
and along side the roaster there remained no need to increase
staff as the roaster operators could manage the mill.
The mill is an overflow mill running at 73 per cent critical
speed initially fed with 80 mm steel balls and operated in a
closed circuit with a Cavex 250 mm cyclone cluster. The mill
circuit pump is a variable speed 75 kW, 150/200 EAH Warman
pump. The mill was fitted with rubber liners which is very
common in Australia. With the coarse feed of -25 mm the mill
liners were a major problem with excessive lifter wear and lifter
damage. The mill was subsequently relined with polyurethane
liners that have steel caps and leading faces to tolerate the coarse
feed. These liners resolved the problem but the cost was close to
steel liners.
The mill was commissioned in 1996 reducing the cost of
grinding by $A 0.8 per tonne while increasing mill capacity by
25 per cent.
Crusher upgrade
The Emperor crusher plant was built in the 1930s and upgraded
in the 1940s. The underground sulfide ore presently and
previously being processed is friable. In the past open cut mining
of oxides and upper levels produced sticky ore that was difficult
to slurry. Emperor handles ore on average four times before
being crushed. The handling by scraper, loader, truck and then
through the shaft passes breaks the ore and creates 90 per cent
fraction passing -70 mm leaving very little work for the surface
jaw crusher but a major problem when mixed with open cut ore
that is sticky.
The old two stage crusher produced an P80 20 mm with a
top size of 25 mm and when open cut ore was crushed screens
were removed allowing -90 mm to enter the milling circuit. The
incoming product was washed over a grizzly and again in a blade
mill and then delivered to a wash screen. This washing was
required as the ore was very fine and when mixed with open cut
ore it become very sticky and difficult to cone crush. The ore
also required washing for the hand picking station. This circuit
operated for 50 years with little change.
The crusher circuit was very labour intensive with eight people
per shift; a full time fitter and supervisor were required to operate
the crusher at low production rates in the order of 400 000 tonnes
per annum. In 1993, the circuit was reviewed, as it was onerous
and expensive to operate. For all this pain of operating the onerous
plant it cost on average $US 0.75 per tonne for the blade mill and
a total cost per tonne crushed of $US 2.05 per tonne.
As the mine was expanding, the crusher required an upgrade
and given the condition and design, it was prudent to replace the
entire two stage crushing plant with a three-stage higher capacity
plant.
409
B WESSON
FIG 8 - Secondary and tertiary crushes.
Requirements for the new plant were:
Reliability in remote areas, technology must be robust and
simple.
Increased crusher capacity from 500 000 to 1 000 000 tonnes
per annum at an availability of 72 per cent.
Reduced metallurgical operating costs and improved down
stream operating efficiencies. The production of a fine
consistent product to mills reduces the cost of milling power
and the cost of liners and balls. Further flotation efficiencies
improve as the mills have a large capacity due to the finer
feed. A consistent ore feed size and subsequent mill
discharge to the flotation improves recovery through flotation
circuits.
The new crusher was built and commissioned January 1997 at a
cost of $US 750 000 million for design, equipment, power
reticulation and control, and installation. The project was very
onerous as it had to be constructed around the existing operating
plant using the same bins and buildings. The project took six
months with the critical path being the importation of a jaw
crusher from France. During a ten day shutdown the old crusher
was stripped out and the new crusher installed and commissioned.
The project was marginally late by 22.5 hours as extra concrete
had to be removed to fit in steelwork and the budget over-run was
4.1 per cent. The new crusher achieved a P80 6 mm and the
operating cost was $US 1.35 per tonne crushed.
Extraneous water
Extraneous hot water was first accounted in the 1950s when a
drive out on 14 level Cayzer mined into a churn hole and nearly
lost the mine due to the ingress of water. The drive on 14 level
encountered 40 to 50C water and had many attempts to seal the
water with little success. Once Emperor delineated resources in
R1 inside the caldera, it was necessary to develop into the R1
orebodies. A detailed hydrological analysis was undertaken by
SRK London and based on this report a decision was made to
dewater the area.
410 Adelaide, SA, 19 - 22 September 2004
FIG 9 - Cooling of extraneous water for reuse at Philip shaft.
A LM55 diamond drill was modified to drill into high-pressure
hot water and fitted with blow out protectors. Holes were drilled
into the water bearing structures and an old churn hole called
F19 was redrilled and an oilfield screwed casing grouted into the
hole from surface down 494 metres to 14 level. A 710 kW Sulzer
8 stage pump was installed on 14 level to pump 68 L/sec through
the casing to surface. The water is sprayed on surface to reduce
the temperature of the water before use in process or discharge.
The dewatering was very successful and a second pump was
installed on 16 level, Cayzer shaft some 574 metres below
surface to pump extraneous matter up a new 350 mm shaft rising
main. Since dewatering the drill hole water pressures have fallen
to an equivalent elevation below 16 level. Pumps have now been
installed in the lower levels of R1 declined to pump to the main
pump chamber on 16 level.
Western Mining Corporation encountered hot water at Philip
shaft, in the shaft, during sinking and on 13 level (430 m below
surface), 1601 (538 m below surface) and 1701 (578 m below
surface). The development of these main drives was abandoned
and the lower levels of Philip shaft partially mined areas that
were not excessively impacted by hot water. In 1995, Emperor
drilled dewatering holes from 1601 and 1701 and installed a
pump station on 16 level Philip shaft to pump the hot water to
surface. The concept is to dewater and reduce the zone pressures
in the vicinity of the haulages to allow cementation to seal the
areas before development takes place.
Emperor is one of the deepest mines in the world with hot 40
to 50C extraneous water in close proximity and at times on the
orebodies. Emperor has innovated and developed cutting-edge
technology to manage the water.
Phase 2 expansion 2002
The Phase 2 mine expansion plan focused on optimising the
company by fully exploiting the large resource base to extend the
mine production plan from a five to a ten years at a higher yield.
The Phase 2 expansion is projected to cost approximately $A 75
million, financed through debt and equity. The focus of the plan
is as follows:
1. Flexibility through increased:
near mine brown field exploration, and
capital development.
2. Improved operational efficiency through:
replacement of aging heavy vehicles with more efficient
and productive machines;
PACRIM 2004

installation of ventilation systems to adequately ventilate
areas; and
reducing tramming distances by deepening shafts and
installing new access ways between shafts.
3. Increase production by 40 per cent.
Flexibility
Exploration
Emperor Gold Mine has operated for 68 years and produced over
6.9 million ounces of gold. Over the years the resources have
increased as new orebodies have been discovered and many of
the larger orebodies previously and presently being mined
remain open on strike or dip. Historically 30 per cent of ore
mined was not specified in the resources (NIR, this occurs as in
many cases resources are discovered while mining is taking place
and steep and other structures are found that were not identified
in the original drilling. On average over the years, gold
estimation in terms of block factors has reconciled well with
actual mining gold content within 5 per cent.
In April 2001, Emperor Gold contracted Dr G Corbett and
J Nethery Corbett&Nethery specialists in epithermal systems to
use modern techniques and review the large geological database
that covered 70 years of mining. This study substantiated that the
area was a world-class gold system that was highly prospective
and drew parallels with Lihir, deeper parts of Porgera and Kelian.
Emperor has a close correlation with other Pacific Rim deposits
that are generally quartz sulfide gold copper style gold
mineralisation with bonanza grades associated with roscoelite.
The study highlighted areas such as Matanagata North East,
Brewster West, Matanagata West, Prince William SE, Prince,
Kava, R1 West and R1 deeps indicated in the Smith section in
Figure 10.
Recognising the large 3.6 million ounce resource base at the
time and given the comments by Corbett and Nethery it was very
evident that the mine had the potential to produce at a higher rate
FIG 10 - Indicative orebody and mine area locations.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
MINING IN FIJI
than the average 120 000 ounce per annum. After intense
planning and consideration, the geologists and mine planners
prepared a plan based on known resources to build up production
to 180 000 ounce over a 30-month timeframe. To make the plan
robust there was a focus on exploratory drilling from surface and
underground to upgrade ore resource in category and stepping
out to locate lateral and down dip extensions of existing
orebodies as indicated outside the mine sections as shown in
cream coloured zones in Figure 10.
Capital development
Capital development is the lifeblood of underground mines, as is
moving back overburden for open cut mines. Emperor Gold
Mining Company Limited defines underground capital
development as development of infrastructure to support mining
and includes all declines and main drives to the point of the first
rise on ore. After this point, the development is classed as
normal development and is costed as part of mining costs in the
area.
The most important functions for a mine are finding and
determining resources, defining these resources and accessing
the area by advanced development before mining. This
development is a continuous process and for higher levels of
production must increase to support the level of extraction and
create flexibility to manage any variability in complex
epithermal narrow vein orebodies.
Advanced capital development in R1 and Philip shaft is more
important than other sections as the hot water shears need to be
dewatered from the level below the mining horizon. As described
above, boreholes are drilled across shears and the water extracted
and pumped to surface. As the water filters from depth through
fractures migrating to the extraction bore holes, the zone pressures
fall quickly. This fall is a function of the small fractures feeding
the area restricting the flow that causes a fall in pressure from the
static high pressure to a dynamic low pressure. The lower pressure
makes it far easier, quicker and low risk to seal the water shears to
allow development to traverse these shears.
411
B WESSON
Efficiency
Ventilation
The Emperor Gold mine is close to the sea and uses a sea level
surface barometric pressure of 101.4 Kpaabs this is important
when considering isometric charts and relative humidity. Fiji
receives over two metres of rainfall per annum from October
through to April and temperatures reach a maximum of 27C wet
bulb at 90 per cent humidity during these months. The
ventilation in the mine must cater for the hot surface
temperatures, humidity, high geothermal gradient of three
degrees per 100 m, heat from hot water leakage into workings
and the heat load from heavy vehicles and broken rock.
In 1992, Emperor had two main surface ventilation shafts,
Cayzer shaft with a collar diameter of 3.4 m fitted with a
subbrace air duct that diverted upcast air away from the brace
into a Howden/Stork axial flow 134 kW fan. The airflow was
87 m3/sec ventilating Cayzer and Smith shaft areas. This area
was suffering from excessive temperatures with poor mining
efficiencies. Borthwick shaft is a 3.4 m circular service shaft
converted into an upcast shaft to service the Philip and decline
sections. The shaft had a 250 kW, 2.2 m axial flow fan fitted to a
bend above the collar, extracting 100 m3/sec. The total mine
extraction was 187 m3/sec.
In 1994, a second axial flow fan was installed on the
Borthwick shaft increasing the extraction to 180 m3/sec from the
decline and Philip shafts. This change increased the total volume
of air extracted to 267 m3/sec for the mine. The R1 section was
developed from Smith shaft and the Cayzer shaft could not
provide sufficient ventilation for the Smith shaft, Matanagata
workings and the new R1. To improve conditions a new
ventilation shaft (Vision shaft) was raisebored in two sections,
one 3.8 m diameter from 16 level Smith shaft to eight level main
Decline and then 3.5 m diameter to surface.
The shaft was fitted with centrifugal fans, as axial flow fans
previously used are low-pressure fans that are not suited to deep
narrow vein high resistance operations as they are susceptible to
surge. In surge conditions, the blades distort and fail. The ideal
fan for a high resistant deep mine is a centrifugal fan that
tolerates high suction pressures, can easily be repaired as the
impeller is made of carbon steel and the motors are out of the
main air stream.
The vision shaft was fitted with two 315 kW, ABB centrifugal
fans extracting 210 m3/sec of air at a pressure of 2.2 Kpa. The
Cayzer shaft was refurbished and down cast to become a service
shaft for the R1 section. The total airflow for the mine increased
from 267 m3/sec to 390 m3/sec.
Phase 2 strategy
The Phase 2 plan is forecasting to produce 800 000 tonnes of ore
per annum, increasing capital development to peak at over 6000 m
per annum and increase the underground heavy machinery by
23 units. To achieve 100 per cent manpower productivity and work
rate the ventilation design for Phase 2 was upgraded from the
previous cooling power of 180 w/m2 to a design of 250 w/m2.
The main drivers of heat in the mine are:
Surface summer temperature reaches 27C wet bulb
temperature at 90 per cent humidity. Surface rock
temperatures are 25C.
Hot 45C water entering workings and the ventilation stream
through drips or leaking water running free in the main
ventilation air stream. This water is controlled but at times
heat enters the ventilation stream.
High geothermal gradient with heat radiating from the
exposed excavation surfaces and from broken rock
transported through the haulages and mine shafts.
Heat from heavy machinery.
412 Adelaide, SA, 19 - 22 September 2004
To a lesser extent:
heat from auto compression of the air flowing down 600 m
mine;
supply of compressed air; and
heat from electrical equipment.
Refrigeration
Prior to 1997 small spot refrigeration plants with a capacity of
200 kWth were used to ventilate very hot areas. Typical
application was to cool drill chambers while drilling hot water
extraction wells. The small machines were not very effective and
practical to cool the production areas. In 1997 the first large 2500
kWth chiller installed to cool the air for Matanagata East area.
The plant performed well for 18 months and was shut down in
the Fiji dry season (cool season) as conditions were acceptable
underground. On re-commissioning, the machine had major
problems that had developed from the negative gas pressure
machine sucking water through cracked tubes and suffering acid
attack. Poor water quality created poor heat transfer and high
condenser pressures translated into compressor surging causing
fatigue of the high efficiency copper tubes, cracking and
developing water leaks. The leaks were not detected during the
months while standing. Water entered the R123 gas through fine
cracks in the tubes and formed hydrofluoric acid that destroyed
the internal components. When the machine was again required,
it was in a very poor state and repairs were not satisfactory due to
the advanced corrosion throughout the machine. The machine
ran inefficiently with continual failures and after evaluation, it
was more economic to replace the machine with a new high
pressure R134a, 2900 kWth York machine as shown in Figure 11.
The new machine was commissioned in the latter part of 2003
and made a significant impact on the working conditions in the
Matanagata East and North East workings dropping stope
temperatures by two degrees and indications are that there has
been a ten per cent improvement in productivity in the area.
FIG 11 - New York 2900kW R134a refrigeration.
In line with the Phase 2 strategy, a further two York YT
centrifugal 2500 kWth machines have been purchased for bulk air
cooling applications on surface at Philip shaft and Cayzer shaft.
These plants are required to reduce the system approach
temperatures and humidity. The high relative humidity in Fiji
makes the air more conductive and heat is picked up faster as the
air moves through the approximate 2 km of declines and drives
over heavy vehicles trucking ore from the workings. To achieve
the design parameter of 250 w/m2, refrigeration is required to
reduce the temperature of the air before it enters the shafts.
A study was carried out comparing the capital and running
efficiencies of bulk underground and surface bulk cooler
operations using chilled water from surface plants. The extra
PACRIM 2004
 MINING IN FIJI

capital cost associated with the underground installations and
operating costs associated with pumping of chilled water,
condensate and the losses in the cooling coils proved that for
shallow application above 1000 m with high unit costs of
electricity a surface mounted plant and bulk air-cooler is the
most appropriate and economic solution. The design for Cayzer
and Philip shafts is for surface bulk air cooling with some
cooling of the service water feeding the workings.
As the operating conditions in Fiji are onerous, the two new
plants have been designed to be robust using R134a refrigerant
(none ozone depleting under Montreal protocol) making the
plants positive pressure and more tolerant to fouled tubes and
subsequent higher condensing pressures. The plant coefficients
of performance are above five. (Coefficient of performance
COP is the thermal heat exchanged over the electrical energy
consumed). Copper nickel tubes and delta clad tube plates have
been installed to tolerate poor water quality, and lack of, or rough
maintenance over the machines life cycle. The copronickle tubes
are thicker and smooth bored to ensure that scale can easily be
brushed out, or in dire circumstances, the machine can be acid
cleaned without reducing the plants life.
Refrigerated air is a very expensive commodity for any mine
especially in Fiji. Power costs at the time were $A 0.15 per kWh.
To operate a single refrigeration plant the electrical power cost is
approximately $A 980 000 per annum before maintenance.
some existing machines refurbished or replaced. The heavy
vehicle fleet is large for the mine size (approximately 35
production machines) as the orebodies are widely spread across
the mine laterally and vertically. The majority of machines are
fitted with very low NOx Detroit 471 and series 40 and 50
engines; however, modelling indicated that the air volume was
insufficient for the proposed machinery or future workings.
Two new 3.5 m ventilations shafts were raisebored, one in the
R1 area called Sautu that would be an intake air shaft and the
other on the boarder between Matanagata east and northeast
called Cagi Vou, an upcast shaft. Two centrifugal 350 kW fans
were installed on Cagi Vou to extract 200 m3/sec from
Matanagata area and Smith shaft, freeing up the existing Vision
shaft to draw air from Cagi Vou through the R1 workings. Both
shafts were completed in the 2003/2004 financial year within
budget constraints. This brings the total installed mine
ventilation capacity to 590 m3/sec.

Ventilation shafts
As can be seen in Figure 12 and Figure 10 the mining operations
have moved away from the main access shafts to mine lateral
extensions of Matanagata North east and the R1 to the south of
Smith shaft. Prior to the Phase 2 strategy, ventilation was
insufficient to supply the future mining needs, and was very
difficult to control, as air was lost through seals when taken in
down Cayzer and Smith shafts across into working over 2 km
away from Smith shaft and return to the vision shaft in close
proximity to Smith shaft. To achieve the higher development and FIG 13 - Matanagata ventilation shaft Cagi Vou
production rates, extra heavy vehicles had to be purchased and (ex bounty gold in WA).

Schematic of main ventilation flows Cayzer and Smith

shaft associated with the Phase 2 plan
Cayzer down cast chilled
air from new
refrigeration unit
100m3/sec Smith shaft down
New down cast vent Existing Vision shaft
shaft cast 100m3/sec
up cast 210 m3/sec New vent shaft for
170 m3/sec Matanagata~Smith
Proposed 200 m3/sec
bulk chilling
on surface

New
Matanagata
North East
Matanagata and
Matanagata East
R1 workings
19 level
Existing bulk chilling on 19
level fed with chilled water
from surface Ref BW~ventP2~023

FIG 12 - Diagrammatic layout of new mine ventilation system for Smith (Matanagata) and R1.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 413

B WESSON
Vertical shaft extensions and upgrades
The Philip shaft was sunk down to an elevation 719 m below
surface to level 21 level and intersected hot extraneous water.
Cement was poured into the shaft to curb the water but the shaft
continued to make water. Pumps were installed to keep the shaft
dewatered. Mining was concentrated in the upper levels above
16 level and the bottom of the shaft was allowed to fill with
spillage. In 1995, the spillage had built up to the 18 level station
and the shaft required bogging to maintain the level of spillage
below the plat for cage over run. Access to the lower levels of
Philip shaft was difficult due to high pressure hot water at
approximately 11 bar and 50C that was accounted in the
development of 1701 (587 m below surface) and 1601 (552 m
below surface).
In 1995, while mining was concentrated on the upper levels of
Philip shaft and in the Borthwick area, a pump station was
installed on 16 level Philip shaft hosting a Sulzer 710 kW, two
pole, eight stage pump that moves 2100 ML/annum of water
from the water bearing shears 567 m to surface. The hot water is
extracted from the water bearing shears through boreholes that
have sufficient pressure to push through drain columns to the
16 level hot water dam.
The zone water pressures have dropped in the lower levels
allowing mining to access the 1601 area and in line with the
Phase 2 project develop down to 18 level. The shaft loading
facilities on 16 level are very limited with a single pass system
for ore and waste which will become a bottle neck as production
and development increases in the bottom levels. In the short
term, development waste is trammed to the 13 level pass up a
decline, which incurs trucking cost and puts heat in the air
stream. To permit the increase in tonnes to be handled through 16
level and reduce tramming costs a decision was make to install a
new loading facility on 19 level. This project will require the
shaft to be bogged and a new loading pocket cut on 19 level in
the shaft barrel with three silos developed back to 18 level for
waste and ore as indicated in Figure 14.
16L Ore passes
17L
1850L
18L 18L
Decline Dev.
20L
Shaft spillage
Prince/Dolphin to be cleaned
System 19L Loading Pocket
FIG 14 - Cross-section of Philip shaft new loading pocket.
No 2 winder replacement
The Smith vertical shaft has two mine winders; one dedicated to
people and material and the other a dual winder that can be used
for mineral, material and people. The mineral winder is used for
moving the day shift as the service winder does not have
sufficient capacity. To move shift the skips are changed to cages
and the reverse for rock hoisting at the end of the shift. This
process results in ten hours of lost mineral hoisting time per day.
The mineral winder travels at 8 m/sec and can hoist 96 tonnes
414 Adelaide, SA, 19 - 22 September 2004
FIG 15 - New number two winder installed at Emperor Smith shaft.
per hour. To achieve the Phase 2 production rates the mineral
winder will be required 24-hours a day for hoisting.
To free-up the mineral winder for shift it was necessary to
replace the small 3 m/sec, AC, 135 kW, 50-year old service
winder as it has insufficient power to joist a larger cage and the
mechanical and electrical controls are coarse and labour
intensive to maintain. To achieve twice the people capacity the
single deck cage was to be replaced with a double deck cage and
cycle calculations determined that 300 kW was required to lift
the full cage single drum from shaft bottom. At the time, the
Bounty Gold Mine in Western Australia had closed and the
service winder was the ideal size for this application.
The Bounty winder was purchased for a nominal amount as
the majority of old designs do not comply with modern
Australian standards and require major modification of drums,
shafts and electrical controls to comply. The Bounty winder is a
double drum 350 kW, DC Ilner type winding system driven from
a Reliance field regulator.
The project was time consuming as the winder, electrical
control and sheave wheels needed redesign. Further, the Smith
shaft headframe built in the 1950s did not comply with
Australian Standards and the existing service winder could
buckle the structure in rope break conditions. To remedy this
situation a finite element analysis was carried out of the
headframe to understand the forces in play. The most
cost-effective solution provided by Combined Resource
Engineering Perth was to install a raker leg between the new
winder and the headframe sheave wheels to remove the stress
from the structure and the lattice beams were boxed to make the
frame more rigid.
Heavy vehicles
Ore is bogged in strike or footwall drives by Toro 151d (1.5 m3)
machines and trammed to main drive loading cuddies. The ore
loaded by EJC 130d loaders with a bucket size of 3.7 m3 and
loaded into EJC 417 or Eimco 985T15 dump trucks with a
nominal 17 tonne or 15 tonne capacity respectively.
Due to the low ore density of 2.5, the loader buckets physically
carry 12 and seven tonnes respectively. Pre-Phase 2 project the
fleet was aging with the average run hours over 30 000 and some
machines exceeded 60 000 hours and had low availability and
high operating costs.
To achieve the Phase 2 projected development and production
rates the heavy vehicle fleet required revitalisation by replacing
old vehicles with new, more productive larger vehicles, low hour
used vehicles or rebuilding existing vehicles. The fleet will
increase to 34 vehicles for the Phase 2 project and once the new
PACRIM 2004
 MINING IN FIJI

FIG 18 - New Toro 006 loader.

TABLE 2
Dump truck capacities.
Low profile truck model EJC 20 EJC 416 Eimco
or 417 985T15
FIG 16 - Raker leg for new winder. Actual capacity tonnes 18 10 7
Average cost per hour $US 17 18 29

ACKNOWLEDGEMENT
This paper does not reflect the views of Emperor Mines Limited
or its subsidiary Emperor Gold Mining Company Fiji. The views
taken are those of the author based on the management of the
projects and five years working at Emperor Gold Mine and seven
years in Fiji.

FIG 17 - New EJC 20 low-profile articulated dump truck.
areas are established and development is reduced, old vehicles
will be removed from service and the fleet will reduce back to
approximately 25 vehicles.
The new areas will have larger drives, 20 tonne dump trucks
selected to reduce tramming costs, achieve the increase in
production and reduce the overall number of vehicles required.
The loaders purchased to load the larger trucks at a higher rate
are Toro 006 LHDs with 4.2 m3 buckets. Fourteen new vehicles
have been purchased for the Phase 2 plan and seven used low
hour vehicles to take the fleet to 33 vehicles by the end of
2003/2004 financial year. At the time of writing this paper, the
EJC 20 dump trucks had operated for 12 months. The operating
cost comparison is shown inTable 2.
Table 2 considers the comparative costs and tonnes costs
associated with operating the newer larger trucks EJC 20 and
EJC 416 are the same but the EJC 20 carried 80 per cent more
tonnes significantly reducing tramming costs.
REFERENCES
Anderson, W B, Antonio, M, Davis, B, Jones, G F P, Setterfield, T N and
Tua, P, 1987. The Emperor epithermal gold deposit, Vatukoula, Fiji.
Colley, H, 1976. Mineral deposits of Fiji; Mineral Resources Division Fiji
Memoir 1.
Colley, H and Flint, D, 1995. Metallic Mineral Deposits of Fiji.
Corbett, G and Nethery, 2001. Comments on the geology and
prospectively of the Emperor Gold Mine.
Dunkin, H H, 1948. Gold Mining at Vatukoula.
Emperor Gold Mine. http://www.emperor.com.au
Emperor Mines Limited, 1996 - 2003. Annual reports.
Fiji Government online portal 2004. http://www.fiji.gov.fj Department Of
Statistics.
Hemp, R, 2002. An assessment of the ventilation and cooling aspects of
the expansion plans of Emperor Gold Mine.
Ibbotson, P, 1967. Petrology of tertiary Caldera Tavua Goldfields
(Government of Fiji Geological survey department).
Lowry, A, 1995. Bateman Kinhill process engineers detailed crusher
design and premiums based on US historical data.
Moodys country credit 2004 ratings, Moodys Sydney.
Rodda, P, 1996. History of Mines other than Manganese and Gold.
Vigar, A, 1998. Tuvatu orebody assessment.
The World Bank Group. http://www.worldbank.org/data/population
statistics in Fiji.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 415

A Three-Dimensional Structural Interpretation of the Olympic
Dam Deposit Implications for Mine Planning and Exploration
H Widdup1, T Fouet2, J Hodgkison1, T C McCuaig3 and J Miller4
ABSTRACT
Olympic Dam is the eighth largest Cu deposit and the largest U deposit in
the world, with approximately 35 million tonnes of contained Cu and
1.2 million tonnes of U. The most recent resource estimate was for
2.95 billion tonnes of mineralised material with average grades of 1.2 per
cent Cu, 0.4 kg/t U, 0.5 g/t Au, 2.6 g/t Ag as contained mineral resources.
Olympic Dam is expected to produce >200 000 tonnes of Cu and 4200
tonnes of U3O8 from 9.8 million tonnes of ore in 2004, making it the 12th
largest Cu producer in the world. One of the impediments to achieving
targeted rates of production has been oversize material (>0.9 m) reporting
to stope draw points. A major contributor to oversize material may be
overbreak along pre-existing structural weaknesses from stope walls and
backs during the production life of a stope. Underbreaking to structures
can also have a detrimental effect on the performance of a stope.
In an effort to understand the influence of structural geology on stope
performance, WMC Resources and SRK Consulting partnered during
2003 to refine the current system of structural mapping, and to use
historical data to produce a structural interpretation covering a large part
of the mining area. The interpretation has been wireframed in
DATAMINETM into a three-dimensional structural framework model. To
date, there are more than 600 structures of >50 m strike length
incorporated into the structural framework model, which covers a study
area 1500 m 600 m 350 m. A back-analysis exercise to assess the
influence of interpreted structures on overbreak and underbreak was
completed for laser-surveyed stopes.
The information in the structural framework model is used as a tool to
estimate potential overbreak and underbreak in planned stopes in an
attempt to minimise the formation of oversize material. There have been
several flow on benefits discovered since the model was completed. The
model has also been applied to interpreting data from a micro-seismic
system. In the Stuart Shelf Geological Province, there is a spatial
association between intersections of WNW and NNE trending structures
and Fe-oxide deposits, which may contain Cu-Au mineralisation. In the
future this association may be applied to exploration for high-grade Cu
shoots within the Olympic Dam deposit.
A structural orientation analysis has been completed of the structural
framework model. Combining the results of this study with the work of
previous authors highlights the existence of structural domains. There is a
reduction in structural density from the top mine level to the bottom mine
level studied. Moving along strike of the orebody from south-east to
north-west, there is a reduction in structural density with structural trends
becoming far more distinct for significant structures (>50 m strike
length). There is also a decrease in the thickness and occurrence of
late-stage barite-fluorite sheared veins from south-east to north-west.
Significant structures are commonly strike-slip, with some examples of
reverse faults. The orientation of these structures is mostly in an arc from
N-S trending, through NW-SE to E-W trending. The pattern of significant
structures is overprinted by many shorter, brittle faults and joints of varying
orientations.
INTRODUCTION
The super-giant Olympic Dam Cu-U-Au-Ag-REE deposit is
situated approximately 560 km north-northwest of Adelaide,
1. MAusIMM, WMC Olympic Dam Operation, Technical Services
Department, PO Box 150, Roxby Downs SA 5725.
2. WMC Olympic Dam Operation, Technical Services Department, PO
Box 150, Roxby Downs SA 5725.
3. MAusIMM, SRK Consulting, 1064 Hay Street, West Perth WA 6005.
4. School of Earth Sciences, Department of Earth Sciences, University
of Melbourne, Parkville Vic 3010.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
near the eastern margin of the Gawler Craton (Figure 1) and is an
example of a Proterozoic Fe-oxide Cu-Au system. Discovered in
1975 by WMC Resources and commenced production in
mid-1988. Following a period of rapid expansion during the
mid-late 1990s, the operation targets a production rate of more
than nine million tonnes of ore a year. Production targets have
been impeded in the past by oversize material (>0.9 m) reporting
to the stope draw points.
Oversize material cannot easily be moved by loaders and is too
large to fit through ore pass grizzlies, creating additional
secondary breakage requirements. The practice of drilling,
charging and blasting oversize material in stope draw points can
create delays to the production schedules. Oversize material may
consist of overbreak from stope walls or backs, or incompletely
blasted material from within the stope design. Fall off can affect
the stability of adjacent stopes, or result in dilution of the mined
ore.
Structural mapping data was used to produce a
three-dimensional structural interpretation which can be used as a
predictor of primary weaknesses that could be causing overbreak
and underbreak within existing stopes. The current structural
interpretation covers a study area that comprises four consecutive
mine areas in the northern part of the mine (Figure 2).
The study area extends over 1500 m strike length and is
approximately 600 m wide (Figure 2). The study was conducted
between the upper most mine levels and the lowest extraction
level, between ~350 m and 700 m below the surface. Production
from the study area constitutes approximately 60 per cent of the
total mine production. The structural model will soon be
extended to cover the entire mine.
This study presents a three-dimensional structural framework
model of a large part of the Olympic Dam mine that was
commenced to investigate the formation of oversize material in
stopes. Numerous flow on benefits that have been gained from
producing the model are also discussed.
REGIONAL GEOLOGY
The Olympic Dam Cu-U-Au-Ag-REE deposit occurs on the
eastern margin of the Gawler Craton, within the Stuart Shelf
geological province (Figure 1). Palaeoproterozoic metasediments
of the Hutchison Group and Lincoln Complex granitoids were
intruded during the Mesoproterozoic by the Hiltaba Suite
granitoids (1600 - 1585 Ma) (Parker, 1990). The Gawler Range
Volcanics are the extrusive equivalent of the Hiltaba granitoids
(Flint, 1993). The Gawler Craton is bounded 75 km to the east by
the NNW trending Torrens Hinge Zone, where undeformed
sediments of the Stuart Shelf and the Gawler Craton basement to
the west are juxtaposed with variably deformed thicker
sediments of the Adelaide Geosyncline to the east. The deposit is
entirely contained by the Olympic Dam Breccia Complex
(ODBC), which is hosted by the Roxby Downs Granite (Reeve et
al, 1990), a member of the Hiltaba suite with A-type affinities
(Creaser and Cooper, 1993).
Detailed descriptions of the distribution of lithologies in the
ODBC have been published by Oreskes and Einaudi (1990) and
Reeve et al (1990). The ODBC represents an area of the Roxby
Downs Granite that has undergone extensive hydrothermal
417
H WIDDUP et al

133 E

137 E
o

o
TO
RR
EN
STUART o

S
30 S
SHELF
OLYMPIC DAM

ROXBY DOWNS

H IN G E
WOOMERA

o
CEDUNA 32 S

ZONE
NT
QLD PORT AUGUSTA
WA
NSW
VIC
TAS

SOUTH AUSTRALIA

OLYMPIC DAM

North
Adelaide

o
34 S
Late Sedimentary cover
Proterozoic

Pandurra Formation
PORT LINCOLN
Middle Gawler Range Volcanics
Proterozoic
Hiltaba Suite

Corunna Conglomerate/ Tarcoola Formation

Ifould Complex
Early
Proterozoic Hutchison Group/ Lincoln Complex/ Mt Woods complex/
Coober Pedy Ridge Complex & equivalents
0 100

Mulgathing Complex/ Sleaford Complex kilometres

Archaean

FIG 1 - Interpreted sub-surface geology of the Gawler Craton (modified after Daly, Fanning and Fairclough, 1998).

418 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A THREE-DIMENSIONAL STRUCTURAL INTERPRETATION OF THE OLYMPIC DAM DEPOSIT

Ar

678000mE

680000mE

682000mE
ea
FNW ss
tu
Re died
ev by
6632000mN ee S 6632000mN
t a ug
l (1 de
ST 99 n (1
UD 0) 99
2);
YA
RE Southern
A Mine Areas
FN
F DNW
DC

ROBINSON SHAFT

CLARK SHAFT
WHENAN
SHAFT
LEGEND
Granite rich breccias

Matrix rich granite breccias

North
Heterolithic granite 6630000mN
and hematite breccias

Hematite rich breccias

Hematite-quartz breccias
(barren core complex)

Volcaniclastics

0 500 1000

metres
678000mE

680000mE

682000mE
FIG 2 - Simplified geology map of the Olympic Dam Breccia Complex. Shows the mine areas covered in this study, as well as the area
studied by previous authors to the south-east. Mining has taken place around the northern margin of the barren core complex and along
strike of the breccia complex to the north-west.

alteration and chemical and mechanical brecciation. Magma
derived Fe-rich mineralising fluids replaced silicates with
hematite to form a lithological continuum. Granite-rich, clast
supported breccias containing hematite along fractures and veins
occur near the margins of the deposit and represent one end
member. A hematite-quartz matrix-rich breccia, which is barren
with respect to Cu and U, occurs in the centre of the deposit and
constitutes the other end member. Between the two end
members, breccias containing varying proportions of granite and
hematite components occur.
Breccia bodies in the ODBC are arranged along a WNW
oriented axis. Hematite content varies enormously along the strike
of the deposit, but breccia bodies are commonly hematite-rich at
the centre grading outwards to granite-rich breccias. All of the
granitic rocks in the ODBC are characterised by a complete
replacement of plagioclase by sericite. Locally, granite-rich rocks
may be altered to sericite-chlorite assemblages, especially where
fluid has infiltrated along faults. Figure 2 shows the simplified
distribution of lithologies within the ODBC.
Mineralisation is zoned around the barren core of
hematite-quartz. A chemical progression from Cu-rich, S-poor
minerals (eg native Cu, chalcocite, bornite) towards S-rich,
Cu-poor minerals (eg chalcopyrite, pyrite) is exhibited moving
away from the barren core. U (in uraninite), Au and Ag are
associated with Cu sulfides in varying concentrations.
The Proterozoic basement that hosts the Olympic Dam deposit
is unconformably overlain by more than 300 m of flat-lying
sediments of the Stuart Shelf Sequence. Previous workers have
used a range of geophysical data, combined with mapping of rare
basement outcrops elsewhere in the Gawler Craton to propose
various structural syntheses of the basement.
ODriscoll (1985) applied a lineament tectonics analysis to
develop prospective Cu targets in the Stuart Shelf geological
province. This work used air photo mosaics and geophysical
images to map linear features, and identified strong NNW and
ENE trending lineaments as well as weaker NNE and WNW
trends. Intersections of lineaments were found to be commonly
spatially associated with known ore deposits. The Olympic Dam
deposit was discovered in 1975 by WMC at the coincident
occurrence of gravity and magnetic highs, the intersection of a
WNW linear trend and a circular feature seen in photo lineament
analyses, and the intersection of the G2 (NNW) and G9C (WNW)
bouger gravity lineaments. These features were interpreted to
reflect major regional structures (ODriscoll, 1985, 1990).
More recently, workers undertaking geophysical studies have
proposed structural syntheses of the basement in the Olympic
Dam region. Hiskins (1999) proposed early WNW to NW faults
cross-cutting Lincoln Complex granites. These faults appear to
have exercised considerable influence on the intrusion of Hiltaba
suite granites. Following the extrusion of the Gawler Range
Volcanics, NE oriented faults truncated some NW trending
features. Structural features oriented WNW to NW have been
subsequently re-activated, and were further exploited by intrusive
dykes of the NW trending Gairdner Dyke Swarm (ca 820 Ma
Wingate et al, 1990). Betts, Giles and Lister (2002) proposed that
the distribution of Hiltaba intrusives was controlled by three
systems of shears. Granites were emplaced adjacent to dextral
ENE reidel shears within an E-W shear that cross-cuts the
Gawler Craton. WNW oriented faults exercised a local influence
on granites, and NNW oriented structures appear to have
localised intrusions. Hiskins (1999) and Betts (1999) both
recognised an association of Fe-oxide deposits at the intersection
of WNW and NNE trending structures.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 419

H WIDDUP et al
Previous structural studies at Olympic Dam were undertaken
in the south-eastern portion of the current study area. Reeve et al
(1990) and Sugden (1992) recognised several major structural
components. These are:
strike-slip faults trend WNW in the north-west of the deposit,
whilst in the central region of the deposit strike-slip faults are
NNW-trending;
reverse faults in the south-east trend NW and dip gently to
the SW and NE, whilst reverse faults dip predominantly SW
in the north-west of the deposit; and
a conjugate set of late-stage barite-fluorite veins are oriented
NE (dextral) and E-W (sinistral). Joint sets measured in the
cover sequence display similar orientations to the late vein
sets. These conjugate vein arrays were interpreted by Sugden
(1992) to have formed during the ca 500 Ma (Preiss, 1987)
Delamarian Orogeny.
Sugden (1992) identified ultramafic and mafic dykes that
occupy NW-NNW and NE-NNE trends, and possibly exploit the
early strike-slip faults. Sugden (1992) also identified a major
diatreme structure along the northern margin of the barren core
complex, at the intersection of two mine scale faults oriented
NNW and WNW. Both Reeve et al (1990) and Sugden (1992)
argued that most of the fault network observed at Olympic Dam
developed post mineralisation, after much of the brecciation
episode had abated.
Cross, Daly and Flint (1993) recognised that individual breccia
bodies are oriented NNW, arranged along a WNW trending axis.
This was interpreted to represent a dextral dilation zone along a
WNW trending structure, which may have caused Olympic Dam
to become the focus for magma and hydrothermal fluid flow.
Hiskins (1999) interpreted an ENE trending structure that
intersected the Olympic Dam deposit that may correspond with
Mashers Fault (Sugden, 1992). Recent geophysical
interpretations (Betts, Giles and Lister 2002) failed to detect
discrete structures that correspond directly with the WNW and
NNW trending lineaments interpreted to intersect at Olympic
Dam by ODriscoll (1985, 1990).
METHODOLOGY
The study area consists of four consecutive mine areas, with
approximate dimensions 1500 m 600 m 350 m (Figure 2).
The structural model is based on interpretations made on mine
level plans, for each of the four connected mine areas. DC is the
most south-easterly mine area in the study area, located
approximately 500 m north west of the barren core. Moving
north-west along the strike of the orebody, DC is followed by
DNW, F and FN. The FN mine area consists of two stoping areas
Amber to the north and Cyan to the south (abutting F). Only
the Cyan stoping area is included in this study, so the component
of FN referred to in this paper will be called Cyan.
Geologists routinely map the backs and walls of development
drives as they are mined, at 1:500 scale. Any structure that can be
traced for >3 m in the backs or walls is mapped, and the type and
intensity of the structure, infill, alteration, and the presence and
type of kinematic indicators is noted. This range of information
has not been systematically recorded in the past. Most historical
mapping data consists only of orientation (dip and dip-direction)
and the type of structure. The information collected is digitised
on to level plan maps of the backs and is stored in Microstation.
Hard copy maps (1:500) for each mine area in the study area
were produced, showing the outline of development drives,
structure and lithology mapping data and stope void outlines. Only
structures that were mapped from one side of a drive to another
were included in the interpretation (unless they were interpreted to
be the weak continuation of another structure). Structures were
420 Adelaide, SA, 19 - 22 September 2004
interpreted between mapped drives where similar features could
be correlated along strike, allowing for realistic variations of strike
and dip. Structures are commonly observed that change dip or
strike orientation by up to 30 from one side of a drive to another,
but generally occupy a roughly planar orientation. Structures were
grouped according to their strike length: ~5 - 30 m, 30 - 50 m and
>50 m. This indicates their potential to effect the stability of a
void. Most stopes at Olympic Dam have less than a 30 m 30 m
footprint. As stopes are unlikely to be affected by small structures,
30 m is considered to be the minimum strike length of a
significant structure.
Mine levels are spaced approximately 30 - 60 m apart, but most
commonly at ~40 m. As a result, cross-sections at 40 m spacing
were prepared showing the outline of development drives and
stope voids with drilling data including lithology, structure,
mineralisation and alteration. Completed level plan interpretations
were transferred to cross-sections and interpretations were made
between levels. In order for a structure to extend from one mine
level to another, it would have to have a dip extent of >~50 m.
The level plan structural interpretation was ground truthed
underground once level to level correlations were made. Very
few corrections or additions were made, adding confidence to
more than a decade of mapping data and giving credibility to the
interpretations. Ground truthing also allowed the observation of
kinematic indicators and alteration characteristics of structures,
and the quality of the rockmass proximal and distal to significant
structures. The ground truthed mine level plans were digitised
and the strings were annotated in DATAMINETM. As well as dip
and dip-direction, structures were given attributes including type,
intensity, name, infill, thickness (where available). These
attributes make it possible to filter strings by structure name to
display the strings for one or more structures at a time.
Wireframes were constructed, linking strings from level to
level and extrapolating structures between levels. Strings were
extrapolated up/down dip where there was no evidence for the
structure mapped on the level above/below, according to a set of
rules. The structures were extrapolated either half of their strike
length up and down dip, or half the distance to the next level
along dip (the smallest distance from both possible projections
was used). Shallow dipping structures (<30) were extrapolated
the lesser of half way to the next level, half the strike distance, or
ten metres.
The resultant model contains wireframes of significant
structures (>30 m strike length). There are over 14 000 structures
with strike length <30 m interpreted in the study area that have
not been wireframed.
STRUCTURAL FRAMEWORK MODEL AND VOID
STABILITY ANALYSIS
Geologists at Olympic Dam provide structural data and
interpretations to geotechnical engineers and mine design
engineers in order to estimate the behaviour of the rockmass
when part of it is mined. Occasionally geologists are asked to
provide an interpretation of an area that is experiencing problems
whilst it is being mined, such as broken ground in development
headings or within stopes. Historically, such interpretations were
done for small areas on an as needs basis. The structural
framework model provides a first stop for engineers and is being
used earlier in the design process to modify stope designs or
firing sequences where there is a structure that has been
interpreted that may form an area of instability with the potential
to form oversize material. The simplest scenario of void stability
being compromised by the presence of structures is where two
moderate (45), opposite dipping structures intersect. If a void
(such as a drive or a stope) were opened beneath the intersection,
an unsupported wedge may also fall out.
PACRIM 2004
 A THREE-DIMENSIONAL STRUCTURAL INTERPRETATION OF THE OLYMPIC DAM DEPOSIT

overbreak
A B
underbreak

120m
th
Nor

C D
over-
break
designed
80m

crown
level

th
Nor

FIG 3 - Examples of mined stopes from within the study area, showing laser surveyed mined shapes versus stope design shape. Arrows
show areas of major breakage to structure planes. (A) Steeply south-east dipping structure has caused overbreak and underbreak towards
the top of the mined stope, shaping both sides of the void. View looks north-west. (B) Identical view to (A) comparing laser surveyed mined
stope shape with design shape. (C) Steep south-east dipping structure has caused overbreak above the crown of the stope. View looks
north-west. (D) Identical view to (C) comparing laser surveyed mined stope shape with design shape.

Figure 3 shows significant planar overbreak and underbreak in
laser surveyed stope shapes to major structures that have been
interpreted as part of the structural framework model. The first
example (Figure 3a) displays slight underbreak along the eastern
wall to a continuous, steeply east dipping structure. This
underbreak is close to the designed wall of the stope (the design
shape is shown relative to the laser surveyed shape in Figure 3b).
The same structure has caused overbreak on the western side of
the crown. In the second example (Figure 3c) an interpreted
structure intersects the crown of the designed stope along the
western edge. Crown failure was exacerbated by the presence of
this structure, which pre-split the western wall and assisted in a
major crown failure (Figure 3d).
Structures are well known to cause overbreak and underbreak
in development drives at Olympic Dam. Figure 3 illustrates how
the same can occur in stopes. When unblasted rock falls out of
crown pillars or stope walls there is a strong possibility this will
be of larger dimensions than blasted rock. The resultant oversize
material is cumbersome to move and in extreme cases may block
stope draw points. Material that originates outside of a stope
design shape can potentially dilute ore grades, or de-stabilise the
walls of an adjacent planned or mined void.
Overbreak can potentially undercut development drives, or
may cause the loss of one that has been drilled ready for firing.
Underbreak may result when a structure pre-splits and prevents
effective blasting. Less than the anticipated volume of rock is
mined as a result.
If a significant structure daylights into a stope shape and the
stope shape is mined, the change of stresses in the area may
cause the structure to move. This may result in a seismic event,
unplanned fall off, or the deterioration of ground conditions
within the stope or in adjacent development. WMC Resources is
in the process of developing and installing a mine wide
micro-seismic system at Olympic Dam, for detecting and
analysing microseismic events in the vicinity of the mine. When
a seismic event is recorded, the location can be calculated using
the arrival times of the P and S waves at three or more sensors.
The location of the event may be correlated to an interpreted
structure from the structural framework model.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 421

H WIDDUP et al
Whilst granitic rocks in the ODBC feature complete
replacement of plagioclase by sericite, intense sericite chlorite
alteration assemblages are commonly found in association with
major structures. Alteration may be limited to the fault gouge
developed within the slip plane, but in many cases also affects
the rock for up to tens of metres. Intensely sericite-chlorite
altered rocks are weaker than most of the lithologies in the
ODBC and may be more prone to collapse or crumbling. The
presence of sericite and chlorite along structure planes
significantly weakens the structure.
Geotechnical control is closely linked to safety of people and
the efficiency of mining activity. If falls of ground, oversize
material in stopes and unexpected seismic events can be
minimised, the potential for process losses, equipment damage
and unsafe conditions can be minimised.
STRUCTURAL SYSTEM
The ODBC is overprinted by a cross-cutting and interconnected
system of brittle faults, joints and veins that most likely formed
late syn- to post-brecciation, and in deformation events
subsequent to this. Faults in the ODBC commonly display a
narrow (<10 cm) zone of gouge or breccia, associated with an
anastomosing fault plane. Damage zones of up to 0.5 m have
been reported. Movement indicators may be obscured by strong
alteration of fault gouge, or reactivation. In many cases multiple
generations of slickenlines are present, and interpretation of the
oldest event is difficult. The network of faults is accompanied by
brittle joints, and veins. Joints are planar, and continuous over
strike lengths of up to 80 m. Joints are commonly present in
subparallel joint sets, bunched in narrowly spaced (0.2 m) zones
up to ten metres thick, or in widely spaced (ten metres) sets that
foliate a much larger area (hundreds of metres). Veins are
commonly <1 cm thick and may contain a combination of
hematite, sericite, chlorite, siderite, barite, fluorite, quartz,
sulfides or pitchblende (Reeve et al, 1990). Later veins may be
infilled with up to 2 m of barite fluorite, chalcopyrite, bornite,
pyrite, siderite.
Data for the analysis of orientations in the structural
framework model was taken from the strings of structures >30 m
strike length used to construct the model. Each string represented
the interpreted trace of a structure on a level, and had dip,
dip-direction, and structure type attributes as well as northing,
easting and RL. This information was output from
DATAMINETM into Microsoft Excel, and was sorted into mine
area groups to assess trend changes along the strike of the
ODBC. Data was also sorted by strike length to assess the
contribution of low continuity versus high continuity structures
to the trends, and finally by structure type. Data is presented on
lower hemisphere, equal angle contoured pole plots with planes
added for significant clusters.
TABLE 1
Structural trends in each of the mine areas within the study area, for structures 30 - 50 m strike length and >50 m strike length.
30 - 50 m
DC Scattered pole clusters
NW trending, steep to shallow dipping
Range of strike directions covering N-S/NW/E-W range
DNW N-S/NNW trending, steep to shallow dipping
E-W trending sub-vertical
F N-S/NNW/NW trending, steep to shallow dipping
E-W trending sub-vertical
Cyan E-W trending sub-vertical
NNW trending, steep to moderate dipping
NW trending, steep to moderate dipping
Entire study area N-S/NNW/NW trending, steep to shallow dipping
E-W trending sub-vertical
422 Adelaide, SA, 19 - 22 September 2004
Table 1 summarises the structural trends for structures
30 - 50 m strike length, and >50 m strike length for each of the
mine areas in the study area. The most common strike trend is
NW-SE, accompanied by lesser N-S and E-W trends. Figure 4 is
a composite form surface map of the 26 and 28 levels (the
uppermost mine level), it has inset contoured pole plots of the
significant structures (>30 m strike length) in the study area (on
the left) and for each mine area (on the right). These stereonets
correspond to data summarised in Table 1.
Moving north-west from DC to Cyan there is an evolution of
structural trends. N-S, NW and E-W trends are evident
throughout the study area. In the south-east (DC) there is a range
of strike trends between N-S and E-W (centred on NW-SE).
Within this there are indistinct clusters of poles suggesting strike
trends at N-S, NNW, NW and ENE. In the north-west similar
trends become far more distinct.
Structures with 30 - 50 m strike length display N-S/NW/E-W
trends in scattered, indistinct clusters with the data spread over a
larger area. This indicates that there is considerable variation in
the orientations of shorter strike length structures (Figure 5a).
The structures with strike length >50 m display trends in a range
from N-S striking structures through NW to E-W striking
structures, with pole clusters that correspond to distinct trends
(Figure 5a). The most dominant strike direction for structures
with strike length >50 m is NW.
Data for faults is dominated by N-S to NW trends (Figure 6a),
the contoured pole plot resembles the plot for the entire study
area (see inset on Figure 4). This is not surprising, as faults make
up ~80 per cent of the structures measured. The stereonets also
highlight a weaker but distinct E-W trending set of faults. Joints
display variable orientation trends, dominated by a strong E-W
trend, with many pole clusters ranging from N-S through NW to
E-W trending (Figure 6b). Veins display strongly clustered poles
for an E-W to NE strike trend with lesser N-S and NW trends
(Figure 6c).
Similar strike trends that are apparent in stereonets are visible
on a structural map of the study area (Figure 4). The most
continuous structures in the study area are NW or NNW striking,
closely followed by E-W oriented structures. On the map, there
is a northward evolution of the dominant trends of major
structures that isnt immediately apparent in the contoured pole
plots. The most continuous structures are vastly outnumbered in
stereonet analyses, but stand out on a form surface map. In DC,
the pattern is dominated by long striking E-W faults and sheared
veins, with less continuous but more common N-S sheared veins
and faults. This pattern is overprinted by shorter striking joints
and faults (<30 m) of various orientations. In some cases they are
subparallel to major structures, though many exhibit complicated
cross-cutting relationships. Moving north-west into DNW and F,
>50 m
N-S trending, steep to moderate dipping
WNW trending, steep to moderate dipping
E-W trending sub-vertical to steep dipping
NW trending, steep to moderate dipping
E-W trending sub-vertical
NNW trending, steep to moderate dipping
E-W/ENE trending sub-vertical
NW trending, steep to moderate dipping
N-S trending sub-vertical
E-W trending sub-vertical
NW trending, steep to moderate dipping
N-S/NNW/NW trending, steep dipping
E-W trending sub-vertical
PACRIM 2004
 A THREE-DIMENSIONAL STRUCTURAL INTERPRETATION OF THE OLYMPIC DAM DEPOSIT

CYAN LEGEND
>50m strike length, may be
interpreted from one level to
another

30-50m strike length

~5-30m strike length

250m
F

North

DNW

Data from entire

study area

Location of form surface map within the ODBC

North

DC

FIG 4 - Composite structure map of the 26 and 28 levels (the top mine level). Lower hemisphere, equal angle contoured pole plots are
inserted that illustrate trends of significant structures (>30 m strike length) for each mine area as well as for the entire study area.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 423

H WIDDUP et al

the dominant strike trend is NW (with less extensive WNW and

A B NNW) steep to moderate west dipping faults. There is a similar
effect of overprinting by shorter striking structures. In the Cyan
area, the most dominant structures trend NW (dipping steeply to
the NE), with cross cutting WNW to E-W structures. There is an
overall decrease in structural density moving from DC towards
Cyan, as well as from the top mine level (26/28 level) to the
bottom mine level studied (57 level).
N = 1589 N = 631 Many of the overprinting shorter striking structures are planar
joints, or faults that display a small amount of normal movement.
Geotechnical point data and general observations show that the
FIG 5 - Contoured pole plots for (A) all structures 30 - 50 m strike rock mass is generally massive in areas away from major
length in the study area and (B) all structures >50 m strike length structures. Blocks formed from secondary jointing are smaller in
in the study area. size proximal to major structures. The structural trends observed
in the study area are in broad agreement with the earlier
A observations of Reeve et al (1990) and Sugden (1992), with a
few important variations and additions.

DISCUSSION
The trends for strike-slip and reverse fault sets identified by
previous authors are in agreement with the fault trends identified
here. The strong E-W strike trend identified of faults in the study
N = 1780
area matches closely the E-W trend of sheared veins that both
Reeve et al (1990) and Sugden (1992) recognised. E-W shears
B that exist in the north-west may represent a family of faults that
were infilled by vein material in the south-east during an episode
of later deformation.
None of the earlier studies comment on joint orientations,
except in close relation to major structures. There are numerous
significant joints (with a strike length that is in excess of 30 m) in
the ODBC that display a planar surface and no discernible offset
or movement indicators. There are numerous examples of zones
N = 116 of subparallel joint sets up to 10 m thick. These joint sets may
represent the weak continuation of a fault. Alternatively, joints
C sets may foliate the intermediate zone between two shears, where
stress has been transferred from one plane to another (Figure 7).
Sugden (1992) and Reeve et al (1990) identified NE and E-W
oriented barite-fluorite veins, many with filled jog zones that
suggested the vein arrays formed a conjugate network. This
study identifies two new vein trends: N-S sheared veins, and NW
veins (with no evidence of shearing reported to date). Thick (up
to 2 m) N-S trending sheared veins are most common in the
N = 324 south-east of the study area in DC and DNW. There is an
abundance of narrower (up to 50 cm) NW and N-S trending
FIG 6 - Contoured pole plots (A) all significant faults (>30 m strike
veins in the north-west of the study area, although they appear to
length), (B) all significant joints (>30 m strike length) and (C) all
occupy mainly extensional fractures that display no evidence of
shearing.
significant veins (>30 m strike length).

A.

B.

Joints form in tension

between two faults

FIG 7 - Joints sets may be associated with faulting, (A) along strike of a fault as the main shear weakens or
(B) taking up extension in the jog zone between two similar faults.

424 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 A THREE-DIMENSIONAL STRUCTURAL INTERPRETATION OF THE OLYMPIC DAM DEPOSIT
The differences between previous studies (Reeve et al, 1990;
Sugden, 1992) and this study illustrate important structural trend
variations within the ODBC. Moving along strike from
south-east towards the north-west there are the following
evolutions:
A WNW structural trend (reflecting the major regional trend)
is most prominent closest to the barren core. WNW trending
structures are absent in all of the mine areas north-west of
DC (which contains rare examples of WNW faults).
A reduction in the amount of reverse faults, with strike-slip
faults dominating the interpreted fault pattern in the
north-west of the study area.
Thick (up to 2 m) sheared veins in DC and DNW give way to
narrower (up to 50 cm), extensional veins in F and FN. There
is no evidence for shearing along veins in the north-west of
the study area. N-S and NW trending veins have been
identified in this study.
E-W trending sheared veins found in the south-east
correspond with similarly oriented faults in the north-west,
these faults feature little or no vein fill.
Discrete structural strike trends at N-S, NW and E-W in
Cyan emerge from a range of strike directions in DC, which
are centred around NW-SE.
There is a reduction in the amount and magnitude (thickness)
of veins. Veins up to 2 m thick that cross-cut the entire
ODBC are common in DC and to the south of the study area,
but less common in Cyan, where veins are up to 50 cm thick.
There is an overall decrease in structural density to the NW
and at depth.
Previous investigations have never gone as far as to interpret
the structural system on multiple levels for such a large part of
the deposit, nor to join these interpretations from level to level in
a three-dimensional structural framework model. The
development of a three-dimensional structural framework model
has added greatly to the understanding of the distribution of
structures and the existence of structural trend domains. Above
all, this model illustrates important structural heterogeneities that
may contribute to the formation of oversize material.
A change in structural density moving along strike of the
ODBC from the barren core towards the north-west as well as
with depth has been recognised. In addition to variations along
strike and with depth, there is a relationship between the
structural model and the characteristics of the rock mass. Major
structures are associated with secondary jointing that tends to
form small blocks. An increased incidence of structures means
there is more potential to form fault bounded blocks that may not
fragment when blasted and could form oversize material.
Likewise, a higher abundance of structures at shallow depths
suggests there is a greater potential for crown pillar failure.
The differences in dominant trends and structural
characteristics from area to area could be used to construct
geotechnical domains. The trends and characteristics of
structures in a domain may be considered to apply globally
throughout the domain. All of the stopes in a domain could then
be designed to allow for potential planar weaknesses in the
orientations established to be the most significant for that area, in
order to minimise unplanned fall off and the formation of
oversize material.
IMPLICATIONS FOR EXPLORATION
Mineralisation is thought to be temporally related to the
development of hematite breccias (Oreskes and Einaudi, 1990;
Reeve et al, 1990; Johnson and Cross, 1995). Fluids and melts
were probably introduced along early structures that may have
formed in an en-echelon gash at a jog zone on a dextral, WNW
oriented regional shear (Cross, Daly and Flint, 1993). The jog
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
zone appears to be associated with an intersection with a NNW
oriented structure. The intersection of a WNW and NNW
oriented structures have been reported to correspond with a
diatreme structure within the ODBC. Intersections of WNW and
NNE structures have been reported to be associated with
Fe-oxide deposits in the Stuart Shelf geological province and the
Mt Woods inlier (Hiskins, 1999; Betts 1999). These associations
(together with the evidence presented by ODriscoll, 1985; 1990)
suggest that intersections of mine-scale WNW with NNW
structures, and WNW with NNE structures be made the target of
future exploration, as they may potentially be associated with
high-grade or high-tenor Cu.
In many deformed ore deposits there is an inherent possibility
that ore has been offset along a large scale structure. This hasnt
previously been accounted for in drilling or exploration at
Olympic Dam. Sugden (1992) suggested that most of the faults
affecting the ODBC featured offsets of <10 m. This magnitude of
structures would not cause much concern in an orebody the size
of Olympic Dam. However, now that a structural model exists,
the most significant cross-cutting features can be identified and
investigated, especially where they truncate ore zones.
CONCLUSIONS
The introduction of metalliferous fluids and or melts may have
been controlled by an ancient structural conduit. The original
system was overprinted and healed by brecciation, chemical
precipitation and emplacement of melts (Reeve et al, 1990;
Oreskes and Einaudi, 1990). On the basis of the distribution of
breccia bodies it may be speculated that a major WNW oriented
structure combined with NNW directed structures (that fit into
the regional structural framework) were prominent in the original
structural geometry at Olympic Dam. The ODBC has been
overprinted by a complex system of cross-cutting and
interconnecting brittle structures that have the potential to form
oversize material, cause dilution or produce bridges in stopes.
A three-dimensional structural interpretation of a large part of
the Olympic Dam deposit has been wireframed, to produce a
structural framework model. Investigations reveal that overbreak
(outside stope design) and underbreak (inside stope design) is
commonly planar, and in many cases corresponds closely with
the position of interpreted structures.
The structural framework model is now used in mine planning
and void stability analysis, to estimate the likely stability of voids
and to attempt to estimate the occurrence of potential oversize
material that may occur in stopes and report to draw points. The
reduction of oversize material would reduce the subsequent
machinery stand-by and remedial secondary breakage that needs
to take place.
Structural orientation analyses of the structural framework
model show an evolution of structural trends from the area
around the barren core complex, moving north-west along the
strike of the ODBC. Trends identified in the south-east of the
study area correspond with the established framework proposed
by earlier workers. This framework holds for the north-west also,
with the addition of two vein trends (N-S and NW), and a
reduction in the significance of E-W sheared veins.
There is an overall reduction in structural density moving
downwards, and moving along strike from the barren core
complex towards the north-west of the deposit. Similarly,
increased secondary jointing is common in association with
major structures. The spatial qualities of the structures in the
ODBC can be used in conjunction with the structural framework
model to characterise geotechnical domains within the mine.
As well as uses in detailed mine planning and void stability
analysis, the structural framework model may have implications
for in-mine exploration. Intersections of structures may be used
to generate drilling targets for high-grade or high-tenor Cu in the
future.
425
H WIDDUP et al
REFERENCES
Betts, P G, 1999. Tectonic and structural evolution of the Mt Woods
inlier: Implications for the Palaeo- to Mesoproterozoic tectonic
history of the Gawler Craton, Australian Crustal Research Centre
publication No 72.
Betts, P G, Giles, D and Lister, G S, 2002. Tectonic setting of the
Olympic Dam Fe-oxide Cu-Au-U-REE deposit, Australian Crustal
Research Centre publication No 89.
Creaser, R A and Cooper, J A, 1993. U-Pb geochronology of middle
Proterozoic felsic magmatism surrounding the Olympic Dam
Cu-U-Au-Ag and Moonta Cu-Au-Ag deposits, South Australia,
Economic Geology, 88:186-197.
Cross, K C, Daly S J and Flint, R B, 1993. Olympic Dam deposit, in The
geology of South Australia, Volume 1 The Precambrian, (Eds: J F
Drexel, W V Preiss and A J Parker) pp 132-138 (Geological Survey
of South Australia) Bulletin No 54.
Daly, S J, Fanning, C M and Fairclough, M C, 1998. Tectonic evolution
and exploration potential of the Gawler Craton, South Australia,
AGSO Journal of Australian Geology and Geophysics, 17(3):145-168.
Flint, R B, 1993. Hiltaba suite, in The Geology of South Australia,
Volume 1 - The Precambrian, (Eds: J F Drexel, W V Preiss and A J
Parker) pp 127-131 (Geological Survey of South Australia) Bulletin
No 54.
Hiskins, D J, 1999. An investigation of the Olympic Dam
Cu-U-Au-Ag-REE deposit, South Australia, Australian Crustal
Research Centre publication No 74.
Johnson, J P and Cross, K C, 1995. U-Pb geochronological constraints on
the genesis of the Olympic Dam Cu-U-Au-Ag deposit, South
Australia, Economic Geology, 90:1046-1063.
426 Adelaide, SA, 19 - 22 September 2004
ODriscoll, E S T, 1985. The application of lineament tectonics in the
discovery of the Olympic Dam Cu-Au-U deposit at Roxby Downs,
South Australia, Global Tectonics and Metallogeny, 3(1):43-57.
ODriscoll, E S T, 1990. Lineament tectonics of Australian ore deposits,
in Geology of the Mineral Deposits of Australia and Papua New
Guinea (Ed: F E Hughes), pp 33-41 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
Oreskes, N and Einaudi, M T, 1990. Origin of rare earth element-enriched
hematite breccias at the Olympic Dam Cu-U-Au-Ag deposit, Roxby
Downs, South Australia, Economic Geology, 85(1):1-28.
Parker, A J, 1990. Gawler Craton and Stuart Shelf regional geology and
mineralisation, in Geology of the mineral deposits of Australia and
Papua New Guinea Monograph 14 (Ed: F E Hughes), pp 999-1008
(The Australasian Institute of Mining and Metallurgy: Melbourne).
Preiss, W V (comp), 1987. The Adelaide Geosyncline Later Proterozoic
Stratigraphy, Sedimentation, Palaeontology and Tectonics (Geological
Survey of South Australia) Bulletin No 53.
Reeve, J S, Cross, K C, Smith, R N and Oreskes, N, 1990. Olympic Dam
Copper-Uranium-Gold-Silver deposit, in Geology of the mineral
deposits of Australia and Papua New Guinea (Ed: F E Hughes), pp
1009-1035 (The Australasian Institute of Mining and Metallurgy:
Melbourne).
Sugden, T J, 1992. Structural geology of the Olympic Dam deposit,
WMC Resources K/3247.
Wingate, M T D, Campbell, I H, Compston, W and Gibson, G M, 1990.
Ion microprobe U-Pb ages for Neoproterozoic-basaltic magmatism in
south-central Australia and implications for the breakup of Rhodinia,
Precambrian Research, 87(3-4):135-159.
PACRIM 2004
Industry Challenges Globalisation, Communities,
Developing Countries, Financing Exploration

Exploration and Project Development Risk Reduction Approaches in the A H Osman 429
PacRim The Indonesian Experience

Industrial Minerals Opportunities and Making Money in AusAsia D Turvey 435

Exploration and Project Development Risk Reduction
Approaches in the PacRim The Indonesian Experience
A H Osman1
ABSTRACT
This paper will review the experiences of both big and small
exploration/mining companies (local and foreign investors), including the
impediments to exploration and project development in Indonesia since
the Contract of Work was first introduced there (1967 Minerals, 1981
Coal).
This is seen through the eyes of a person who has been employed in a
managerial capacity in large and small concerns, and more recently as an
independent consultant. Comment will be made on the political,
legislative and social constraints imposed by the presence of a large
group of disparate artisans (illegal miners) who now form a significant
component of the Indonesian mining industry; the Bre-X debacle; the
impact of the recent Regional Autonomy legislation; and changes to the
Mining, Forestry and Environmental Acts.
An overall view (including statistics) will be presented as to how
recent Indonesian developments have affected international confidence in
investing in its mineral sector over the past decade and it will be placed in
an international competitive context.
INTRODUCTION
Indonesia is a multicultural Islamic-majority country covering
some 17 000 islands of which some 6000 are inhabited. The
population is around 220 million and it is the worlds most
populous Islamic nation. Approximately half the population lives
on Java which is overcrowded. At Independence on 17 August
1945 the country inherited, inter alia, from the previous Dutch
colonial administration a European Roman-Dutch legal system,
ie Indonesia is a Civil-law jurisdiction which is dissimilar to
Australias Common-law system. The national Language is
Bahasa Indonesia which is a strongly developing, (similar to
Malay) language with many borrowings, including Arabic,
Hindi/Sanskrit, Dutch, Portuguese, English and Japanese. The
international commercial language is English.
According to the Finance Minister, Mr Budiono (May 2004),
the Indonesian economy was anticipated to show a 5.0 per cent
growth with 6.0 per cent inflation for 2004 with an expected
5.0-5.5 per cent growth and 5.0 per cent inflation for 2005.
The PricewaterhouseCoopers Indonesian Mining Survey 2003,
(IMA 2003/188) noted that investment on new mines and
capacity expansions, as well as replacement capital, remained a
fraction of the level of previous years. Indeed from 1996 to 1999,
spending on development and fixed assets averaged more than
US$1500 million per annum. The spending levels of the last two
years represents less than 20 per cent of these levels. Despite
Indonesias high prospectivity, spending on greenfields
exploration activity in the last two years was substantially lower
than it has been over the seven preceeding years. Greenfields
exploration expenditure is defined here as exploration spending
by exploration CoW companies. In 2002 it was only $7 million,
the same level as 2001. PricewaterhouseCoopers (2003)
comment that this expenditure effectively includes corporate and
administrative costs, meaning that field expenditure is even
lower. It does exclude however the spending of a few explorers
with advanced properties.
Currently Indonesias share of global exploration spending is
about one per cent, compared with almost three per cent three
years ago.
1. FAusIMM, Minconindo Pty Ltd, 24 Stanhope Road, Killara NSW
2071. E-mail: antonyosman@aol.com
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
HISTORY OF MINING IN INDONESIA AND ITS
PLACE IN THE WORLD TODAY
Indonesias mining history extends back to antiquity. The
country occurs in a regional tectonic setting consisting of an
extremely complex zone caused by the interaction or three
crustal lithospheric plates (Indo-Australia, Philippine and
Pacific) moving in different relative directions and impacting on
a fourth, the Eurasian plate. This has resulted in a geological
scenario suited for diverse mineral deposition. Tin operations
commenced in the 1700s and large-scale coal mining
commenced in the mid 1800s with most other larger mining
activities getting under way after Independence in 1945.
Van Bemmelen (1949) has provided an outline of the mineral
data base and history of exploration and mining development in
Indonesia to that point of time. Theo Van Leeuwen et al (1994)
have put together a superb document which covers the last
quarter century of mining and exploration activities in Indonesia.
The doctoral dissertation of Sigit (1996) draws together nicely
the geology and mining history of this country from the point of
view of the aspirations of the Indonesian people.
Singer (1995) noted the following concerning the mining
industry of Indonesia and its place in the world. The scenario
relates to just prior to the turn of the century:
seventh largest gold producer and first in the world in gold
discovered in porphyry copper systems;
19th in the world (first in South East Asia) in silver
discovered in Comstock epithermal vein systems and sixth in
the world (first in South East Asia) in silver discovered in
porphyry copper systems;
12th in the world in (first in South East Asia-just ahead of the
Philippines) in terms of total discovered copper but fifth in
the world in copper discovered in porphyry copper systems;
second largest tin producer; and
country possess around 15 per cent of world nickel reserves
and ranked fifth in world production.
Indonesia is currently around third in world for steam coal
exports. Bauxite, diamonds, base metals, industrial minerals are
also known.
Potter (1998) succinctly noted that the prospectivity of
Indonesia for more mineral discoveries related to the following
factors.
Sufficient land area to allow for the discovery of additional
world-class orebodies.
Possessing one of the longest Cenozoic vocano-plutonic arc
extents in the world, which results in an extremely favourable
and diverse geologic terrain.
Contains a magmatic environment that has proven production
coming from rocks of the most favourable composition and
age. Indonesias known producers are from igneous rocks
where the bulk of the worlds gold, silver and copper are
associated.
The country has a brief exploration history but already seven
world-class deposits can be identified viz four copper and/or
gold, two nickel and one skarn gold.
The country is under explored.
The geology database for Indonesia is still growing.
429
A H OSMAN
REGIONAL AUTONOMY
Under the system prevailing up to 1999 the revenue from the
provinces flowed through to the central government coffers in
Jakarta. From Jakarta a defined percentage of such revenues was
supposed to then be remitted back to the provinces. The flow
back was often far less than the defined amount. This
unsatisfactory situation was always a bone of contention
between the regional and central governments.
With the departure of long serving President Soeharto and the
arrival of President Habibie the opportunity was taken to redress
this situation with two major pieces of legislation viz Law No 22
of 1999 concerning Regional Government and Law No 25 of
1999 concerning the Financial Ratio between the Central and
Regional governments. With these pieces of legislation the
formerly centralised management of natural resources was
decentralised. Because of the above legislation and because Law
No 11 of 1967 (The General Mining Law) was seen to be no
longer in line with the development of science and technology,
accelerated access to information, trade development and
investment in the era of globalisation, then it should be amended.
A new mining law has been proposed but finally has been
returned by Parliament for redrafting.
STRUCTURE OF CURRENT MINING SYSTEM
(UNTIL ALTERED BY A NEW MINING LAW)
The current mineral development policy is founded, via the 1945
Constitution, on the principle that all minerals in the territory of
Indonesia in the form of natural resources are a part of the
nations wealth, are owned by the people and administered by the
Government on their behalf. The legal regulations are set out in
Law No 11 of 1967 (the general mining law) and Law No 1 of
1967 (Foreign Capital Investment).
Minerals have been divided into three categories viz:
A. Strategic Minerals eg petroleum, coal, uranium, nickel, tin.
B. Vital minerals eg iron, bauxite, copper, lead, zinc, gold,
diamonds. The mining of strategic minerals can, in
principle, be carried out only by the State. However, under
the concept of production sharing or Contract of Work,
foreign and domestic private companies can be involved in
the exploration for and development of these minerals.
C. Minerals not included in A or B, eg beach sands, industrial
minerals, etc.
Foreign companies are required to establish a PMA (foreign
capital investment) PT (limited liability) company and domestic
companies a PMDN (domestic capital investment) PT company.
Such companies act as contractors to the State and in the mineral
sector enter into a Contract of Work.
The current structure of the mining industry is as follows.
State mining companies
These include:
PT Aneka Tambang tbk the general mining company which
is now listed on the stock exchanges of Indonesia and
Australia (nickel, bauxite, gold, etc),
PT Tambang Timah, tin mining, and
PT Tambang Batubara Bukit Asam.
These companies may enter into joint ventures with other
foreign companies.
Contractor companies
These are the major foreign (PMA) companies most often with
an Indonesian partner in the shareholding structure. Domestic
contractor companies are PMDN. Generally a company holding
430 Adelaide, SA, 19 - 22 September 2004
now is up to 100 000 ha. A licence to mine is issued by the State
for 30 years after date of deemed start-up of production.
The very first mineral Contract of Work was issued to PT
Freeport (Irian Jaya/Papua) in around 1967 and was reissued
around 1997.
The very first coal Contractor company PT ARUTMIN
Indonesia was established on 31 October 1981 and the process of
incorporation and attaining full limited liability status achieved
some nine months later.
National private mining authorisation (KP)
holders
Kuasa Pertambangan (KP) are held by small private Indonesian
entities or persons. KPs are not for foreign investment purposes.
Each KP is around 2500 ha with a total holding per
person/company at 10 000 ha Reserves/Resources within each
holding restricted.
At the bottom end of the mining scale in Indonesia are found
the KUD (village level cooperatives) and family mining
operations. The former operate with a KP and the latter operate
with some form of KPPRakyat (peoples mining authorisations).
Both are small area tenements which are now granted at
Kabupatan/District level.
Kooperasi Unit Desa (KUD)
These are Village Level Cooperative Units of restricted size and
resource base.
Pertambangan Rakyat
Peoples Mining tenements for families, local fossickers.
Artisanal mining.
Illegal mining/PETI/informal mining
Peti (Pertambangun tanpa Izin = mining without licence, ie
illegal mining).
In the past gold (mostly alluvial), diamonds (alluvial) and tin
were the main commodities of interest. Now coal has become the
major commodity for illegal mining. A myriad of small
operations with a combined output of somewhere between
6 - 9 million tpa (a good sized mine). The local persons involved
in illegal mining operations are often supported financially by
other outside organisations.
The relationship between the various levels of mining
organisations/companies and other aspects of their operations is
shown in Table 1, Osman (1998).
BRE-X DEBACLE
Bre-X was a small Canadian company which took interests, in
company with an Indonesian partner, in gold exploration KPs
and CoWs in East Kalimantan. Over a period of time in the
mid-late 1990s Bre-X announced increasingly optimistic gold
resource grade/tonnage discoveries. This resulted in local and
world-wide attention, a soaring share price and increased interest
(to fever pitch) by other parties in taking up equity positions in
the Bre-X concessions and the general geological/geographic
area. At this point in time when interest was at its height the
mining world was rocked by the revelations that check drilling,
by an independent party, had failed to confirm previous grade
levels combined with the mysterious decease of the chief
geologist in an apparent fall from a helicopter. In the event it was
seen by many as an elaborate scam.
The Bre-X debacle embarrassed whole layers of Indonesian
society and the Government. Many aspects of the companys
operations (as reported to, and as seen by, an outsider) are clearly
PACRIM 2004
 EXPLORATION AND PROJECT DEVELOPMENT RISK REDUCTION APPROACHES IN THE PACRIM
TABLE 1
Relationship between size of mining enterprise and other aspects of mining in Indonesia.
Size Status Funds
Large, medium PMA Foreign CoW, CCoW Large
Medium PMDN Moderate
Small KP, CoW CCoW Small
KUD, Pert Co-ops Little
Rakyat, Artisanal Families, Groups Nil
Illegal/non-formal miners Illegal Nil
Medium- to small-scale national mining companies may seek a larger foreign or PMA (foreign investment) or national PMDN (domestic investment)
company to joint venture their concession Osman (1998).
indicative of how not to manage an exploration company and its
operations in a country like Indonesia, and especially so when a
major discovery is believed to have occurred. It is hard to
separate fact from fiction in this case but the following points are
worthy of particular comment:
1. Tight control of management, administration and technical
operations at all levels is a prime necessity with the
president director (the chief operating officer or his/her
equivalent) of the company residing in-country. The
management and administration of a newly created
company and its exploration program is a very
time-consuming matter. A low key approach to
announcements of discoveries is eminently desirable.
2. Tight control of staff (especially expatriates), their
movements and their activities is also a prime necessity.
3. An understanding of the culture of the country is again a
prime necessity. As an example an Indonesian colleague
once told this writer two decades ago that when establishing
say, the office workforce then the different ethnic/linguistic
groups should be balanced so that no one group can gang
up on others. In the Bre-X case a foreign speaking (Tagalog)
ethnic group was allowed to have complete control of a most
important part of the operations (viz sampling preparation
and dispatch) with disastrous results.
4. Foreign companies generally operate as a shareholder of a
PMA (foreign capital investment) PT (limited liability)
company under the terms and conditions of a Contract of
Work. It is more than useful to understand what is, and
what is not, allowed and required. The company is a
contractor to the Government or one of its agencies. The
PMA CoW company does not own the commodity being
mined. The CoW documents as well require that the
company instruct its staff and its families, as well as those
of sub-contractors, in the laws and customs of Indonesia.
IMPEDIMENTS TO EXPLORATION AND MINING IN
INDONESIA
The current system of Regional Autonomy along with a
perceived set of hostile Forestry legislation are seen as the major
impediments to exploration and mining activities by foreign
companies in Indonesia at the present time.
Before 1999 the Indonesian Government had issued more than
150 licences to domestic and foreign capital mining companies
to operate in designated areas. At that time the concessions were
granted in areas designated non-forest or classified production
forest in which open pit mining was allowed. With the change of
government and political system Law No 41 of 1999 on Forestry
was issued. Simultaneously many parts of the above mentioned
concessions were declared protection forests in which no open
pit mining was allowed.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Regulatory requirements Duty of care to the community
Numerous, stringent Very high
Numerous, less stringent High-Moderate
In effect
Some, relaxed Restricted-Nil
Nil Nil-Negative
This matter remains not entirely resolved to this day.
The Jakarta Post website news for 19 May 2004 made the
following comment concerning a disputed, non-mining, court
ruling in Sumatra viz:
Analysts say Indonesias corrupt and ineffective
judicial system has emerged as the principal
deterrent to overseas investment in this Country.
INDONESIAS PLACE IN WORLD INVESTMENT
ATTRACTIVENESS
The Fraser Institute of Canada produces an Annual Survey of
Mining Companies and in this survey rates the policy
attractiveness and mineral attractiveness of mining jurisdictions
around the world. From the results of these surveys the Institute
creates several indices, viz:
1. The Mineral Potential Index (MPI). This index rates a
country or its states/provinces according to its
attractiveness for new investment based on its geology.
2. The Policy Potential Index (PPI). This is a composite index
that measures the effects of government policies, eg
regulatory, land use, etc in attracting new Investment.
3. Investment Attractiveness Index (IAI). This is an overall
index which accounts for both the aforementioned factors.
To arrive at this overall index the mineral Policy Potential
factors were weighted 40 per cent and Mineral Potential
60 per cent.
The results for selected countries are illustrated in Table 2,
showing Investment Attractiveness Index (IAI) trends for the
years 2003/2004, 2002/2003 and 2001/2002. The Mineral
Potential Index (MPI) is shown for selected countries for
comparison with Indonesia. The Policy Potential Index (PPI) for
all the selected individual countries is not tabulated here but
Chile, No 1 ranking country on the IAI, with an MPI of 92, has a
PPI of 85. Indonesia No16 ranking country on the IAI has a MPI
of 91 and a PPI (by calculation) of 24. Indonesia would need a
PPI of around 93 to equal Chile on the IAI rating scale.
Chile is seen to be on the way up for mineral investment but
Indonesia is seen to be stalled.
Indonesia (91) is in the top six for perceived mineral potential
attractiveness and only five points behind Chile (96). Given that
the area of archiapelagic Indonesia is 1 919 440 km2 some two
and a half times the area of Chile (756 950 km2) it is interesting
to speculate, given Potters (1998) opinion that Indonesia is
under explored whether Chile is perceived as more explored,
because of its lesser area, or less explored than Indonesia.
Speculating further if the perception is that Indonesia is less
explored than Chile, then perhaps with increasing exploration the
MPI for Indonesia would increase (or decrease!).
431
A H OSMAN

TABLE 2
Indonesia investment attractiveness in the world context. Selected countries with data, courtesy Fraser Institute, Canada (2004).
Mineral Potential Index (MPI) Investment Attractiveness Index (IAI)
Country/Jurisdiction Score Rank Trend
03/04 02/03 01/02 03/04 02/03 01/02
(/100) (/53) (/47) (/45)
Russia 100 74 63 59 9 14 16 Up
Western Australia (PPI 73) 98 88 * * 3 * * n/a
Peru 96 82 84 73 5 5 8 Up
Chile (PPI 85) 96 92 94 86 1 1 4 Up
Nevada, USA (PPI 89) 92 91 86 83 2 4 6 Up
Indonesia (PPI 24) 91 64 49 52 16 24 20 -

Quebec, Canada (PPI 80) 89 85 90 90 4 2 1 Down

Brazil 85 82 83 86 5 6 4 -
South Africa 81 66 64 61 13 12 12 n/a
Northern Territory 77 76 * * 8 * * n/a
BC, Canada (PPI 30) 70 54 54 60 26 20 15 Down
China 70 62 64 58 17 12 18 Up

Philippines (PPI 22) 64 47 32 25 32 39 41 -

Kazakhstan 62 52 49 39 9 14 16 Up
Queensland 62 69 * * 12 * * n/a
NSW 55 66 * * 13 * * n/a
India 45 44 24 * 37 43 * Up
South Australia 43 59 * * 19 * * n/a
Tasmania 32 52 * * 26 * * n/a
New Brunswick, Canada (PPI 73) 32 48 46 37 30 29 34 Up

Victoria 28 46 8 * 34 * * n/a
Colorado, USA 21 24 44 33 49 32 38 Down
Turkey 21 35 * * 41 8 8 n/a
California, USA 9 12 36 35 52 38 35 Down
New Zealand (PPI 71) 8 27 22 * 47 45 * -
Ireland 8 33 * * 42 * * Up
Washington, USA 2 11 22 15 53 45 45 -

IAI = PPI ( 0.4) + MPI ( 0.6), eg:

Chile 92 = 85 ( 0.4) + 96 ( 0.6)
Indonesia 64 = 24 ( 0.4) + 91 ( 0.6)
NB: Nevada had the top PPI perception for the Seventh Annual Survey

RECENT DEVELOPMENTS companies. This of course clearly reflected the inability of

This writer is indebted to Mr Noke Kiroyan (former President
Director of PT Kaltim Prima Coal and PT Rio Tinto Indonesia)
and a Board Member of the Indonesian Mining Association
(IMA) for the substance of the following comments (Kiroyan
2004) occurring in an excellent article in the April issue of the
Jakarta based IABC News. These comments are those of an
Indonesian executive intimately involved and experienced with
the mining industry and were prepared in March 2004. I quote
extensively here from this article.
Over the past several years to date and with the exception of a
few of the largest international mining companies there has been
virtually no mineral exploration taking place in Indonesia.. As a
flow on from this the membership of IMA had declined from
over a hundred to around a dozen or so. It was noted that active
mining operations are being carried out by only a handful of
Indonesia to generate interest among foreign investors.
(Small-scale mining development involving pegging of KPs by
local companies have been particularly active in the coal sector).
Towards the end of 2003 steps were commenced to try to do
something about this moribund investment situation. On 30
December there was a meeting between the Executive Board of
IMA and the Minister of Energy and Mineral Resources, H E Dr
Purnomo Yusgiantoro. Dr Purnomo requested that
representatives of the mining industry prepare a Road Map for
the revival of the Indonesian mining industry. The industry
moved quickly and on 2 January 2004 a meeting involving the
newly appointed Director General for Geology and Mineral
Resources (Dr Simon Sembiring) and other senior department
officials, industry representatives and PERHAPI (The
Association of Indonesian Mining Professionals) prepared the
groundwork for a proposal to resurrect the mining industry. The

432 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 EXPLORATION AND PROJECT DEVELOPMENT RISK REDUCTION APPROACHES IN THE PACRIM
opinion was noted that the industry had been one of the
mainstays of the Indonesian economy for many years and the
envy of many other developing countries prior to the South East
Asian financial crisis.
Intensive meetings involving the same participants took place
on a regular basis afterwards in the new year.
The responsibility for preparing the road map was given by the
Government to the Faculty of Economics at the University of
Indonesia (LPEM-FEUI) and was still in progress at time of
writing and thus the final conclusions were not then available.
However the problems afflicting the mining industry and
identified by LPEM and forming the basis upon which the
research was conducted are as follows:
1. absence of exploration activities;
2. low index of confidence within the industry;
3. steadily deteriorating business climate following the
financial crisis in 1997;
4. extremely high entry cost;
5. Indonesias high ranking in the Corruption Perception
Index;
6. excessive bureaucracy; and
7. ambiguity about oversight powers over the mining industry.
The PricewaterhouseCoopers Indonesian Mining Survey 2003
Report added the specifics of the need for improvement in:
the competitiveness of the tax and royalty system;
resolution of conflicts between the CoW and the Forestry
Law;
restoration of the long term certainty in the CoW system;
reduction in illegal mining; and
level of fairness in divestment of foreign interests.
In May, LPEM was in the final stages of completing the study
of possible solutions for the individual problems outlined above,
well known by the mining industry but seemingly ignored by the
Government for years! It has also been recognised and the point
clearly made, that political will on the part of the Indonesian
government is a major prerequisite for the revival of the
Indonesian mining industry.
With reference to the afore mentioned meetings Mr Kiroyan is
of the opinion that the mining industry has been reassured that
the government is aware of the need to reaffirm its commitment
to the industry and is willing to take the necessary measures in
concert with other government agencies. A case in point is the
necessity for close cooperation with the Department of Finance
to remove the current fiscal disincentives to investment, and
amending existing legislation. The new Parliament will only be
able to consider such changes in 2005.
The former draft mining law which had been submitted to the
State Secretarys office several years ago has been withdrawn
and a task force consisting of officers of the Department of
Energy and Mineral Resources is now required to draft a new law
incorporating the conclusions and recommendations of LPEM.
An informed source believes it will be the middle of 2005 before
Parliament itself can review the new draft.
In the meantime, to accommodate companies that are prepared
to continue exploration in the anticipation of ultimately applying
for a CoW (or whatever name is given to an entity of similar
purpose under the new mining law), an interim measure is being
sought that recognises the current situation under Regional
Autonomy laws and covers the period of time until a new mining
law has been promulgated. The Directorate General for Geology
and Mineral Resources along with IMA, PERHAPI and APBI
(Indonesian Coal Mining Association) are working together on
such an interim solution.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Mr Kiroyan concluded his article with the opinion that a
Ministerial Decree (KepMen) or a specific Government
Regulation would most likely be the legal instrument that
provides a suitable framework for Contact of Work situations in
this interim period.
WHAT OF THE FUTURE?
The future of mining in the near term is to a great degree
dependent on the composition and attitude of the incoming
parliament and incoming President. (It is worthwhile here
recalling that when President Soeharto came to power in the
mid-sixties he inherited a country in economic disarray. Under
his guidance in 1967 the first law passed by Parliament was the
Foreign Investment Law (1/1967) to be shortly followed by the
General Mining Law (11/1967). This enabled the revitalisation of
the petroleum and mining industries to be accomplished).
On Monday 5 April 2004 the Indonesian voters went to the
polls to vote on their new parliament make-up with 24 registered
parties to choose from. In a nation of some 220 million people
around 148 million were registered to vote. At the declaration of
the poll 124.4 million votes had been cast of which 10.9 million
were invalid. GOLKAR the party of previous President Soeharto
received 21.58 per cent of the votes, PDI-P, the party of
incumbent President Megawati Soekarnoputri received 18.53 per
cent of the votes (massively down from the 34 per cent received
in the 1999 elections) and with PKB-The National Awakening
Party in third place with 10.57 per cent of the votes cast (Jakarta
Post, 6 May 2004).
No one party received a clear majority. In the 5 July 2004
Presidential elections no one person received the clear majority
required, and thus in September the nation went to the polls
again to elect a new President.
CONCLUSIONS RELATING TO RISK REDUCTION
Indonesia is a mineral rich country with much to offer the
foreign investor when certain elements of the current system
have been improved.
Know the country, its political history, its peoples, its
languages, its cultures, legal system (civil law jurisdiction
not common law jurisdiction), business systems. Be aware of
Indonesias strengths as well as its weaknesses.
Know the countrys geology, mining history and relevant
legislation.
Take time to review your companys needs, and be especially
careful in selecting and getting to know your Indonesian
partner before committing. This is going to be a long-term
commitment. Verbal understandings and trust play a great
part in Indonesian business operations.
Retain the best and most experienced law and notarial firms
in Indonesia to draft your agreements and your articles of
association. Even though the written word does not have the
same practical strength in Indonesia as elsewhere it is best to
have in writing everything your company wants to happen.
Having a legal document silent on important matters is not
a good situation.
Select appropriate and mature staff for company operations.
Be prepared to lead and instruct in a culturally appropriate
fashion.
Understand and be prepared to spend a great deal of time
managing and administering the company and its partners.
Make certain offshore Head Office staff aware that operating
in Indonesia is not the same as operating back home.
Remember the three Ps viz: Patience, Persistence and
Politeness.
Be prepared for the long haul, or dont bother!
433
A H OSMAN
REFERENCES
Anon/Fraser Institute, 2004. Seventh Annual Survey.
Anon /Fraser Institute, 2004. Geology and Politics determine mineral
investment climate, Mining Engineering, 56(3):28-30.
IABC News, Jakarta, April 2004.
Kiroyan, N, 2004. A ray of hope for the Indonesian mining industry,
IABC News, Jakarta, April 1, 4.
Osman, A H, 1998. Balancing the social and cultural impacts of mining
projects, in Proceedings Mining Indonesia Conference 1998,
(Indonesian Mining Association: Jakarta).
Potter, D R, 1998. Elephant hunting in Indonesia the search for the next
generation of Indonesian mineral deposits, in Proceedings Mining
Indonesia Conference 1998, (Indonesian Mining Association:
Jakarta).
434 Adelaide, SA, 19 - 22 September 2004
PricewaterhouseCoopers, 2003. Indonesian mining survey, in Indonesian
Mining Association Newsletter 2003/18.
Sigit, S, 1996. Indonesias mineral potential and the awakening of its
mining industry, 66 p (Scientific oration presented before the
Bandung Institute of Technology open senate meeting held at the
bestowal of title of Doctor Honoris Causa in engineering science,
9 March) (Indonesian Mining Association with the contributions of
P T Adaro Indonesia, P T INCO Indonesia and P T Ingold
Management, Jakarta. Translated from the original Indonesian text.)
Singer, D A, 1995. World class base and precious metal deposits A
quantitative analysis, Economic Geology, 90:88-104.
Van Bemmelen, R W, 1949. The geology of Indonesia, Vol 2, Economic
Geology (Government Printing Office: The Hague, Netherlands).
Van Leeuwen, T M, Hedenquist, L P and Dow, J A S (Eds), 1994.
Indonesian Ore Deposits Discoveries of the past 25 years, J
Geochem Explor, 50:142.
PACRIM 2004
Industrial Minerals Opportunities and Making Money in
AusAsia
D Turvey1
ABSTRACT
This paper provides an expanded definition of industrial minerals that
challenges the way we look at exploration, development and the
understanding of mineral industry structure and profitability. Included in
industrial minerals are bulk commodities such as iron ore and coal
alongside the more traditional industrial minerals such as salt, soda ash,
silica sand and clay.
The profitability of industrial minerals is reviewed from a producers
perspective and shown in many instances to be significantly higher than
for precious and base metal industries. The key drivers of profitability are
examined along with some highly profitable industrial mineral industries
such as borate, potash, chromite and natural soda ash.
Demand for industrial minerals in Australia and Asia is reviewed, with
a focus on supply bottlenecks and the opportunity this represents for
investment by Australian companies. The establishment of new price
benchmarks is considered and the possibility of new export grades for
industrial mineral products.
Investment opportunities in industrial mineral projects are outlined,
including some exploration concepts in Australia and Asia. Industrial
mineral project evaluation methods are discussed and the challenge
issued to companies (exploration, production and finance) to become a
part of the profitable action in the world of industrial minerals.
DEFINITION OF INDUSTRIAL MINERALS
Industrial minerals are defined as mineral products that are
mined, processed and traded for use in manufacturing and
industrial applications. Any mineral that is mined and not
changed in chemical composition by processing prior to its use
in an industrial application is an industrial mineral.
This definition distinguishes industrial mineral products from
value added mineral products such as alumina, magnesia or
titanium dioxide (essentially industrial chemicals) and also metal
alloys and metals. In the latter cases the mineral composition has
been altered by processing rather than just concentrated.
Minerals such as iron ore (hematite, magnetite and/or
limonite), manganese ore, chromite, coal and bauxite are
considered to be industrial minerals alongside the more
traditional industrial minerals such as salt, gypsum, soda ash,
potash, phosphate rock, sulfur, borates, silica sand, feldspar, clay
(bentonite, kaolin, ball clay), mica, wollastonite, etc.
Base metal concentrates, especially if they are predominantly a
single mineral species such as a galena lead concentrate, are
essentially industrial minerals.
Unlike precious metals, the evaluation and development
methods for industrial mineral projects include consideration of
market application rather than just price, especially if the mineral
is unaltered by the later industrial process, eg kaolin as filler or
coating mineral for paper.
Market price is not only based on product purity (contained
metal, element or oxide) and contaminant elements but often on
physical properties of the crude or processed industrial mineral,
eg fluidity in coking coal, decrepitation properties in magnesite,
aspect ratios in wollastonite or talc.
Many people consider industrial minerals to be low unit value
commodities. Indeed silica sand, gypsum and salt are low unit
1. MAusIMM, Equant Resources Pty Ltd, 3 Stannington Avenue,
Heathpool SA 5068. E-mail: dturvey@senet.com.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
value minerals ($US 15 - 20/t fob) and their commercial viability
is significantly affected by the cost and availability of
transportation.
However, industrial minerals also include high unit value
minerals such as potash ($US 120/t fob), borates ($US 250 - 350/t
fob) and zircon ($US 350 - 450/t fob).
MARKET VALUE AND PROFITABILITY
The market value (industry size) and profitability of most well
known industrial minerals is summarised for the 1990s in Table
1.
The market value of industrial minerals is calculated based on
global annual ore production (tonnes) at a weighted average
production grade and $US/mtu fob price. Profitability for ore
producers across the global industry (EBIT/Sales) considers
issues of industry structure and integration, domestic and export
subsidy, and transfer pricing. The 1990s are considered to be a
representative time slice that includes periods of low and high
commodity demand and price, covering the boom and bust of
Asian economies and the emergence of the Chinese industrial
economy.
The market value and profitability of industrial minerals is
compared with that of gold, copper and zinc industries, where
their market value is calculated from an estimate of ore value ex
mine and/or physical beneficiation plant.
People are often surprised at the large size of industrial
mineral markets and even more surprised by their profitability.
For example, the fertilizer mineral industries (phosphate, potash
and sulfur), salt and soda ash industries have an annual market
value >$US 3 billion. This compares with zinc ore production
~$US 2 billion and gold ore production ~$US 5 billion.
Further, the profitability of industrial minerals in many
instances is significantly higher than for precious and base metal
industries, though this fact is poorly understood and lacks
recognition in Australia. Several industrial minerals are highly
profitable over long periods and economic cycles, including
borates, potash, natural soda ash, iron ore, chromite, titanium
minerals and zircon where EBIT/Sales margins are >25 per cent.
This compares to lower profit margins for gold and base metal
industries of <10 per cent during the 1990s.
The standout profit performers of the industrial minerals are
borates (32 per cent EBIT/Sales), potash (28 per cent EBIT/Sales)
and iron ore (26 per cent EBIT/Sales).
The key drivers of profitability for industrial minerals include:
1. Limited sources of supply. This leads to price leverage in
negotiations by individual producer companies or collective
industry parties.
2. Limited substitution.
3. Limited vertical integration between ore producer and end
use industry, ie producing and marketing minerals, not
manufacturing metals or industrial products.
4. Consistent quality and product differentiation.
5. Low production costs and transport costs.
6. Effective dealing with the right partners, both owners and
customers.
435
D TURVEY

TABLE 1
Industrial minerals market value and profitability, 1990s.

Industrial mineral Market Profitability of ore producers

Ore production Ore production Mtpy @ Ore traded $US billion EBIT/sales per cent
$US billion $US/mtu fob
Phosphate rock 4.3 143 Mt @ $30 1.4 20
Potash 3.3 25 Mt @ $132 2.2 28
Sulfur 2.8 55 Mt @ $50 1.0 12
Salt 3.3 185 Mt @ $18 1.0 20
Gypsum 2.1 103 Mt @ $20 0.5 18
Borates 1.0 2.8 Mt @ $350 0.6 32
Soda ash 3.2 32 Mt @ $100 0.5 25
Silica sand 2.2 180 Mt @ $12 0.5 15
Kaolin 2.8 28 Mt @ $100 1.0 16
Ca carbonate# 1.0 10 Mt @ $100 0.2 22

Iron ore 19 950 Mt @ $20 8.4 26

Manganese 2 20 Mt @ $100 0.4 23
Chromite 0.8 10 Mt @ $80 0.2 25
Coking coal 20 500 Mt @ $40 7.0 18
Bauxite 3 120 Mt @ $25 0.8 23*
Magnesite 0.8 16 Mt @ $30 0.2 20
Ti minerals 1 2.8 Mt @ $180 0.2 25
Sources: Company reports, analysis from financial institutions, primary research and estimates
In comparison+
Gold 5 72 Mt @ $70/oz N/A <10
Copper 5 500 Mt @ $10 N/A <15
Zinc 2 80 Mt @ $25 N/A <10

Notes:
Profitability of 'free market' producers (unsubsidised), especially listed public companies.
Natural soda ash producers.
# Calcium carbonate for paper industry.
* Includes transfer pricing from integrated alumina and aluminium producers.
+ Market value is based on estimate of ore value ex mine and/or physical beneficiation plant.

A key feature of industrial mineral profitability is greater
stability and less volatile price cycles due to their relationship
with global industrial growth, rather than supply driven markets
and price cycles. This compares with metal markets, which have
experienced over supply during the 1990s from many producers
with significant surplus production capacity. As such, metal
prices and profitability have been low during this period.
However, high profitability cycles occur for both industrial
minerals and metals during times of high demand such as years
2003 and 2004, related to high economic growth in China and
India and general economic recovery in developed economies.
We have all seen the frantic markets in iron ore and coal
during the last 12 - 18 months with prices reaching record levels
and new benchmark prices being established. The same market
demand is evident for the less glamorous industrial minerals such
as salt, soda ash and potash. The profitability of companies in
these industrial mineral sectors will correspondingly be high for
several years.
By understanding the structure of an industrial mineral
industry and optimising your competitive advantages, then
long-term, high margin and stable profitability is possible. For
example take RIO TINTO, who make significant profits from
industrial minerals such as borate, talc, salt, ilmenite and zircon,
iron ore and coal at high profit margins.
Indeed, borates are arguably the worlds most profitable
mineral industry.
This compares to RIOs gold and copper investments that
made lower profit margins during the 1990s. Further, RIO does
not make investments in end-use industries such as ceramics and
fibreglass (borates), paper and plastic (talc), chloro-alkali and
plastics (salt), pigments (ilmenite and zircon), and steel (iron ore
and coking coal).
GROWTH AND DEMAND
Demand for industrial minerals is related to economic growth
and investment in manufacturing and infrastructure sectors.
Strong industrial growth in Asia during the past two decades has
seen high demand for industrial minerals and significant
increases in domestic production, imports and prices. However,
infrastructure bottlenecks and lengthy lead-time for evaluation
and development of mineral projects has resulted in supply
imbalances in many minerals.
These supply imbalances have lead to investment opportunities
in exploration and development of industrial minerals, especially
in developed countries that have pre-existing infrastructure,
technical expertise and financial backing. In Australia we are

436 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 INDUSTRIAL MINERALS OPPORTUNITIES AND MAKING MONEY IN AUSASIA

experiencing significant investment activity in iron ore, chromite,
manganese and salt, while in Indonesia and China there
continues to be rationalisation and investment in the coal
industry and many other industrial minerals.
Increased development and trade of industrial minerals in
Australia is forecast in the short term (two to five years) due to
strong regional economic growth. However, in the medium to
long term (three to ten years), industrial mineral deposits located
in countries with the greatest demand will require to be
developed, as transport costs and trade imbalances will adversely
effect industry profitability. This represents an excellent
opportunity for Australian geoscience professionals, and
exploration and mining companies to participate in development
of larger, commercially sound industrial mineral industries in
countries such as India, China and Indonesia.
The type and grade of traded industrial mineral products can
change quite rapidly due to new technologies (mining,
processing and/or end use application) and market structure
(trade imbalances, cost structure, industry profitability and
ownership). A good example of such changes has occurred in the
phosphate industry, where historically (1950s - 1980s) the
international trade was dominated by rock phosphate, whereas
now the preferred traded products are phosphoric acid and
Di-Ammonium Phosphate (DAP). These changes were driven by
several factors:
1. demand and trend towards the use of high analysis
fertilizers, especially in developed countries with large
agricultural sectors;
2. better economics of value added products relative to
transportation costs; and
3. environmental issues in reducing wastes at fertilizer
manufacturing facilities (often located in heavily populated
industrial centres in developed countries) and keeping
wastes at or near the phosphate rock mine site (often
located in developing countries with lower populations).
However, this value added trend can also work in reverse,
where trade of mineral products increases. This is commonly
seen where higher price benchmarks are established (currently
the case for iron ore and coal), thereby providing the opportunity
for production and trade of new export grades of industrial
mineral products. These new products are often lower grade or
different physical properties (eg sizing) and they are used as
blended feedstock by the end user relative to existing or new
manufacturing technologies.
For example, currently in China the steel industry is blending
local siliceous limonite ores (~23 per cent Fe), local magnetite
fines (~50 per cent Fe, <1.0 mm), local magnetite pellets
(~68 per cent Fe), with imported lump and fine hematite ores
(60 - 64 per cent Fe). The average blast furnace feedstock grade
is <50 per cent Fe, especially in the northern and central
provinces of China. Further, steel and ferroalloy producers in
China are using feedstock of mixed heavy mineral concentrates
for electric arc furnaces (Ti, Fe, Cr including titano-magnetite
concentrates) and direct shipping mineral ores are common,
especially base metal oxide or silicate ores and specialty metal
ores (Be, Ta, Sn, Nb, In, etc).
CONCEPTS AND INVESTMENT OPPORTUNITIES
Exploration concepts and development projects for industrial
minerals in Australia and Asia are summarised in Table 2.

TABLE 2
Industrial minerals concepts and investment opportunities.

Industrial mineral Concept and/or project Location Investment opportunity

Phosphate rock 1. High-grade carbonatite Eppawala, Sri Lanka Exploration, R&D

2. Under utilised sedimentary phosphate assets Lao Cai, Vietnam M&A recapitalisation
Potash 1. World-class, high-grade, and well defined resources Udon Thani, Thailand M&A development

2. Good geology and logistics Danakil, Ethiopia Exploration

Salt 1. Solar salt fields NW Western Australia M&A recapitalisation
Gypsum 1. Develop agriculture and road construction products WMCF operations R&D
Borates 1. Under utilised assets Pacific South America M&A recapitalisation

2. Known deposits Central Iran and NW China Exploration

Soda ash 1. Sodium bicarbonate brines Roma district, Qld R&D evaluation

2. Trona deposits Southern Queensland Exploration

Silica sand 1. Export glass sand to Asia and develop specialty sands Nth Qld and Bangka Indonesia M&A recapitalisation
Kaolin 1. Evaluation and development of value added products Belitung, Indonesia M&A recapitalisation
Iron ore 1. Titano-magnetite sands NZ and Indonesia M&A recapitalisation

2. Fe-W-B skarn and exhalite Malaysia, NE China Exploration R&D

Manganese 1. Good exploration potential and logistics Northern Australia and Eastern Exploration and evaluation
Indonesia
Chromite 1. Current operations Southern Iran M&A recapitalisation

2. Known deposits Philippines, Indonesia Exploration R&D

Bauxite 1. Known deposits Indonesia, Saudi Arabia, Northern Exploration and evaluation
Australia
Ti minerals 1. Rutile in acid volcanic and Na altered granites Malaysia, Thailand and SE China Exploration

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 437

D TURVEY
The occurrence and development potential for industrial
minerals in Australia and Asia relates to regional geology, local
geology and weathering processes. As general rules, China is
relatively deficient in high-grade ferrous ores (Fe, Mn, Cr from
Archaean and Proterozoic deposits), whereas Japan and
Philippines are deficient in silica sand and potash (calc-alkaline
volcanic island arc environments). Australia is deficient in
potash, natural soda ash and borate deposits, mainly due to a lack
of recent volcanism and recent intracratonic sedimentary basis.
Several industrial mineral exploration and development
projects warrant detailed description as follows.
Soda ash
Bi-carbonate brines, Roma area, Queensland
In the 1970s, unique bi-carbonate rich subsurface brines were
discovered during oil and gas exploration in the northern Surat
Basin near Roma, southeast Queensland. These brines contain
high concentrations of bi-carbonate (~14 g/litre) and very low
concentrations on sodium chloride (~1 g/litre) and alkali sulfate
species. The brines were likely formed by dissolution of trona
(Na2CO3. NaHCO3.2H2O) or nahcolite (NaHCO3) deposits with
preservation in a sealed aquifer. Precedent brine operations
include Dow Chemicals magnesia operations in Michigan State.
Conventional solar evaporation and fractional crystallisation
technology could produce high purity sodium bi-carbonate for
use in food, glass, mineral processing (aluminium) and chemical
industries. The commercial potential is high as Penrice is a
monopoly supplier to Australian markets with strong growth in
Asian markets. Sodium bi-carbonate is a high value and high
margin mineral product (price ~$US 300/t fob).
Trona deposits, southern Queensland
Good exploration potential for trona (Na2CO3. NaHCO3.2H2O)
or nahcolite (NaHCO3) deposits exists in the northern Surat
Basin, Queensland. Precedent exploration model is the Green
River basin Wyoming USA, which contains the worlds premier
trona deposits. Natural soda ash producers are highly profitable
(eg FMC reported an average 28 per cent EBIT/Sales margin
during 1990 - 1996, EBIT ~$US 150 million).
Iron ore
Titano-magnetite sands, New Zealand and Indonesia
Titano-magnetite sands are being mined from coastal deposits in
northwest New Zealand (2.4 Mtpy @ ~58 per cent Fe and ~7 per
cent Ti). These titano-magnetite concentrates are used as iron ore
feedstock by BHP Steel and also exported directly to Japan.
Large resources of titano-magnetite sands occur in coastal
deposits in Indonesia, especially on the north coast of Irian Jaya.
Similar iron sands are currently mined by PT Aneka Tambang in
Java with target sales of ~140 000 tpy to Chinese steel customers.
Newly developed mineral sand processing technology
(including Austpacs ERMs process) may be useful in
beneficiation of titano-magnetite sands; pelletising may be done
in China (surplus capacity in northern China).
IronTungstenBoron skarn and exhalite deposits,
Malaysia and NE China
Large iron (magnetite) skarn deposits and bedded
magnetite-hematite exhalite deposits (low base metal contents)
occur in Palaeozoic rocks in northeast Peninsular Malaysia and
northeast China. The Japanese mined these deposits during
1930s-40s as feedstock for steel manufacture. The deposits
438 Adelaide, SA, 19 - 22 September 2004
warrant exploration and evaluation using modern methods and
technologies. Infrastructure and logistics are excellent.
Conventional mining and processing technology could produce a
magnetite concentrate (~65 per cent Fe) that could be shipped for
pelletising or direct feedstock to Chinese steel producers.
Potash
Sylvinite resources, Udon Thani, northern Thailand
Large, high-grade potash deposits (sylvinite) were discovered in
northern Thailand in early 1980s. Detailed exploration and
evaluation during 1994 - 2003 has confirmed a commercially
viable project, supported by strong market demand in Asia and
competitive advantages in freight.
Mining permits are in progress the project is world-class.
Danakil area, Eritrea and Ethiopia
Sylvinite mineralisation was discovered during oil exploration in
1970s. The deposits warrant exploration and evaluation using
modern technologies. Infrastructure and logistics are good.
Manganese
Sedimentary deposits, northern Australia and
eastern Indonesia
Sedimentary manganese deposits were discovered during 1970s
in the Wellesley Island area and Borroloola areas in Northern
Australia. Manganese occurrences are also recorded on
Sumbawa, Flores, Timor and Sumba Islands in Indonesia. The
deposits warrant exploration and evaluation using modern
technologies. Infrastructure and logistics are good.
Many mineral deposits in Asia are poorly explored and
operations are often significantly undercapitalised, certainly not
optimised. Many Asian industrial companies are cashed-up and
keen to invest in mineral evaluation and development projects in
Australia. This represents an excellent opportunity for Australian
geoscience professionals, and exploration and mining companies
to participate in mineral development projects in countries such
as India, China and Indonesia.
EVALUATION METHODS AND CHALLENGE
Exploration and project evaluation methods for industrial
minerals must recognise the interplay between technical
(resource, mining and processing), market requirements and
structure, and commercial parameters. Infrastructure and
transport logistics must be considered in commercial evaluation,
as many industrial minerals are bulk, low unit value
commodities.
Published mineral product specifications and prices should
only be used as a guideline because significant changes in market
structure (end use industry) may result in new mineral product
specifications. Market and product development work, in
conjunction with key customers, is generally required and often
successful in business development of new industrial mineral
products.
Due to Australias world recognised mining industry, there
exists immediate opportunities for investment by exploration and
mining companies and/or for geoscientists to provide consulting
services.
Australian geoscience professionals and mining companies
have much to offer Asia we are well respected and are well
liked. We see many adventurous souls investing in exploration
and development of gold and base metal projects in Australia and
Asia, especially when the market is bullish, indeed we are metal
PACRIM 2004
 INDUSTRIAL MINERALS OPPORTUNITIES AND MAKING MONEY IN AUSASIA

centric. The best examples of recent successful development of 3. our financial community does not understand the
industrial mineral operations in Australia are Consolidated profitability of well-managed industrial mineral businesses;
Minerals (manganese, chromite) and Portman Mining (iron ore).
At present we are being out manoeuvred by competitor 4. poor technical and commercial relationships with Asian
exploration and mining companies from Europe, UK and North entrepreneurial mining industry individuals or companies;
America for development of the best industrial mineral assets in and
Asia. 5. we are poor business negotiators and relatively poor
Why are Australian companies slow in evaluating investment teachers in Asia.
opportunities in industrial minerals in Asia?
1. Our geoscience community has a poor understanding of Challenge
how to explore and evaluate industrial minerals; There are plenty of opportunities in industrial minerals we just
2. our mining culture is predominantly metal centric; need to speak the same language.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 439

Looking to the Future Sustainability, Education and Skills

An Environmental Collaborative Research Program Field Studies A Black, T Clemens and 443
Contributing to the Sustainability of New Zealands Mineral Industry D Trumm

Water Quality Trigger Values for Suspended Mud in the Jabiluka Mine K G Evans, D R Moliere, 447
Receiving Catchment, NT, Australia M J Saynor, W D Erskine
and M G Bellio

Pioneering Community Collaboration in Mineral Exploration C M Horn and I B Miller 453

Southern Flinders Ranges

Environmental Auditing as a Driver for Continuous Improvement in K W Perry 459

Environmental Performance
An Environmental Collaborative Research Program Field
Studies Contributing to the Sustainability of New Zealands
Mineral Industry
A Black1, T Clemens1 and D Trumm1
ABSTRACT
Adopting environmentally acceptable practices in New Zealand is a
crucial requirement for mineral wealth development. Recent planned
expansions in the coal and gold industry have increasingly placed mining
in the media spotlight. A lack of appropriate standard environmental
guidelines has resulted in inconsistent responses and assessments of
current and proposed mines. These inconsistencies have lead to delays in
decisions on developing economic mineral deposits that may have only
minor short-term impacts, and have also allowed unacceptable
environmental management practices to continue elsewhere.
A research collaboration between CRL Energy, Landcare Research,
University of Canterbury and University of Otago aims to provide
regulator groups and the mining industry with a unified approach to the
requirements of the Resource Management Act 1991, the legislation that
largely drives the nations environmental policies and guidelines. The
tangible outcome of this research program will be water quality and
impact management guidelines integrated into a decision-making
framework. The result of this unification will assist in the development of
viable mineral deposits and minimise environmental impacts.
The framework will be derived from a synthesis of three objectives.
Objective 1 identifies potential water quality impact. This objective is
based on a targeted geochemical study of the lithology and waterways of
the West Coast and Southland regions. Objective 2 categorises the degree
of impacts on the ecology of receiving aquatic ecosystems and
determines the processes that drive aquatic ecosystem recovery. Objective
3 identifies best strategies to sufficiently manage impacted ecosystems,
and to prevent future unacceptable impacts. Wide consultation with
representatives of the mining industry and regulator groups will occur
frequently during the development stages of the framework. It is
anticipated that this multidisciplinary approach will provide a robust and
coherent, standardised system that will assist in achieving industry and
environmental targets.
INTRODUCTION
Mining has been an integral part of New Zealands landscape
since the 1800s and has approximately $80 billion in potential
wealth in known mineral deposits, comprised mainly of coal,
gold, aggregate, and iron sands (NZMIA, 2003). Currently it is
predicted that the $1 billion per annum gained from minerals will
double within ten years. This is to be achieved by increasing coal
production to six million tonnes ($400 million), precious metals
production to three million ounces ($750 million), aggregate to
37 million tonnes ($700 million) and iron sands to 3.5 million
tonnes ($100 million) (NZMIA, 2003).
High profile cases have created a negative public image of
mineral extraction. Recent media attention has focused on the
both the positive (economics) and negative (environmental)
aspects of the coal industry, particularly with the increase in
planned coal production to meet the international market demand
for low ash and sulfur coal. With the cancellation of the proposed
hydro scheme planned for the South Island, coal is being touted
as the medium term solution to the countrys increasing energy
needs (SENZ, 2003). New Zealands largest coal producer Solid
Energy New Zealand Ltd has committed to improve its
environmental management practices, with a statement in the
Annual Report of an environmental no net negative impacts
goal (SENZ, 2003).
1. CRL Energy Ltd, PO Box 29-415, Christchurch, New Zealand.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Since the passing of New Zealands Resource Management
Act (RMA, 1991), many industries in New Zealand, including
the minerals industry have made major operational and
management changes. Prior to 1991, mineral exploration, access
and extraction were the primary steps involved in generating
mineral wealth. Subsequently, there has been an increasing
emphasis on a further step, namely managing the environmental
impacts.
While environmental management is an essential part of any
development, the lack of standard guidelines coupled with
inconsistent procedures among parties has lead to confusion and
significant delays in developing economically viable mineral
deposits. Often regulators request further information and
clarification on the geochemical, remedial and ecological aspects
of the management proposal. Since there are no standardised
methods of assessing these aspects, interpretations are often
subjective.
The successful development of a standard system to provide
industry and regulators with a uniform process to base important
environmental decisions upon is a key element in cultivating
New Zealands mineral resources. Recognising this, the New
Zealand Government, through the Foundation of Science
Research and Technology, is financially supporting a research
collaboration between CRL Energy Ltd, Landcare Research,
University of Canterbury School of Biological Sciences and
University of Otago Department of Geology to provide industry
and regulators with a standardised decision making framework
for the West Coast and Southland Regions.
This program has been awarded $NZ 475 000 per annum for
six years (July 2004 - June 2010). The goal of this program is to
reduce impacts from current and future mining operations,
minimise environmental degradation and to facilitate processes
involved in developing mineral wealth and environmental
sustainability. A decision-making framework will be the overall
outcome of this collaborative research program and will be
derived from three objectives that focus on establishing
effects-based water quality threshold targets and management
strategies.
During the first three years a targeted detailed study of the
West Coast and Southland region will occur. These investigations
will identify the likelihood of water quality impacts by
examining the geochemical characteristics of the mineral
deposits. Impacts on the receiving ecosystems will be categorised
according to their degree of disturbance and thresholds of
tolerance and processes that drive ecosystem recovery will be
determined. This program will also identify strategies to mitigate
such impacts and expand on remediation techniques that have
already been tested at an historic site on the West Coast (Trumm
et al, 2003a).
A program governance panel, comprised of key strategic
individuals, will provide feedback and help to ensure that the
framework is sound and will be adopted by the identified end
users.
RESEARCH METHODOLOGY
This research is focused primarily on determining and managing
the likelihood of water quality impacts to be incorporated into
443
A BLACK, T CLEMENS and D TRUMM
the decision-making framework. The research has been divided
among three objectives:
1. establishing a baseline of water quality and rock
geochemistry to determine the likelihood of impacts from
mine related disturbances;
2. categorising the degree of ecological impact on algal and
macro-invertebrate communities through ecotoxicity
studies to develop water quality threshold limits; and
3. identifying the most appropriate impact management
strategy to remediate impacted ecosystems.
The sites chosen for this study are constrained within the West
Coast and Southland regions of the South Island, New Zealand
(Figure 1), where much of the growth in coal and gold mining is
planned.
Determining potential impacts
In the West Coast region, acid mine drainage (AMD) from coal
mining is one of the main sources of aquatic ecosystem
degradation (Trumm et al, 2003a). Historic and current coal
mining operations have resulted in a legacy of impacted streams
around the northern part of the region, namely catchments within
the acid producing Brunner Coal Measures (James, 2003). To
date, very little AMD or mine related environmental data is
available from the Southland region.
While gold mining has also been documented as a producer of
AMD in New Zealand and Internationally (Morell, et al, 1995;
Ritchie, 1993), it does not have any reported occurrences in the
South Island. Gold mining does however cause an increase in
bioavailability of trace elements in the streams (Webster-Brown
and Craw, 2004).
The South Islands geology, ecology, climate and topography
vary considerably from region to region (Black et al, 2004). As a
result, mines in different areas have different discharge water
quality and environmental impacts. The impact of pH and
dissolved trace elements from various mine discharges can range
from minor to severe, with the presence of environmentally
significant trace elements varying from site to site (Harding,
2004).
Tasman Sea
North Island
0 200
Kilometres
South Island
New Zealand
Invercargill
FIG 1 - The locality of the West Coast and Southland regions, South Island, New Zealand.
444 Adelaide, SA, 19 - 22 September 2004
While the science behind the geochemical processes of acid
generation and consequent contaminant problems is well
understood (Gray, 1997), the extent of the likelihood of
environmental impacts from different lithologies in the targeted
regions of New Zealand remains largely undefined. A significant
body of rock (geochemical) and water data (quality, flow,
rainfall) already exists for some of the South Islands regions
(Black 1999, 2000, 2001; Craw et al, 2000; Campbell, 2001;
Craw and Pacheo, 2002; James, 2003), and this information will
be collated into a consistent format.
Rock units containing economic amounts of coal and gold
from up to six new sites in the West Coast and Southland regions
will be geochemically assessed to determine the potential for
environmental impacts through mine drainage. The collation of
existing data and geochemical assessments will lead to
identification of areas that have the potential for high,
medium or low risk to water quality. The critical factor is to
select appropriate sites and design an adequate spatial and
temporal sampling regime. It is planned that a geochemical GIS
map and database of the regions will be developed once all
required information has been collated.
Identifying the impacts on the aquatic ecosystem
Impacts of AMD on stream systems are well documented
(Bradley, 2003). However, much of this research has focused on
assessing the ecology of macro-invertebrate communities rather
than determining the mechanisms responsible for the impacts, or
the processes that drive the recovery of impacted systems
(Harding et al, 1998, 1999, 2000, 2003; Stark, 2001).
Aquatic ecosystem community recovery includes a complex
set of scenarios that require more than just correcting the
physico-chemical parameters (OHalloran, 1998, 1999, 2001;
Eason, 1999; Cussins, 2000; Harding, 2004). It is necessary to
identify the biological factors that drive the recovery of impacted
streams so that methods designed to assess site-specific effects
are transferable to other areas.
From existing data on water quality, stream ecology and
ecotoxicity studies for the West Coast and Southland, a draft set
Westport
Tasman Sea
West Coast
Region
Christchurch
Pacific Ocean
Southland
Region
Scale: 100km
PACRIM 2004

of water quality thresholds will be formulated. Extensive algal
surveys in a range of impacted systems will be used to identify
impact tolerance levels. Experiments designed to identify
conditions limiting microbial processes will also be conducted
and ecotoxicological data on a representative range of endemic
invertebrate taxa will establish pH and metal tolerances. From
these results the levels of ecosystem recovery will identify
thresholds for water quality parameters.
The effect of mine drainage on aquatic taxa in West Coast
systems is complicated by the fact that many West Coast species
are highly endemic tolerate a wide range of pH and metals
(Bradley, 2003). This suggests an adaptation to naturally acidic
streams on the West Coast (Winterbourne et al, 2000).
Managing the risk
Based on findings from existing work at a historic West Coast
coal mine site (Lindsay et al, 2001; Black et al, 2003;
OHalloran et al, 2003; Trumm et al, 2003a, b) a methodology to
determine strategies for minor, moderately and severely impacted
sites will be formulated. Small-scale trial remediation systems
using passive remediation techniques will be trialled at other
minor and moderately impacted historic coalmine sites. The
results of these trials and previous research will be used to
develop, an impact management strategy and its success
measured against the water quality and ecological threshold
limits developed in Objectives 1 and 2.
Ultimately the best methods of managing the potential and
actual impacts is to either prevent or treat mine discharge so that
the physico-chemical parameters of the receiving water do not
exceed any ecological thresholds. The level of treatment required
will depend on the amount and duration of discharge in
combination with the natural background characteristics of the
receiving water.
Building the framework
The purpose of the framework is to provide appropriate water
quality and impact management guidance, which will allow
environmentally responsible extraction of mineral resources.
These guidelines will be designed to compliment existing
policies and generic guidelines currently used by regulators and
industry, particularly in the preparation of an assessment of
environmental effects, which accompanies resource consents (as
required under the New Zealand RMA, 1991).
Consultation with end users in industry and regulatory
organisations to discuss optimal ways of combining the results
from the three objectives into a useable framework will occur on
a regular basis. An evaluation of the various types of frameworks
and best management practices from other countries will provide
the initial structure (Cavanagh et al, 2003). This will be trialled
by end users at sites within the West Coast and Southland
regions. Modifications to improve the framework will occur as
required once the results of trials and feedback from end users is
collected.
The framework is initially envisaged as a comprehensive
document, containing a series of feedback loops between the
three main research themes. The initial output will be a
document detailing the rationale of the framework as well as
milestones answered in Objectives 1 - 3. Although the
framework will be built with water quality impacts from coal and
gold mine drainage, the process developed should be applicable
to the aggregate industry.
An effective decision-making framework must be based upon
a thorough scientific understanding of geochemistry and
ecosystem interactions and its response to natural and
anthropogenic induced stress and remediation techniques. It is
anticipated that recommendations will be developed to help
guide future work in a number of areas of potential extraction
sites and environmental sustainability.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
AN ENVIRONMENTAL COLLABORATIVE RESEARCH PROGRAM
CONCLUDING SUMMARY
The strategic objective of this research is to provide a unified
decision process to meet the requirements of the RMA and to
promote sound environmental practices in the mining industry.
Current guidelines require mining operations be completed in an
environmentally responsible way. However, it is recognised that
temporary localised adverse effects may be unavoidable during
mining and that management and operational practices need to
ensure that long-term there are no net negative environmental
impacts.
Responding to regulatory and public concerns, the New
Zealand mining industry has improved environmental
management practices over the last ten years. Management of the
environmental effects of mining in New Zealand has become
more publicised as a result of the increase in developments.
Mining in New Zealand, particularly for coal has a negative
image, despite the economic reliance on coal as an export
commodity and an energy source. In part, this image is
understandable because some historic sites still noticeable and
significantly impact on water quality in streams (Trumm et al,
2003).
The final framework, incorporating the geochemical
information, ecosystem recovery threshold targets and
management strategies for the West Coast and Southland regions
will be completed and in place by June 2010.
ACKNOWLEDGEMENTS
The authors would like to thank the Foundation for Research,
Science and Technology for investment in this new contract and
to also thank Solid Energy New Zealand Ltd, for financial
support and encouragement of this new initiative.
REFERENCES
Black, A and Craw, D, 1999. Environmental effects of open cast coal
mining and the in situ processes of natural remediation: Wangaloa
open cast mine, South Otago, New Zealand, in Proceedings The
Australasian Institute of Mining and Metallurgy New Zealand
Branch 32nd Annual Conference, pp 121-130 (The Australasian
Institute of Mining and Metallurgy: NZ Branch).
Black, A and Craw, D, 1999. Seasonal variation in arsenic, copper and
zinc in surrounding streams and wetlands receiving coalmine
drainage, Wangaloa open cast mine, southeast Otago, in Proceedings
8th New Zealand Coal Conference, pp 263-267.
Black, A and Craw, D, 2000. A preliminary environmental impact
assessment on historic coal mining in Reefton, West Coast, New
Zealand, Earth Science Research Report 00/01, 44p (Department of
Geology, University of Otago).
Black, A and Craw, D, 2001. Arsenic, copper and zinc occurrence at the
Wangaloa coal mine, southeast Otago, New Zealand, International
Journal of Coal Geology, 45:181-193.
Black, A, Craw, D, Youngson, J and Karubaba, J, in press. Natural
rehabilitation of historic mine tailings discharged into a river system,
Shag River, east Otago, New Zealand, Journal of Geochemical
Exploration.
Black, A, Trumm, D T, Cavanagh, J, Harding, J, de Joux, A, Moore, T A
and OHalloran, K, 2003. Developing remediation strategies for New
Zealand sites impacted by Acid Mine Drainage (AMD). A case
study, Sullivan Mine, West Coast, New Zealand, in Proceedings Sixth
International Symposium on Environmental Geochemistry, 7 - 11
September, Edinburgh.
Black, A, Trumm, D, and Lindsay, P, in press. Coal mining in New
Zealand and impacts on water quality: case studies from West Coast
and Otago, in Metal Contaminants in New Zealand: From Sources
and Transport to Effects on Ecology and Human Health, (Eds: T A
Moore, A Black, J A Centeno, J S Harding, D A Trumm).
Bradley, A, 2003. Stream communities in acid mine drainages:
investigating longitudinal recovery and remediation possibilities,
Master of Science in Zoology thesis, University of Canterbury,
Christchurch, New Zealand.
445
A BLACK, T CLEMENS and D TRUMM
Campbell, R N, Lindsay, P and Clemens, A H, 2001. Acid generating
potential of waste rock and coal ash in New Zealand coal mines,
International Journal of Coal Geology, p 163.
Cavanagh, J E and OHalloran, K, 2003. International legislative
frameworks for managing contaminated land derivation and
application of numeric values, Landcare Research Contract report:
LC0203/110, 40 p.
Craw, D, Chappell, D, Reay, A and Walls, D, 2000. Mobilisation and
attenuation of arsenic around gold mines, east Otago, New Zealand,
New Zealand Journal of Geology and Geophysics, 43:373-383.
Craw, D and Pacheco, L, 2002. Mobilisation and bioavailability of
arsenic around mesothermal gold deposits in a semiarid environment,
Otago, New Zealand, Scientist World Journal, 2:308-319.
Cussins, T, OHalloran, K, Lilburne, L, Cochrane, P, Hickey, C,
Sheppard, D, Gough, J, Wickstrom, M, Eason, C and Phillips, C,
2000. Ecological risk assessment at contaminated sites in New
Zealand, in Contaminated Site Remediation: From Source Zones to
Ecosystems: Proceedings of the 2000 Contaminated Site
Remediation Conference, (Ed: C D Johnston), 4 - 8 December,
Melbourne, pp 17-24.
Eason, C, Svendsen, C, OHalloran, K and Weeks, J M, 1999. An
assessment of the lysosomal neutral red retention test and immune
function assay in earthworms (Eisenia andrei) following exposure to
chlorpyrifos, benzo-a-pyrene (BaP) and contaminated soil,
Pedobiologica, 43:641-645.
Gray, N F, 1997. Environmental impact and remediation of acid mine
drainage: a management problem, Environmental Geology,
30(1/2):62-71.
Harding, J S, 2003. Historic deforestation and the fate of endemic species
in streams, NZ Journal of Marine and Freshwater Research,
37:333-345.
Harding, J S, Benfield, E F, Bolstad, P V, Helfman, G S and Jones, E B
D, 1998. Stream biodiversity: the ghost of land-use past, in
Proceedings National Academy of Sciences, 95(25):14843-14847.
Harding, J S, in press. Impacts of metals and mining on stream
communities, in Metal Contaminants in New Zealand: From Sources
and Transport to Effects on Ecology and Human Health (Eds: T A
Moore, A Black, J A Centeno, J S Harding and D A Trumm).
Harding, J S, Quinn, J M and Hickey, C W, 2000. Effects of mining and
forestry production, in New Zealand Stream Invertebrates: Ecology
and Implications for Management (Eds: M J Collier, and M J
Winterbourn) (New Zealand Limnological Society: Christchurch,
New Zealand).
Harding, J S, Young, R G, Hayes, J W, Shearer, K A and Stark, J D, 1999.
Changes in agricultural intensity and river health along a river
continuum, Freshwater Biology, 42:345-357.
James, T, 2003. Water quality of streams draining various coal measures
in the North-central West Coast, in Proceedings 36th Annual AusIMM
Conference, 2 - 5 September, Greymouth, New Zealand (The
Australasian Institute of Mining and Metallurgy: NZ Branch).
Lindsay, P, Mowatt, C and Trumm, D, 2001. Proposed remediation of
contaminated sites using zeolites, in Proceedings Engineering and
Development in Hazardous Terrain, New Zealand Geotechnical
Society Symposium, University of Canterbury, 24-25 August.
446 Adelaide, SA, 19 - 22 September 2004
Morell, W J, Gregg, P E H, Stewart, R B, Bolan, N and Horne, D, 1995.
Potential for revegetating base metal tailings at the Tui Mine site, Te
Aroha, New Zealand, in Proceedings PACRIM 95, pp 395-400 (The
Australasian Institute of Mining and Metallurgy: Melbourne).
NZMIA, 2003. New Zealand Minerals Association website:
http://www.minerals.co.nz
OHalloran, K. 2001: Ecotoxicity thresholds for hazardous waste.
Landcare Research Contract Report LC0001/66.
OHalloran, K, Ahokas, J T and Wright, P F A, 1998. Response of fish
immune cells to in vitro organotin exposures, Aquatic Toxicology,
40:141-156.
OHalloran, K, Ahokas, J T and Wright, P F A, 1998. The adverse effects
of aquatic contaminants on fish immune responses, Australasian
Journal of Ecotoxicology, 4:9-28.
OHalloran, K, Black, A, Bradley, A, Cavanagh, J, Harding, J, de Joux,
A, Moore, T and Trumm, D, 2003. Assessment and remediation of A
New Zealand West Coast Acid Mine Drainage (AMD) site, Society
for Environmental Toxicology and Chemistry, 28 September -
1 October, Landcare Research, New Zealand.
OHalloran, K and Booth, L H, 2001. Terrestrial toxicity tests for soil
fauna and flora, in Organic contaminants (Eds: Kookana, Sadler,
Sethunathan and Naidu), pp 55-74, New Delhi, IBH Oxford.
OHalloran, K, Booth, L H, Hodge, S, Thomsen, S and Wratten, S D,
1999. Biomarker responses of the earthworm Aporrectodea
caliginosa to organophophates: Hierarchical tests, Pedobiologia,
43:646-651.
Ritchie, A I M, 1993. Waste rock dump geochemistry, in Proceedings
Symposium of the Management and Rehabilitation of Waste Rock
Dumps, (Eds: S J Riley, P W Waggitt and C McQuade), pp 69-90.
SENZ (Solid Energy New Zealand Limited), 2003. Annual report, p 47.
Stark, J D, Boothroyd, I K, Harding, J S, Maxted, J R and Scarsbrook, M
R, 2001. Protocols for sampling macroinvertebrates in wadeable
streams, New Zealand Macroinvertebrate Working Group Report No
1, 57 p (Ministry for the Environment: Wellington).
Trumm, D A, Black, A and Gordon, K, 2003a. Remediation strategies for
acid mine drainage (AMD) at Sullivan Mine, West Coast, New
Zealand, in Proceedings 36th Annual AusIMM Conference, 2 - 5
September, Greymouth, New Zealand (The Australasian Institute of
Mining and Metallurgy: NZ Branch).
Trumm, D T, Black, A, Cavanagh, J, Harding, J, de Joux, A, Moore, T A
and OHalloran, K, 2003b. Developing assessment methods and
remediation protocols for New Zealand sites impacted by Acid Mine
Drainage (AMD), in Proceedings Sixth International Conference on
Acid Rock Drainage, pp 223-230 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
Webster-Brown and Craw, in press. Examples of trace metal mobility
around historic and modern metal mines in New Zealand, in Metal
Contaminants in New Zealand: From Sources and Transport to
Effects on Ecology and Human Health (Eds: T A Moore, A Black, J
A Centeno, J S Harding, D A Trumm.
Winterbourne, M J, McDiffet, W F and Eppley, S J, 2000. Aluminium
and iron burdens of aquatic biota in New Zealand streams
contaminated by acid mine drainage: effects of trophic level, The
Science of the Total Environment, 254:45-54.
PACRIM 2004
Water Quality Trigger Values for Suspended Mud in the
Jabiluka Mine Receiving Catchment, NT, Australia
K G Evans1,2, D R Moliere1, M J Saynor1, W D Erskine3,4 and M G Bellio1
ABSTRACT
The Jabiluka uranium mine was constructed in the Ngarradj catchment, a
right bank tributary of Magela Creek in the Northern Territory. The
Jabiluka Mineral Lease is adjacent to Kakadu National Park, near an area
listed as Wetlands of International Importance under the Ramsar
Convention and recognised under the World Heritage Convention.
In the late dry season of 1998 and prior to the 1998 - 1999 wet season,
a stream-monitoring program was implemented in the Ngarradj
catchment to collect baseline data on sediment transport within the
catchment to assess and proactively manage associated impacts caused by
mining at Jabiluka.
Stream suspended mud concentration, stream discharge and rainfall
data were collected for four wet seasons (1998 - 2002) giving an almost
continuous data set. The stream system has high natural variability
dependent on rainfall intensity and subsequent discharge and pre-wet
season fire distribution and intensity. In general, mud concentration is
elevated at the commencement of the wet season until about February.
This is caused by first-flush removal of surface material detached during
the dry season by agents such as bioturbation (including anthropogenic
activity), wind erosion and surface desiccation.
1. Environmental Research Institute of the Supervising Scientist,
Supervising Scientist Division, Environment Australia, GPO Box
461, Darwin NT 0801.
2. School of Environmental and Life Sciences, The University of
Newcastle, Callaghan NSW 2308.
3. Office of the Supervising Scientist, Supervising Scientist Division,
Environment Australia, GPO Box 461, Darwin NT 0801.
4. Present Address: Environmental Management and Forest Practices
Directorate, State Forests of New South Wales, PO Box 100,
Beecroft NSW 2119.
FIG 1 - Study site.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Water quality trigger values downstream of the mine were derived
using upstream percentiles and the before-after-control-impact, paired
site design. Both analyses showed no observable impact from mine
construction in 1998 or during the study period. However, the impacts of
dry season fires could be clearly detected. The data set provides good
baseline information for future assessment at Jabiluka and demonstrates
the need to view measured parameter values on a catchment-wide basis
with knowledge of rainfall, discharge and fire distribution.
INTRODUCTION
In the 1998 dry season, construction started on the Jabiluka
uranium mine located in the Ngarradj catchment, a right bank
tributary of Magela Creek. The Jabiluka Mineral Lease (JML) is
adjacent to Kakadu National Park (KNP). The Magela Creek and
floodplain are located within KNP, and are listed as Wetlands of
International Importance under the Ramsar Convention and
recognised under the World Heritage Convention. KNP is one of
the largest and most environmentally diverse national parks
worldwide and is managed for conservation, tourism and natural
and cultural features and values (Kakadu Board of Management,
1998). Ngarradj has its headwaters on the Arnhem Land Plateau,
flows through lowlands within the JML, and debouches into the
Magela Creek floodplain in KNP (Figure 1).
A stream-monitoring program was implemented in the
Ngarradj catchment by the Environmental Research Institute of
the Supervising Scientist (eriss) in the late dry season of 1998
and prior to the 1998 - 1999 wet season (Erskine et al, 2001).
Baseline data on catchment sediment movement were collected
to assess and proactively manage associated impacts caused by
mining at Jabiluka.
447
K G EVANS et al
Rainfall, stream suspended mud concentration and stream
discharge data were collected for four wet seasons (1998 - 2002).
The data have allowed:
1. determination of baseline characteristics of water quality
with respect to suspended mud in the catchment; and
2. assessment of impact on water quality resulting from
construction of the mine site.
STUDY SITE
The region is in the wet dry tropics of the Northern Territory,
Australia. High-intensity storms and rain depressions occur
between October and April (wet season) with little rain falling
during the remainder of the year (dry season). Jabiru Airport,
approximately 20 km to the south west of the Jabiluka mine site
receives an average annual rainfall of 1483 mm ( = 302.5 mm)
and Oenpelli, approximately 20 km to the northeast receives an
average annual rainfall of 1397 mm ( = 284.5 mm) (Bureau of
Meteorology, 1999).
Construction of the mine included a retention pond, portal,
infrastructure such as roads, offices and drainage systems and a
small waste rock dump. The site is located in small
sub-catchments to the west of the main Ngarradj channel and it is
likely that erosion products from the site will eventually enter
Ngarradj (Figure 1). The site was rehabilitated in 2003 as there
are no immediate plans to mine the orebody.
The Ngarradj catchment primarily comprises highly weathered
Middle Proterozoic sandstone in the upper reaches (Arnhem
Land Plateau) (Figure 1), with Tertiary and Quaternary alluvium
and colluvium comprising the flood inundated western branch
and northern floodplain of the creek.
MONITORING DESIGN
Three river gauging stations were installed in Ngarradj (Figure 1)
prior to the commencement of the 1998 - 1999 wet season
(Erskine et al, 2001):
FIG 2 - Discharge, suspended mud concentration and daily rainfall at the SC gauge during the 1999-00 wet season. Mud concentration
trigger levels based on upstream percentiles are also shown.
448 Adelaide, SA, 19 - 22 September 2004
1. On the main channel upstream of the mine site, referred to
in this paper as UM (Upper Main). Data from the
un-impacted catchment upstream of the mine were
collected from this site.
2. On the East Tributary channel upstream of the confluence
with the main channel, referred to in this report as ET (East
Tributary). Data from the un-impacted East Tributary
catchment were collected from this site.
3. On the main channel referred to in this report as SC (Swift
Creek). This station is downstream of the mine site and
collected JML data.
The experimental design was that changes in the western part of
the catchment, where the mine is located, should be seen
downstream at SC when compared with the upstream ET and UM
unimpacted data.
SUSPENDED MUD CONCENTRATION
Suspended mud (<63 m >0.45 m diameter) is the sum of the
silt and clay size fractions and is an important indicator of stream
health because it is associated with contaminant transport,
increases in turbidity and adverse affects on aquatic ecosystems
(Walling and Webb, 1985; Neal, Jarvie and Oguchi, 1999; Pentz
and Kostaschuk, 1999; Bonta, 2000). An example data set
showing the variability of suspended mud concentration, [mud],
with rainfall and stream discharge is given in Figure 2. Flow and
[mud] descriptive statistics are given in Table 1. The mean
suspended mud concentration, [mud], reduces from the smallest
catchment, ET, to the largest, SC, indicating that [mud] is
inversely proportional to catchment size. But [mud] at SC and
UM are very similar being 10 ( = 16) mg L-1 and 10 ( = 12)
mg L-1 respectively.
The annual descriptive statistics are presented in Table 2. In
general, annual mean mud concentration (m) increased from
the largest SC catchment to the smallest ET catchment which
conforms with sediment delivery ratio theory (Robinson, 1977;
Walling, 1983). Analysis indicates that none of the data are
normally distributed.
PACRIM 2004
 WATER QUALITY TRIGGER VALUES FOR SUSPENDED MUD

TABLE 1
Mean, median and range for SC, UM and ET gauging station data.

SC UM ET
Parameter Mean Median Max Min Mean Median Max Min Mean Median Max Min
[mud] (mg L-1) 10 6 205 0.3 10 6 91 0.3 18 8 404 0.3
Flow (m3 s-1) 5.25 4.38 21.7 0.01 3.04 2.01 12.9 0.001 2.01 1.35 8.11 0.003

TABLE 2
Descriptive statistics for suspended mud concentrations (mg L-1).

Untransformed Flow duration

Site/Year n Annual Sk Median Mode Min Max Start Finish
Mean (dd:mm) (dd:mm)
SC 98-99 374 16 20 2.3 8 6 0.3 108 9/12 20/4
SC 99-00 451 7 9 3.1 5 2 0.3 69 22/11 28/4
SC 00-01 402 7 6 2.4 4 3 0.3 44 28/11 06/4
SC 01-02 177 15 29 4.4 7 4 0.3 205 31/12 09/03
UM 98-99 299 9 10 2.2 6 2 0.3 65 14/12 08/4
UM 99-00 411 11 10 1? 6 2 0.3 52 20/11 26/4
UM 00-01 421 11 13 2.5 6 3 1 91 04/12 09/04
UM 01-02 190 11 13 2.5 6 6 0.3 74 31/12 11/03
ET 98-99 343 26 39 4.3 12 4 0.3 404 21/12 27/4
ET 99-00 410 23 34 2.8 10 4 0.3 226 20/11 25/4
ET 00-01 389 8 11 3.7 4 3 1 88 28/11 02/4
ET 01-02 198 13 28 5.3 6 2 0.3 228 31/12 10/03

n = count; = sample standard deviation (mg L-1); Sk = skewness.

After the first flush, when concentrations are generally
elevated, [mud] at SC reached a relatively constant level with a
monthly median of <10 mg L-1 by January (Figure 3). At the
nearby Ranger mine site, sediment loss from vegetated erosion
plots has been observed to reach a relatively constant level by
mid-January (Saynor and Evans, 2001). Elsewhere in the region
soil erosion was observed to decline as vegetation cover
increased during the wet season although rainfall momentum
remained constant (Williams, 1969). The constant-level [mud] is
considered to be the unimpacted baseline level at the downstream
SC site. At UM, [mud], is reasonably consistent between years
during the monitoring seasons (Figure 3). There is first-flush
evidence at UM in December 2001 but [mud] is much less then
that recorded at SC for the same period. At ET, [mud] is less
consistent than at SC and UM between years and months
although 1998 - 1999 follows similar trends as 2000 - 2001 but at
an elevated level (Figure 3). ET generally has higher monthly
maxima and medians than UM and SC.
The [mud] levels at UM and SC are consistent during the
monitoring years during the principal flow months of January,
February and March, except for January 1999 and 2002 at SC.
The elevated [mud] at SC in December 1998, 2001 and January
1999, 2002 were not observed at UM or ET. The source of the
elevated [mud] at SC in December 1998 and January 1999
appears to be the western part of the catchment where two major
impacts occurred during the 1998 dry season prior to monitoring.
These were the construction of the mine and an extremely hot
fire, which burnt out most of the western catchment (Figure 4) in
the late dry season. A fire also occurred in the 2001 dry season
(Figure 4), which probably contributed to the elevated [mud] at
SC in December 2001 and January 2002. Figure 4 indicates that
both of these dry season fires had relatively little impact on the
catchment areas upstream of UM or ET. There have been a
number of studies, which found that significant increases in
erosion occur post-fire (Atkinson, 1984; Zierholz, Hairsine and
Booker, 1995; Evans, Saynor and Willgoose, 1999), which would
subsequently result in elevated stream sediment concentrations.
Mud concentrations at UM and SC appear to have similar
responses to natural catchment conditions, but ET concentrations
appear to change independently.
WATER QUALITY TRIGGER VALUES TO ASSESS
IMPACT
The Australian and New Zealand water quality guidelines
(WQG) (ANZECC and ARMCANZ 2000) were used to
determine numerical mud concentrations for SC, which when
exceeded, will trigger a management response. Two methods
were used:
1. comparison of the downstream SC site with percentile
limits at the upstream UM site; and
2. a Before-After-Control-Impact, paired difference design
(BACIP) where the upstream site UM is before impact in a
spatial sense and the downstream site SC is after impact in
a spatial sense.
In both cases the ET site can be used to confirm whether an
observed elevated measurement at SC not observed at UM is:
1. from the mine-site catchment, or
2. from ET and therefore a natural occurrence.
If elevated values are not observed at ET it is assumed that the
source is from the mine-site catchment and investigations are
required to identify the source.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 449

K G EVANS et al

FIG 3 - (A) Monthly suspended mud distribution in SC. (B) Monthly suspended mud distribution in UM.
(C) Monthly suspended mud distribution in ET.

FIG 4 - Fire distribution in the dry season 1998 and the dry season 2001.

450 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


Upstream limits
Where data are not normally distributed, the WQG recommend a
trigger value of the 80th percentile of parameter values of a
suitable reference. In this case the reference site of the
downstream site, SC, is the upstream site, UM. This means that
values measured at SC should be lower than the 80th percentile at
UM. The Supervising Scientist Division, in discussion with
stakeholders, is establishing trigger values for the Alligator
Rivers Region including Ngarradj. These values are, 80th, 95th
and 99.7th percentiles representing different levels of
interventions by supervising authorities and the mining company.
The levels of intervention are still under discussion.
The trigger values for SC have been determined using the
complete data set from four years monitoring (Table 1). The
trigger values are given in Table 3 and are plotted on the SC data
presented in Figure 2. A number of SC data points (Figure 2) are
above the trigger levels. This is due to event related variability.
TABLE 3
Trigger values for the downstream SC site derived using the
upstream UM reference site.
Percentile [Mud] (mg L-1)
80 16
95 36
99.7 70
Although they have the same median (6 mg L-1), SC [mud] is
higher than the 80th percentile at UM less then 20 per cent of the
time (Table 4). This indicates that with respect to sediment, SC
flows at a lower concentration than UM. Mud yields at SC, UM
and ET are 11, 27 and 61 mg ha-1 s-1 respectively. The trend in
these yields concurs with sediment delivery theory, and the data
indicates that for similar disturbance conditions, larger area
catchments generally have lower concentrations than smaller area
catchments as is the observed situation in Ngarradj. Therefore the
use of upstream percentiles is conservative. SC is at concentrations
higher than the UM 99.7th percentile concentration >0.3 per cent
of the time. These relatively infrequent concentrations are
probably due to the first flush and fire-effected spikes observed at
SC and not UM. The data show sediment and turbidity spikes that
exceed the 99.7th percentile only during the fire-affected years of
1998 - 1999 and 2001 - 2002 during events early in the season.
99
[mud]
95
90
80
70
Percent
60
50
40
30
20
10
5
-5
mg/L
FIG 5 - Probability distributions of SC UM for the measured parameters. AD* = Anderson-Darling test statistic (AD0.05 = 2.49). AD* for
these plots indicates that the SC UM population for all parameters approximates a normal distribution.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
WATER QUALITY TRIGGER VALUES FOR SUSPENDED MUD
TABLE 4
Percentage of SC data points exceeding trigger values given in
Table 3.
Trigger percentile Percentage data points above trigger value (%)
80 15.8
95 5
99.7 1.6
BACIP
This assessment uses a before-after-control-impact, paired
differences design (BACIP) (Stewart-Oaten, Murdoch and
Parker, 1986; Stewart-Oaten, Bence and Osenberg, 1992;
Humphrey, Faith and Dostine, 1995). For the BACIP method,
UM and SC are treated as paired sites and the comparison of
differences is used to assess impact.
The time period chosen for comparison was one calendar
month. The parameter used to assess impact was the monthly
median ()~ parameter value at UM and SC because distributions
were generally skewed (ANZECC and ARMCANZ 2000, Table
6.9a). The medians of two distributions at UM and SC were
compared ie ~ of all samples collected in a month.
The test parameter used was the inference about the difference
after (downstream) and before (upstream) impact ie the
difference between the median concentration at SC and UM (SC
UM) (ANZECC and ARMCANZ 2000, Table 6.9b). Outliers,
which were considered to be those points outside two standard
deviations from the mean of the SC UM population were
removed and power analysis was conducted to determine the
probability of Type I and II errors occurring for the number of
samples available in this study. There were 18 monthly data
~ The WQGs recommend that assessment be
points available for .
based on the 24 most recent data values. However, power
analysis indicates that for the number of samples available in this
study for an effect size of one standard deviation from the mean,
the probability of a Type II () error with alpha of 0.05 is 0.20
with a power of 80 per cent.
The distribution of the SC UM sample population was
assessed. If the distribution was normal then standard deviations
were used to set trigger values, otherwise the trigger values were
set as percentiles. Since the distribution of the population of SC
UM is normal (Figure 5), trigger values were set as +1, +2 and
+3 respectively. Figure 6 shows ~ for UM and SC and SC
UM. The trigger values are also plotted. Outliers usually result
Mean -0.888889
StDev 1.88234
Goodness of Fit
AD* 1.433
0 5
451
K G EVANS et al
30
[SC]-[UM ]
[UM ]
25
Concentration (mg L )
-1
[SC]-[UM ] 2nd trigger = 3.0 mg/L
20
15
10
-5
Nov-98 Apr-99
FIG 6 - Temporal variation of qSC UM for the measured parameters and first, second and third order trigger values. Monthly median values
for UM and SC are also shown.
from first flush, fire-impacted events at the start of the
1998 - 1999 wet season and the 2001 - 2002 wet season.
SC UM occasionally exceeds +1 but rarely exceeds +2.
Values exceeding +2 are infrequent and irregular and probably
result from rainfall and hydrograph variation as expected in this
variable natural system.
CONCLUSIONS
The four years of monitoring have provided a high-resolution
data set that is almost continuous. The system has high natural
variability dependent on rainfall event and subsequent discharge
and pre-wet season fire distribution and intensity. There are some
similarities between parameter populations between SC and UM
but ET displays different population distributions. ET is an
important sampling point to confirm whether elevated levels at
SC come from upstream of the mine site or from the mine site
catchment.
In general, parameter values are elevated at the
commencement of the wet season until about February. This is
caused by first-flush removal of surface material detached during
the dry season by agents such as bioturbation (including
anthropogenic activity), wind erosion and surface desiccation.
Water quality trigger levels at SC were derived using UM
percentiles and the before-after-control-impact, paired site
design. Both analyses showed no observable impact from mine
construction in 1998 or during the study period. The impacts of
dry season fires could be clearly seen.
The data set provides good baseline information for future
assessment at Jabiluka and demonstrates the need to view
measured parameter values on a catchment-wide basis with
knowledge of rainfall, discharge and fire distribution.
REFERENCES
ANZECC and ARMCANZ, 2000. Australian and New Zealand guidelines
for fresh and marine water quality, National Water Quality
Management Strategy Paper No 4, (Australian and New Zealand
Environment and Conservation Council and Agriculture and Resource
Management Council of Australia and New Zealand: Canberra).
Atkinson, G, 1984. Erosion damage following bushfires, J Soil Cons
NSW, 40:4-9.
Bonta, J V, 2000. Impact of coal surface mining and reclamation on
suspended sediment in three Ohio watersheds, J Am Wat Resourc
Assoc, 36(4):869-887.
452
[SC]
[SC]-[UM ] 1st trigger = 1.0 mg/L
[SC]-[UM ] 3rd trigger = 4.9 mg/L
[mud] mean [SC]-[UM] = -0.9 mg/L
Oct-99 Apr-00 Oct-00 Apr-01 Oct-01 Apr-02
Date
Bureau of Meteorology, 1999. Hydrometeorological analyses relevant to
Jabiluka, Supervising Scientist Report 140, Supervising Scientist,
Canberra.
Erskine, W D, Saynor, M J, Evans, K G and Boggs, G S, 2001.
Geomorphic research to determine the off-site impacts of the
Jabiluka Mine on Swift (Ngarradj) Creek, Northern Territory,
Supervising Scientist Report 158, Supervising Scientist, Darwin.
Evans, K G, Saynor, M J and Willgoose, G R, 1999. Changes in
hydrology, sediment loss and micro-topography of a vegetated mine
waste rock dump impacted by fire, Land Degradation and
Development, 10:507-522.
Humphrey, C L, Faith, D P and Dostine, P L, 1995. Baseline
requirements for assessment of mining impact using biological
monitoring, Australian Journal of Ecology, 20:150-166.
Kakadu Board of Management and Parks Australia, 1998. Kakadu
National Park Plan of Management (Australian Government Public
Service: Canberra).
Neal, C, Jarvie, H P and Oguchi, T, 1999. Acid-available particulate trace
metals associated with suspended sediment in the Humber rivers: a
regional assessment, Hydrol Processes, 13:1117-1136.
Pentz, S B and Kostaschuk, R A, 1999. Effect of placer mining on
suspended sediment in reaches of sensitive fish habitat, Environ
Geol, 37(1-2):78-89.
Robinson, A R, 1977. Relationship between soil erosion and sediment
delivery, Inter Assoc Hyd Sci, 122:159-167.
Saynor, M J and Evans, K G, 2001. Sediment loss from a mine waste
rock dump, Northern Australia, The Aust Geographer, 39(1):34-51.
Stewart-Oaten, A, Murdoch, W W and Parker, K R, 1986. Environmental
impact assessment: Pseusoreplication in time? Ecology,
67:929-940.
Stewart-Oaten, A, Bence, J R and Osenberg, C W, 1992. Assessing
effects of unreplicated perturbations: No simple solutions, Ecology,
73:1396-1404.
Walling, D E, 1983. The sediment delivery problem, Journal of
Hydrology, 65:209-237.
Walling, D E and Webb, B W, 1985. Estimating the discharge of
contaminants to coastal waters by rivers: some cautionary comments,
Marine Pollution Bull, 16(12):488-492.
Williams, M A J, 1969. Prediction of rainsplash erosion in the seasonally
wet tropics, Nature, 222:763-765.
Zierholz, C, Hairsine, P and Booker, F, 1995. Runoff and soil erosion in
bushland following the Sydney bushfires, Aust J Soil and Water
Cons, 8: 28-37.
Adelaide, SA, 19 - 22 September 2004 PACRIM 2004
Pioneering Community Collaboration in Mineral Exploration
Southern Flinders Ranges
C M Horn1 and I B Miller2
ABSTRACT
Magnesium International Limited (MIL) has broken new ground in their
desire to integrate their mineral development activities in the Southern
Flinders Ranges of South Australia in to the communities of the region,
despite there being no statutory requirement to do so. In their efforts to
define a magnesite resource adjacent to the proposed site for a
magnesium metal processing facility, MIL opened their exploration
program up for public consultation.
In conjunction with the Port Pirie Regional Development Board
(PPRDB), a community group was formed to review the impact of
exploration and mining on the ecologically sensitive Southern Flinders
region. This voluntary move towards full community consultation for
mineral exploration was a first in South Australia and probably in
Australia.
The Community Focus Group (CFG) worked closely with MIL,
Primary Industries and Resources of South Australia (PIRSA) and the
Department for Environment and Heritage (DEH) in the development of
an exploration program and Declaration of Environmental Factors (DEF)
acceptable to all stakeholders. Collaboratively, the community concerns
were discussed and addressed, with the community gaining a very
comprehensive understanding of the exploration process and possible
future mine development. A balance of social, environmental and
economic factors was developed that was accepted by all parties and
encapsulated in the final DEF for approval by the regulatory authority.
The consultation process was a new experience for all of the
stakeholders, each of them effectively exploring new ground in addressing
information dissemination, risk management and the social versus
economic equation. MIL developed not only a viable and acceptable
exploration program but also substantially reinforced its position as a
responsible corporate citizen with the Southern Flinders communities.
INTRODUCTION
In 1998 Magnesium International Limited (MIL), formerly Pima
Mining NL, acquired mineral exploration licences in the Leigh
Creek area to explore identified magnesite deposits as a possible
source of raw material for a proposed magnesium metal
processing facility.
The SAMAG project progressed rapidly between 1999 and
2002, firstly outlining a significant magnesite resource at Leigh
Creek and subsequently acquiring a licence to utilise the Dow
Chemical Companys electrolytic process technology. SAMAG
entered into a Metal Sales Agreement with Thyssen Krupp
Mettalurgie GmbH and prepared a comprehensive Environmental
Impact Statement which received approval in July 2001. In
August 2002 a Bank Feasibility Study was completed.
Port Pirie was selected for the location of the proposed
SAMAG plant because of the proximity to the magnesite
deposits, existing environmental and native title approvals,
availability of support services and skilled personnel and the
strong support for the project from the Port Pirie community.
MIL, as part of an on-going desire to reduce the projects
operating costs, examined known deposits of magnesite located
approximately 20 km east of the proposed plant site to define the
potential resource in terms of quality and quantity, (Figure 1).
1. FAusIMM, Geological Consultant, Hornet Resource Assessment
Services Pty Ltd, 22 Colman Road, Goolwa SA 5214.
2. FAusIMM, Economic Development Officer, Port Pirie Regional
Development Board, 85 Ellen Street, Port Pirie SA 5540.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
PROJECT BACKGROUND
During 2001, Magnesium Developments Limited (MDL)
reviewed available geological data and past mining records with
respect to deposits of magnesite known to occur in the Southern
Flinders Ranges east of Port Pirie. As a result MDL applied for
and was granted Exploration Licence 2828 on 17 July 2001
covering an area of 1415 km2. A second smaller Exploration
Licence 2944 of 69 km2, was granted on 15 May 2002, and
covered known extensions of the magnesite beds in the Germein
Gorge area (Figure 1).
In early 2002 MDL contracted a HyMapTM high-resolution
hyperspectral airborne scanner survey over the magnesite
deposits to assist with geological interpretation. A limited
surface rock chip sampling of the outcropping magnesite beds
confirmed a similar chemical and geological character to the
Leigh Creek deposits. The Collaby Hill area at the south-western
end of EL 2828 was selected for detailed exploration and
possible mining, as this area provides the closest deposits by
road to the proposed Port Pirie plant site (Figure 1).
The sensitive locality of the magnesite project site and the
biodiversity and natural history of the Southern Flinders Ranges
indicated a possible community concern with exploration and
mining in the area. The Company, together with the Port Pirie
Regional Development Board (PPRDB), held a public meeting to
determine the issues of concern to the local residents, in
particular those in the nearby Beetaloo Valley. As a result of the
public meeting, MDL, in conjunction with the PPRDB, initiated
a formal community consultation group to overview the
proposed exploration in the Collaby Hill area. The authors
believe that the initiative was a very innovative approach and a
first in South Australia. Public consultation is not required by
legislation at this stage of the development.
LEGISLATIVE REQUIREMENTS
Mining Act 1971
The purpose of this legislation is to facilitate and regulate
responsible mineral exploration and mine development. The Act
provides for secure title for exploration and development of
mineral deposits and provides land owners, occupiers and other
land users with the right to ensure that environmental disturbance
is minimised and mined land appropriately rehabilitated and
where necessary land owners adequately compensated.
Development Act 1993
This legislation is administered by Planning SA to facilitate the
planning and development of South Australia and provides an
integrated suite of development plans that implement the
planning strategy. The planning strategy sets out the State
Governments vision for development and covers the full range
of social, economic and environmental issues.
Collaby Hill prospect falls within the Ranges Zone
administered by the Northern Areas Council and the Port Pirie
Regional Council. Principals of development control relevant to
the development of mineral deposits within the Ranges Zone are:
453
C M HORN and I B MILLER

FIG 1 - Interpreted magnesite outcrop and HyMap survey Port Pirie area, SA.

454 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 PIONEERING COMMUNITY COLLABORATION IN MINERAL EXPLORATION SOUTHERN FLINDERS RANGES
Mining operations should not be undertaken
unless the deposits are of such State or National
significance that their utilisation may over-ride
all other considerations, and, if developed,
mining operations should be subject to stringent
safeguards to protect the landscape and other
natural features of the environment.
The Development Plan also states:
the development of any identified mineral deposit
in scenic and remnant bush land areas should
only proceed following a full evaluation of the
benefit of removing the deposit, when balanced
against the likely environmental damage and
other potential uses of the land.
Water Resources Act 1997
The Department of Water, Land and Biodiversity Conservation is
responsible for the effective administration of the SA Water
Resources Act which encourages use of water for highest and
best return within sustainable limits. The Act seeks to ensure that
activities do not detrimentally affect another persons right to
take water from the same resource and ensures that there is no
building, destruction of vegetation or depositing of rubbish in
water courses.
The Beetaloo Reservoir reserve is located approximately 6
kilometres north of the Collaby Hill prospect near the northern
extremity of Beetaloo Valley. The catchment area in the north of
the valley is managed by SA Water Corporation whose primary
concern is the protection of water quality which involves the
preservation of the native vegetation and soil erosion control.
Native title
Part 9B of the Mining Act 1971, sets out what procedures must
be undertaken prior to the conduct of mining activities on land in
respect of which native title exists or might exist and is an
alternative State scheme to that offered under the
Commonwealth Native Title Act 1993. While much of the land
in the Southern Flinders Ranges region where MIL wished to
explore is freehold land and therefore not subject to Native Title,
the company held negotiations with the Nukunu Peoples Council
to enter into an agreement to ensure their involvement in
formulating appropriate exploration programs and ensuring
Aboriginal Heritage site avoidance.
Aboriginal Heritage Act 1988
The Act provides protection for all Aboriginal objects, remains,
sites of spiritual, archaeological, anthropological and historical
significance whether they are registered or not. Prior to
conducting exploration a proponent must liaise with the Division
of State Aboriginal Affairs (DOSAA). With respect to the
Nukunu Aboriginal community, agreement was reached between
DOSAA, the Nukunu Peoples Council and the company which
resulted in an agreed set of consultation procedures.
Native Vegetation Act 1991
This legislation provides incentives and assistance to land
owners in relation to the preservation and enhancement of native
vegetation. The Act also controls the clearance of native
vegetation and encourages reestablishment of native vegetation
where it has been cleared.
The Act is administered by the Native Vegetation Council and
provides for the Minister to enter into a heritage agreement
with the owner of the land with the objective to prevent further
reduction of the biodiversity and degradation of land and soil.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
The legislation also aims to prevent the loss of critical habitat
and to preserve and properly manage native vegetation. In the
case of mineral exploration and mining operations, removal and
destruction and reestablishment of native vegetation under this
legislation is administered by delegation to the Minister for
Mineral Resources Development and is approved under a mining
and rehabilitation program.
GEOLOGY
Regional geology
Skillogalee Dolomite is a persistent stratigraphic unit in the
Adelaide Geosyncline, and occurs over a distance of 700 km
from the Mount Lofty Ranges in the south, through to the
Willouran Ranges in the north.
Deposited in the Adelaide Geosyncline, an extensive late
Pre-Cambrian sedimentary basin, during an extensional tectonic
regime, magnesite mud precipitated in a low energy marine
lagoon/ephemeral lake, similar in nature to the current day
Coorong area in south east South Australia. The magnesite
deposits formed as interbeds within the Skillogalee Dolomite, a
wide spread sedimentary carbonate sequence within the Late
Proterozoic Burra Group sediments in the Adelaide Geosyncline.
Magnesite interbeds are concentrated in the upper part of the
Skillogalee Dolomite.
Local geology
The geological structure of the Skillogalee Dolomite in the
Southern Flinders Ranges east of Port Pirie is dominated by two
north/south trending, 30 km long, anticlines with a syncline in
between (Figure 1). The magnesite interbeds are interpreted from
the HyMapTM airborne survey to outcrop closest to the proposed
SAMAG magnesium plant site on the western flank of the
Southern Flinders Ranges at Germein Gorge in the north (15 km
northeast of the SAMAG plant site), and near Collaby Hill in the
south (15 km southeast of the SAMAG plant site) (Figure 1).
These beds have been regionally sampled at eight locations
over the 30 km strike length. At least nine magnesite beds with a
maximum width of 1.5 m have been identified near Germein
Gorge and at Collaby Hill. In the project area at Collaby Hill,
Burra Group sediments strike northwest-southeast and dip
between 60 and 70 to the northeast. Table 1 provides the
average partial leach analyses of 16 primary magnesite outcrop
samples.
TABLE 1
Average partial leach analyses.
Mg Ca Fe Mn Si S B Sr Insol Total
% % ppm ppm ppm ppm ppm ppm % %
24.9 2.1 3381 107 2097 440 95 100 6.2 99.2
Historical mining
Small scale mining has previously been undertaken at Germein
Gorge and Beetaloo. At Germein Gorge, BHP Ltd sporadically
worked Mining Leases between 1947 and 1950. Approximately
1500 tonnes of magnesite grading 25 per cent Mg were mined
from a single 1.5 m wide magnesite bed by shallow open cut
workings and an underground adit (King, 1956).
Magnesite was also quarried near the Beetaloo Homestead
between 1916 and 1966. Approximately 50 tonnes were
excavated from diggings 40 m long and 3 m deep (Crettenden,
1985).
455
C M HORN and I B MILLER
ENVIRONMENTAL ASPECTS
Following initial consultation, MDL revised their exploration
program to take into account community concerns regarding
impact on the physical, biological and human environment.
Areas of significant vegetation, eg The Wirrabara Forrest and the
Telowie Gorge Conservation Park were excluded from any
exploration activities.
As a result of concerns expressed by the local community,
MDL decided to concentrate exploration efforts in an area near
Collaby Hill about 15 km south-east of Port Pirie. The area is
hilly and well dissected with elevations ranging from 200 m to
350 m in the south and 350 to 430 in the north. Numerous
ephemeral creeks drain the southern area to the west and the
northern area into the Rocky River catchment.
Vegetation cover in the exploration area has been in part
cleared for grazing and cropping with approximately 25 per cent
of the area remaining as remnant open Mallee woodland,
comprising Eucalyptus oderata (peppermint box) and Eucalyptus
socialis (red beaked mallee) as the dominant over-storey.
The Southern Flinders Ranges contain a relatively large and
diverse range of reptiles, birds and mammals. The biodiversity
plan for the Northern Agricultural District (NAD) lists 437
vertebrate species recorded within the NAD. There are ten fauna
species which occur in the NAD listed as nationally significant,
comprising one mammal, three reptiles and six birds. The NAD
provides habitat for 313 species of birds, 12 of which are listed
as nationally threatened.
At the initial public meeting in September 2002, residents of the
Beetaloo Valley expressed significant concern regarding the
impact of mineral exploration and mining on their lifestyle, in
particular the high conservation value and sensitivity of the
Southern Flinders Ranges, the rare and endangered flora and fauna
species, and the impact on ground water resources in the area.
Concerns were also expressed about machinery noise, drilling and
blasting activities, and generation of dust, trucking disturbance,
visual impact, soil erosion, ground water contamination,
introduction of weeds and feral animals to the area and the impact
on the grazing and cropping activities in the area.
The area included within ELs 2828 and 2944 is covered by
Native Title claim SC 96/5, registered to the Nukunu Peoples
Council, and on 23 March 2001 SAMAG signed a Work Area
Clearance Agreement with the Nukunu Peoples Council with
respect to the proposed plant site and negotiated an agreement
which involved the appointment of a Heritage Survey Committee
to cover exploration activities at Collaby Hill.
CORPORATE
In 1999 MIL developed an Environmental Policy which
recognised environmental protection as an integral part of the
companys management strategies and the need to protect and
enhance the environment for the overall benefit of the
community, stakeholders and employees. MILs commitment in
environmental care and management goes beyond compliance
with existing legislative requirements and the Company is
currently benchmarking against worldwide Best Practice (ISO
14000).
MIL is committed to operating within the environmental
guidelines for mineral exploration activities in South Australia
developed by the Office of Minerals and Energy Resources
(PIRSA).
MIL also developed a Community Relations Policy (CRP)
which recognised:
the importance of active participation, mutual
trust and open relationships in providing
tangible benefits to the communities of the
regions in which the company operates.
456 Adelaide, SA, 19 - 22 September 2004
This CRP was widely conveyed to the local community and all
company employees.
MIL is firmly committed to their Indigenous Peoples Policy:
...which requires the company to develop and
maintain relationships of mutual understanding
with the Indigenous people of the area in which
they operate and to respect Aboriginal traditions
culture and heritage.
CONSULTATION PLANNING
The Port Pirie Regional Development Board (PPRDB) is an
independent incorporated body, funded by State and Local
Government to facilitate social and economic development in the
Southern Flinders region. In particular the PPRDB has a strong
focus on encouraging sustainable resource-processing
development and as such has worked closely with Magnesium
International Limited (MIL) in the development of the Smelter
project.
The PPRDB believes that all project proponents require from
the community a Social Licence to Operate (SLTO) and this
must be earned and maintained. MIL had clearly defined their
position and their desire to be an integral part of the community,
and to earn their SLTO, initiated the consultation process.
The PPRBD had previously provided substantial assistance for
MIL to interface with the local community, and in partnership
developed the Community Consultative Network (CCN). This
initial community review facility focused on the smelter project
and provided an avenue for community audit of the project. It
should be noted however that the CCN was formed after the
preparation of the smelter Environmental Impact Statement, the
development approval, and the social versus economic debate
had all but been concluded.
Through this process the PPRDB was well equipped to support
a further consultative processes and MIL approached the PPRDB
to facilitate community consultation to assist the exploration
program in the Southern Flinders Ranges. We believe that this
move by MIL to public consultation for exploration is a first for
this state and possibly nationwide.
The exploration consultation commenced at a much earlier
stage in the process than the smelter CCN. In fact the community
consultation began when very little concrete information was
available, which provided additional challenges and necessitated
a different approach. This required a great deal of information
sharing before any community evaluation could effectively be
done. In fact the critical issue was managing the misinformation
and assumptions during the early stages of formal consultation.
It was deemed at this stage that the exploration project, and
potentially mining activities, was likely to cause a higher, rather
than lower level of community outrage. The consultation model
adopted was designed on this basis to:
engage major stakeholders,
maximise project information flow,
maximise the management of anger and conflict, and
incorporate an effective complaints management system.
Consultation process
Consultation began with a well-advertised public meeting
(September 2002) chaired by the PPRDB and attended by some
110 community members. Each attendee was encouraged to
register their contact details for further follow up by MIL. An
extensive display of regional topographic maps with key
references to the exploration leases EL 2944 and EL 2828 was
provided by MIL, and supported with PIRSA displays of the
exploration process.
PACRIM 2004
 PIONEERING COMMUNITY COLLABORATION IN MINERAL EXPLORATION SOUTHERN FLINDERS RANGES
It was the intention of the meeting for MIL to assist the
community to understand the exploration process and how any
such activity might impact on current land use, community
activities, and the ecology of the region. PIRSA supported the
discussion with reference to the regulatory and supervisory role.
However, at the meeting the community simply ignored the focus
on the exploration process and went direct to a very emotive and
ill-informed discussion concerning the impact of mining.
At this stage it was apparent that a community consultation
group needed to be formed and the Southern Flinders Community
Focus Group (CFG) emerged. A Terms of Reference document
was prepared and used as working guidelines during the operation
of the CFG. The key objectives were:
provide a credible forum to achieve mutual understanding
and support through an open and honest exchange of views
on all relevant MIL related issues that impact on the
Southern Flinders Ranges;
enable MIL to gain a clearer understanding of public
perceptions, allowing MIL to prioritise genuine community
concerns;
provide the community with a clearer understanding of
MILs exploration operations and factors that influence its
business activities;
review MILs contribution to, and involvement in, the local
community and identify new opportunities to benefit the
Southern Flinders;
ensure that accurate and timely feedback from group
discussion is transmitted by its members to the community
and all stakeholders; and
share the results of research into Southern Flinders
community attitudes.
The CFG was formed by calling for Expressions of Interest
from the broader community and also soliciting interest from
specific groups such as regional councils, conservation groups,
key landowners and traditional landowners. The group members,
unlike the CCN, were geographically much more tightly
concentrated around the proposed locality and brought with them
distinct individual perspectives.
The composition of the CFG was 14 members:
eight adjacent or directly affected landowners,
two nearby residents (Beetaloo Valley),
two councillors (Port Pirie Regional Council and Northern
Areas Council),
one community group representative (Gladstone), and
one Forestry SA representative.
The process of the formation of the CFG was accomplished in
effectively six weeks and the authors believe that it was essential
to keep this as compact as possible to allow the consultation to
emerge and keep pace with the exploration agenda of the
proponent.
At the inaugural meeting it became apparent that the dynamics
of the CFG were much different to those of the CCN. Unlike the
CCN, where the level of public information and understanding
was much greater, the CFG issues often promoted vigorous
debate and the development of group maturity was slower. The
meetings were facilitated by the PPRDB with formal agenda and
recording of minutes with the inclusion of visitors as required.
The CFG met monthly from January 2003 until May 2004 when
the exploration program was suspended by MIL to concentrate
resources on activities directly related to the smelter project.
The traditional landowners, represented by the Nukunu
Peoples Council, chose not to participate in the CFG and
preferred direct consultation with MIL. Relevant issues were
amicably agreed and where appropriate conveyed to the CFG.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
Exploration program
In accordance with the conditions of the Exploration Licence,
MDL was required to submit a proposal outlining the exploration
activities to be undertaken.
Collaby Hill was determined to be a sensitive area and a full
Declaration of Environmental Factors (DEF), including external
consultation and assessment by other Government agencies, was
deemed necessary where the field exploration would involve the
use of bulldozers, excavators and drilling rigs.
The aim of the DEF is to identify the impacts the proposed
exploration activities may have on the environment and how the
potential impacts may be mitigated. The DEF is an important
document as it facilitates effective program planning and can
reduce the cost of rehabilitation on completion of the work.
MDL formulated an exploration program at Collaby Hill
which involved a two phase program of work comprising
costeaning and drilling. The exploration program was developed
with stakeholder involvement through the Community Focus
Group (CFC). The primary objective of the program was to
manage the exploration activities in such a manner as to prevent
unnecessary environmental impacts and to minimise land and
vegetation disturbance. The work program was designed to work
around seasonal land use requirements of the local agricultural
community.
Liaison with landholders
Mining companies and prospectors are able to access land for
exploration purpose as authorised by the Mining Act 1971. The
legislation clearly recognises the rights of landowners and MDL
made every effort to consult with the landowners in the Collaby
Hill area. MDL served each landowner with the standard 21 days
Notice of Entry and discussed their proposed program of work and
the likely impacts upon the land surface and agricultural activities.
MDL appointed a Liaison Officer to maintain a close liaison
with the landowner/occupier and to ensure that all company
personnel and contractors clearly understood the landowners
stock and crop management programs.
In addition to providing the landowner with comprehensive
details of the proposed exploration program, the company was
able to determine the landowners needs and concerns such as
access routes, use of water, lambing, stock movements and
habits, crop sowing and harvesting programs, noxious weed and
livestock disease and prevention, bushfire precautions and
hazards and the existence of any Heritage Agreements or
management plans or other statutory obligations.
The company formulated a Landholder Agreement which
recognised the rights and concerns of the farmers and the rights of
the explorer to determine the existence of economic mineral
deposits. Four Agreements were signed with relevant landholders.
Outcomes
Early in the consultation process it became apparent that the
level of outrage within the community was high and that this
was based on a whole range of issues, but primarily this
community had not previously been exposed to exploration or
mining activities. The Southern Flinders region is essentially a
rural agricultural area with heavy industry very much
concentrated in Port Pirie. The notion of exploration or mining
occurring in the actual Flinders Ranges was considered contrary
to existing land use and lifestyle.
The rural communities had accepted the industry located at
Port Pirie and also the proposed SAMAG magnesium smelter,
appreciating the social and economic benefits of each, but were
more averse to having operations much closer to the Southern
Flinders Ranges. Initially the social issues were more dominant
than economic ones.
457
C M HORN and I B MILLER
Representatives from PIRSA Minerals and Energy Division,
attended several meetings to outline the exploration approval
process and the regulatory function to protect the interests of all
stakeholders. The CFG were able to review the Mining Act and a
range of pertinent PIRSA data sheets and were able to walk
through the process and identify critical impact areas.
The ecological and social sensitivity of the area resulted in
MIL being required to develop a full Declaration of
Environmental Factors (DEF) as the exploration planning
submission and it was through this process that the CFG had
their major input. The CFG was able to provide a great deal of
local knowledge, supported by a local environmentalist Paul
Slattery. A more detailed and very recent study, The
Biodiversity and Natural History of Beetaloo Valley was also
supplied by the CFG. The use of this locally prepared data by
people with an intimate knowledge of the area provided
significant benefits to the development of trust and ultimate
achievement of MILs SLTO.
The draft DEF was reviewed by the CFG resulting in several
modifications and a list of some 20 issues was prepared that
required further discussion and explanation. Each of these issues
was addressed individually by MIL with input from PPRDB and
the final responses reviewed and agreed by the CFG.
Senator Meg Lees (Australian Progressive Alliance) visited the
region and expressed some concerns regarding insufficient
community involvement in both the SAMAG smelter and
exploration projects. The Senator visited the Collaby Hill site
with the CFG members and was most impressed with the depth
of the consultation processes, the collaborative DEF preparation
and the resolution of the outstanding list of issues.
The ultimate outcome was a sound workable DEF that was
approved by PIRSA, agreed by the community, and accepted by
the landholders on which the exploration leases extended. The
DEF included seasonal land use factors to suit minimal
disruption to cropping and grazing activities. In addition long
term relationships had been established, a reasonable level of
trust had been developed, and many of the community myths
regarding exploration and mining had been dispelled.
The consultation process enabled the community to clearly
understand that a licence to explore does not necessarily mean a
licence to mine and that the two processes are quite separate and
have very separate approval procedures.
CONCLUSIONS
Mineral explorers are just one of many competing land users and
mineral exploration companies should commit to stakeholder
consultation to ensure all traditional owners, environmental,
heritage, pastoral, agricultural, water conservation and other
conflicting land-use issues are taken into account prior to
formulating exploration programs.
The consultation process was intense, but only for a short
duration. The real test of effectiveness would have been
addressing issues during the costean and drilling program and
subsequent rehabilitation. The authors believe that with the
ongoing development of the maturity of the CFG, the exploration
program would have been very successfully monitored and
executed. However the CFG has been placed in recess until the
exploration program is reactivated.
458 Adelaide, SA, 19 - 22 September 2004
The authors believe it is essential to establish good relations
with land owners and to appoint a Liaison Officer to ensure that
the landowner is always notified of any entry by company
personnel to the land. A flexible field work program should be
prepared by the company that allows the proposed company
activities to work in with the land owners activities. Consultation
with the traditional Aboriginal custodians is essential in order to
avoid disturbance to any identified Aboriginal sites.
A formal evaluation of the effectiveness of the consultation
process was conducted by the PPRDB in the form of an exit
survey of members. The key points emerging were:
information sharing was extensive and essential;
contentious issues were successfully raised;
debate was extensive and often vigorous;
the ability to influence change was real;
MILs openness and honesty impressive;
MILs information was well scrutinised and tested; and
consultation is now a community expectation.
The PPRDB had dual roles managing the consultation process
in that the PPRDB had the community as stakeholders and MIL
as a client. As such, the PPRDB was keen to champion the
project from an economic viewpoint but required an open and
honest approach from MIL to ensure that the social requirements
of the community were achieved. The authors believe that
through the CFG process a win-win situation was achieved and
would strongly recommend the consultation process.
The issues raised were often contentious, however, it was
considered far better to flush these out for discussion and
resolution as far as possible. All of these contributed to a more
substantial mutual understanding of exploration activities in a
rural, ecologically-sensitive area and the ultimate achievement of
MILs Social Licence to Operate in the Southern Flinders Ranges
region of South Australia.
REFERENCES
Crettenden, P, 1985. Magnesite in South Australia: An historical review,
1915-1984, Report Book 85/62 (South Australian Department of
Mines and Energy).
Hiern, M N, Grant, I C and Horn, C M, 1978. Managing mineral
exploration and mining, in Proceedings Focus on our Southern
Heritage, 16 - 22 September, pp 56-61.
Horn, R and Biggins, S, 2002. SAMAG Project continues to move
forward, MESA Journal, 25(April):12-14.
King, D, 1956. Survey of magnesite deposits near Port Germein and Port
Augusta, Mining Review Adelaide, 99:97-112.
NSW Minerals Council, Guidelines for best practice community
consultation, in New South Wales Mining and Extractive Industries.
Office of Minerals and Energy South Australia, 2001. Environmental
guidelines for mineral exploration in South Australia, Earth
Resources Information Sheet M33.
Office of Minerals and Energy South Australia, 2002. Statement of
environmental objectives for mineral exploration in South Australia,
Earth Resources Information Sheet M34.
Philpott, V, 2003. The Biodiversity and Natural History of Beetaloo
Valley (Nature Conservation Society of South Australia).
Slattery, P, 2003. Assessment of Proposed SAMAG Exploration Sites near
Collaby Hill (Nature Conservation Society of South Australia).
PACRIM 2004
Environmental Auditing as a Driver for Continuous Improvement
in Environmental Performance
K W Perry1
ABSTRACT
Continuous improvement in environmental management assists in the
achievement of sustainability goals for the mining and processing of
minerals. The use of regular internal environmental audits (scored or
unscored) which form an integral component of an Environmental
Management System (EMS), provide demonstrable evidence of this
improvement.
The audits described in this paper involve working with site personnel
to develop and implement major improvement projects, providing advice
on how to improve environmental practices and facilities, and
disseminating innovative ideas and practices across the site and to other
operations within the same group. Essentially the lead auditor assumes
the role of a skills coach as opposed to a policeman, which is quite
often the case for many audits.
The audit process has been developed from involvement with a number
of major coal mines in the Hunter Valley, where regular internal
environmental audits have been undertaken every two to three months,
over periods varying from 18 months to five years and more recently,
with a metalliferous mine in the Northern Territory. Examples of the
benefits of improved environmental performance (driven by the audit
process) to sustainable development are provided.
The auditing process as described is applicable to metalliferous mines,
the processing of mineral ores, coal mines, coal handling and preparation
plants, quarries, and oil and gas projects.
WHY ENCOURAGE CONTINUOUS
IMPROVEMENT IN ENVIRONMENTAL
PERFORMANCE?
To date, the major thrust in environmental protection and
sustainability has been at the Environmental Impact Statement
(EIS) stage with less emphasis during the operational phase.
Sustainable development is about ensuring that development
meets the need of the present without compromising the ability
of future generations to meet their own needs (World
Commission on Environment and Development, 1987).
This paper demonstrates that continuous improvement in
environmental performance provides benefits to the sustainable
development of mine sites and associated processing to achieve
value-added products. Environmental auditing provides a measure
against what we say we are going to do, in order to achieve
sustainable development in mining and mineral processing.
WHAT ARE THE COMPANIES THAT STRIVE FOR
ENVIRONMENTAL EXCELLENCE DOING?
Companies that strive for environmental excellence have
established programs that enable the site(s) to perform at a
standard that exceeds merely conforming with regulatory
requirements and environmental legislation. Typically they have
an EMS that is either certified to ISO14001 or has been prepared
to meet the requirements of ISO14001. As part of the EMS, there
are requirements for continuous improvement enshrined in the
objectives and targets, the environmental management programs,
1. FAusIMM, Managing Director, Pacrim Environmental Pty Ltd, 2/120
Darby Street, Newcastle NSW 2300.
Previously published in Proceedings Green Processing Conference
2004, pp 169-172 (The Australasian Institute of Mining and
Metallurgy: Melbourne).
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
and the need for monitoring and measurement, and undertaking
environmental management system audits as part of checking
and corrective action.
In all cases where there is an internal, performance-based
environmental audit, there has been a substantial improvement in
environmental awareness across all personnel, from management
to operator level. This has translated into significant
environmental investment on-site. This has been achieved
through:
physical improvements (eg water management improvements,
bunding/storage of hazardous materials, waste management/
recycling, etc); and
changes to the processes used on-site (eg increasing the level
of environmental performance specified in procedures, risk
management assessments undertaken prior to carrying out
high risk activities, environmental conditions in contracts,
examination of the role of contractors and suppliers, etc).
PERFORMANCE AUDITING AND CONTINUOUS
IMPROVEMENT IS A REQUIREMENT OF
ISO14001
The following sections (International Standard ISO14001,
1996a) demonstrate the need for continuous improvement and
performance auditing on a site that is certified to ISO14001:
Section 4.2(b) Environmental Policy requires a
commitment to continual improvement and prevention of
pollution.
Section 4.5.1 Monitoring and measurement requires that
the organization shall establish and maintain documented
procedures to monitor and measure, on a regular basis, the
key characteristics of its operations and activities that can
have a significant impact on the environment. This shall
include information to track performance .
Section 4.5.4 Environmental management system audit
requires that an organization shall establish and maintain (a)
program(s) and procedures for periodic environmental
management system audits to be carried out.
Figure 1 shows the Environmental management system model
for ISO14001 (International Standard ISO14001, 1996b).
All the mine sites audited in this program have an EMS either
certified to ISO14001, 1996 or one which has been prepared to
meet the requirements of ISO14001, 1996.
THE AUDIT PROCESS
What is an audit?
An environmental audit is defined in ISO14010, 1996c as:
a systematic, documented verification process of
objectively obtaining and evaluating audit evidence
to determine whether specified environmental
activities, events, conditions, management systems,
or information about these matters conform with
audit criteria, and communicating the results of this
process to the client.
459
K W PERRY
FIG 1 - Continual improvement and ISO14001.
In a nutshell:
define what you are going to do;
use a pro-forma to ensure a systematic coverage; and
prepare a report.
Type of audit
Audits managed by Pacrim Environmental to facilitate
continuous improvement are internal audits of the Environmental
Management System involving:
1. the review of system elements (eg training, environmental
management programs, corrective action register,
achievement of objectives and targets, complaints,
incidents, inspection records, monitoring results, etc);
2. the field verification of the effectiveness of the system
elements, the observation of processes against documented
procedures, the evaluation of environmental performance;
and
3. the verification of compliance issues associated with those
areas and items audited.
Logistics
Being an internal audit with an emphasis on working with site
personnel to achieve continuous improvement, the audits are
undertaken either bi-monthly or quarterly and are normally of
one to two days duration depending on the size of the site. The
audit team comprises a consultant lead auditor and one or more
site or head office environmental personnel.
An agenda is prepared (based upon a systematic schedule that
lists all the environmental issues and system elements relevant to
the site) and is forwarded to the site normally two weeks or more
in advance of the audit. The systematic schedule is filled out
after the audit, listing the places visited, issues investigated and
system elements checked. Because the audit is of a short
duration, not all areas and issues can be investigated each audit
and therefore the systematic schedule is essential to keep track of
the audit process over a 12-month period. The frequency of
auditing issues is related to the risk rating of each
activity/location.
A comprehensive field pro-forma, which is based largely upon
the sites environmental procedures, is used to ensure systematic
coverage of the site and consistency amongst auditors.
460 Adelaide, SA, 19 - 22 September 2004
Being internal, the audits are undertaken with an emphasis on
working with site personnel, involving:
reviewing and assisting with the development and
improvement of the system elements described previously;
interviewing operators, observing operating practices;
assessing compliance issues, eg licensed water discharge
points, dust control systems on drilling rigs, etc;
providing advice on how to improve environmental practices
and facilities; and
disseminating innovative ideas and practices across the site
and to other operations within the same group.
This audit process is in contrast to an external audit, normally
undertaken at between one and five-year intervals, where a them
and us mentality often prevails and continuous improvement is
less likely to result.
Departmental focus
A departmental focus is used as this aligns with the
responsibilities empowered in the EMS. Each manager is
responsible for the environmental performance of their
department and the results of audits are reported to their teams
via communication sessions. The audit process reinforces the
environmental requirements for each department specified in the
site EMS.
Scoring versus unscored audits
The transition from unscored to scored audits at a number of
sites, combined with the departmental focus, has resulted in an
element of healthy competitiveness and an eagerness to
determine where improvements can be made. The final audit
score is comprised of a percentage score for each department
audited that are combined into an overall site score.
Being a performance-based audit, (as opposed to a compliance
audit where the majority of questions result in a yes or no
answer), scoring of internal audits tends to be largely subjective
as it is difficult, apart from a few instances, to arrive at objective
measures for scoring. There is an agreed maximum score in the
audit pro-forma for each main issue and activity audited with
scoring being based loosely upon the following criteria.
Overall scoring criteria %
Exceptional performance, industry best practice, 100%
innovative and good performance under extreme
conditions.
Good performance and meeting requirements of 80%
EMS.
Acceptable performance, some criteria not satisfied. 60%
Improvement required. 40%
Poor performance, most criteria not satisfied. 20%
Unacceptable performance, doesnt satisfy criteria. 0%
Scoring of audits can be useful for a number of reasons
(provided the degree of subjectivity in scoring is realised):
it provides a measure of continuous improvement in
environmental performance over time;
it provides a comparison of environmental performance
between departments and between mines within the same
group; and
for some sites it is used as part of the environmental
component of performance payments.
PACRIM 2004
 ENVIRONMENTAL AUDITING AS A DRIVER FOR CONTINUOUS IMPROVEMENT IN ENVIRONMENTAL PERFORMANCE

Feedback encountered on an environmental audit and some of the

A close-out meeting is held at the completion of the field
component of the audit, where a verbal report is given of
environmental performance and preliminary scores. This forum
is useful for the exchange of ideas amongst departments and
general discussion on ways of improving environmental
performance.
Reporting
The emphasis in reporting is in providing a short, well-illustrated
document in a timely manner. The report is typically six to eight
pages, of which half comprises photographs, the scoring matrix
and score chart. The text focuses on a compliance statement,
areas of good performance, areas for improvement, observations
and recommended actions the latter items being placed on the
corrective action register. A draft report is provided to the site for
review and comment with the report being finalised to achieve a
target of a week from the time of the audit.
Review
A review of the audit process is held every 12 months and this
includes evaluation of the scoring system, the audit pro-forma,
and the frequency and duration of audits.
Recent developments
A further refinement of the audit process for one group of mine
sites is the focusing on a particular issue as the underlying theme
for the audit, eg noise one audit, water management for another,
etc. This has the advantage of enabling a more detailed
investigation to be undertaken over a vertical slice of the site. To
date this methodology has provided dividends in identifying
potential exposures that had not been encountered previously
under the broader-brush coverage of the audits.
Training of personnel
An additional benefit of the audit is the progressive training of
site personnel that occurs through their role as accompanying
personnel on an audit. This training is not to enable them to be
auditors, but rather a familiarisation of the issues that are
suggestions for improvement in environmental management.
Increasing the environmental knowledge of site personnel in this
manner is the next level of training once the initial environmental
awareness training has been completed.
HOW TO MEASURE THE SUCCESS OF THE
PROGRAM
All the successes in continuous improvement in environmental
management cannot be attributed to auditing alone. Rather, it is
the success of the implementation and on-going maintenance of
the EMS, including training and auditing that result in real gains
in this area.
Some of the indicators are:
talk to the operators, superintendents and team leaders over
a period of time it will become obvious from the answers to
questions on environmental management that their
knowledge and enthusiasm is improving;
the number of innovative ideas for physical improvements
and better ways of doing things will increase, particularly
when the initiatives come unsolicited from operator level;
preventative measures, eg modification to or cessation of
activities during adverse weather conditions will prevent
complaints and the likelihood of agency prosecutions, the
bunding or secondary containment of field tanks, etc become
the standard rather than something that is considered after
the spill and this results in cost savings from not having to
clean up spills and the loss of product;
reduced waste fees from increased recycling and ensuring
that bins are not contaminated with the incorrect materials;
improvement in the level of compliance as issues are
identified on a regular basis and can be addressed prior to a
pollution event occurring or waiting until an infrequent
compliance audit is undertaken;
audit scores, as shown in Figure 2, can provide a measure of
the level of continuous improvement occurring but it must be
remembered that scoring can be subjective; and
innovative physical improvements and best practice
examples in environmental management are shared across
the site and with other mines within the group.

Typical Score Card

100%
89% 84% 85%
90%
80% 80% 81% 81% 81% 82% 83%
80%
target value
Performance Score

70%
60%
60%

50%

40%
2002 Target Score 75%
30%
2003 Target Score 80%
20%
2004 Target Score 82%
10% Actual Score
0%
A/M J /J A/S O/ D/J M/ M/ J /A O/ D/J A/M
02 02 N0 /F A0 J0 03 N0 /F/ /J 0
02 2 02 3 3 3 M
/03 04 4

Audit

FIG 2 - Typical score card.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 461

K W PERRY
CONCLUSIONS
The auditing process as described here is applicable to
metalliferous mines, the processing of mineral ores, coal mines,
coal handling and preparation plants, quarries, and oil and gas
projects.
Audits:
provide the opportunity for regular contact with departmental
managers, superintendents, team leaders and operators so
that environmental issues and improvement projects are
discussed and progressed;
identify the level of environmental performance and whether
continuous improvement is taking place;
confirm if the EMS is being maintained and is still relevant
and satisfying the needs of the site; and
identify whether compliance is being maintained.
From our experience, the number of sites adopting this internal
environmental auditing process is increasing. Scoring of audits
certainly imparts a degree of healthy competitiveness, as borne
out by the level of attendance at close-out meetings and the
enthusiasm evident in discussion.
The benefits of improved environmental performance (driven
by the audit process) to sustainable development can be seen in
areas such as:
462 Adelaide, SA, 19 - 22 September 2004
prevention of soil contamination caused by hydrocarbons and
chemicals, thereby reducing a future legacy for on-going
pollution of soil and water resources, and the costs of
clean-up;
reduced pressure on landfill space by the use of correct waste
management processes;
conservation of ecologically important areas and the
retention of faunal species through prevention of disturbance
to significant areas and the rehabilitation of disturbed areas;
and
improved amenity in the surrounding community through
better management of air quality, water quality, noise
emissions and lighting.
REFERENCES
Standards Australia and Standards New Zealand, 1996a. AS/NZS
ISO14001,1996a Environmental Management Systems
Specification with Guidance for Use, pp 2, 4, 5.
Standards Australia and Standards New Zealand, 1996b. AS/NZS
ISO14001,1996b Environmental Management Systems
Specification with Guidance for Use, p vi.
Standards Australia and Standards New Zealand, 1996c. AS/NZS
ISO14010, 1996 Guidelines for Environmental Auditing General
Principles.
World Commission on Environment and Development, 1987. Our
Common Future (Oxford University Press: London).
PACRIM 2004
Transition Issues Discovery to Profitable Operations

Sampling and Analysis Without Fear or Favour C C Brooks 465

Assessing Ore Loss and Ore Dilution Quantifying the Effects of P C Dare-Bryan and 473
Geological and Technical Uncertainties P A Dowd
Sampling and Analysis Without Fear or Favour
C C Brooks1
ABSTRACT
Sampling and analysis of mineral deposits need to be executed without
fear or favour of financial or personal impacts.
Examples are given to illustrate the effects of departures, witting and
unwitting, from correct use of appropriate techniques and/or equipment
with emphasis on exploration and mining of porphyry copper deposits of
the Pacific Rim.
Some of the results of errors on the low and high sides of true value in
exploration, resource or reserve estimations and in production caused by
sampling and/or sample preparation mistakes, faulty practices, equipment
and/or analysis are considered.
INTRODUCTION
Sampling and analysis without fear or favour are essential in our
industry, otherwise, as shown here, a few age-old problems in the
minerals industry, which some think no longer exist, become
vitally important.
Unfortunately modern technology has further obscured their
realities instead of clarifying them.
Managerial short-term accounting vision and blame-shifting
culture frequently lead to denial and inaction where budgets
become challenged. Errors that have no direct and immediately
apparent human or financial effects tend not to receive
managerial attention. Incredibly, high financial risks not
appreciated when they could result in huge cash flows foregone,
(as distinct from lost in an accounting sense) are seldom
considered, much less quantified. That is, evidence of foregone
cash flows, no matter how big or small, never shows up in the
accounts. So, commonly, no corrective and/or preventative
management decisions are taken.
We need to remember that risk recognition is as important in
this field as risk tolerance. Without experience, value judgements
are difficult to make.
Around the Pacific Rim many valuable ore deposits have
significant chalcocite and/or free gold contents. These pose
challenges at the ore reserve definition and operational phases in
sampling, sample preparation and assaying, which are commonly
misunderstood or ignored in the haste to please top management
by getting projects into production as quickly and cheaply as
expected. Incorrect resource and reserve estimates may lead to
incorrect allocation of mined blocks to millable ore, leach areas
or waste.
Let us look at some examples firstly of sampling, then of
sample preparation and finally of analysis with due
consideration of the financial impacts of allowing people without
sufficient relevant experience to control decision-making related
to each.
DRILLING AND GEOLOGICAL MODELLING
Writing on ore reserve estimation, King et al (1982) stated, inter
alia:
It will be seen that the hazards are found much
more in the realms of assumption and judgement
in the geological and mathematical fields than in
the computations.
1. FAusIMM, Exploration Consultant, 10 Winifred Avenue, South
Plympton SA 5038. E-mail: colbrx@attglobal.net
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
They further warned that geo-statistically developed answers
refer only to the available numerical data, not to the total reality
of any deposit (King et al, 1982).
Bougainville copper
King et al (1982) made the elementary observation that
Bougainvilles vertical drilling pattern was of a kind of drilling:
that in say, Broken Hill or Mount Isa, would
be regarded as useless in evaluation. and It
follows that a quick look at an ore reserve
estimate is almost worthless.
At Bougainville that drilling pattern missed small, high-grade,
vertical breccia pipes and critical vertical lithological boundaries,
giving rise, fortunately, to some pleasant surprises as the mine
developed (see Figure 1).
FIG 1 - Leucocratic high-grade vertical intrusive not detected
during 130 m spaced vertical drilling of the Bougainville porphyry
copper deposit: 20 m width of bench face illustrated.
Twenty years earlier, another of the traps described in King
good drilling data that is too sparse to truly indicate continuity
and width variability caught the company developing the CSA
mine at Cobar, NSW. The ultimate realised grade was 18 per cent
lower than the reserve of 2.7 per cent Cu, at only 2.2 per cent Cu.
The mining engineer in charge of the project would not
acquiesce to the companys chief geologists recommendation
that the deposit be opened on a trial basis to determine ore
geometry. He went ahead with development, disregarding the
advice of the mathematical statistician consulting to the company
who had warned, in a rider to his mathematical appraisal of the
results, that there was nowhere near enough drilling for the
assaying results to be used in a geometric interpretation with any
confidence where the early head grade was estimated at 2.7 per
cent Cu (Joseph, 1962).
Alumbrera
A major porphyry copper mine in Argentina, Bajo de La
Alumbrera, is another example. The board of the controlling
company, as Trevor Sykes wrote in 2003, had comprehensively
lost the plot but ten years earlier control of that deposits
development was vested in an executive very new to the mining
industry, who relied on what the author stated in 1993 to the
465
C C BROOKS
company to be misconceptions as to grade continuity propounded
by an in-house senior geologist and a very senior geological
consultant, both in the academic sphere for most of their careers,
neither of whom had sufficient relevant real world experience.
The writer commented earlier (Brooks, 2000):
Alumbrera was put into production with less than
half the minimum diamond drill hole meterage
most of us would consider necessary for
calculation of resources and reserves, and most
of the drilling done was pattern drilling.
Field inspection by the author, prior to the operating
companys financial involvement, leave alone to project
construction, showed the drill pattern to be parallel to a major
fracture set and inspection of the core showed chalcopyrite
veining parallel to core axis in many holes (see Figure 2). The
interpreted lateral continuity between holes was believed by the
author to be most unlikely at the hole spacing used thereby
invalidating the mathematics used to calculate the grade of the
resource. Although this was reported, the message was not
carried up the line for fear of upsetting higher management.
FIG 2 - Drillhole orientation parallel to a mineralisation control.
Structure has profound influence on how data should be managed
and interpreted. Native gold chalcopyrite veins (dark grey) follow
core axis orientation. Exploration drilling at Bajo de la Alumbrera
porphyry copper deposit, Argentina.
In October 1999, six years later, Alumbrera reported that
identification of a barren intrusive deleted 11 per cent
(57 Mt) from the estimated in-pit ore reserves. The full
quotation, Business News Americas, 1999, is given below.
A minority holder, Rio Algom, not the operator, carried out the
first phase of infill drilling earlier in 1999. Consequently that
company and North Broken Hill wrote down their respective
25 per cent holdings of Alumbrera, which represented many
hundreds of millions of dollars, to zero value.
The controlling company did not take a write-down. Here is
the relevant news release that ultimately undermined public and
investor confidence (27 October 1999):
Argentinas Alumbrera mining company was
surprised by a barren intrusion identified by an
infill drilling campaign at its Bajo de la
Alumbrera copper-gold mine in Catamarca
province, north Argentina, a Buenos Aires-based
company spokesperson told BNA.
Toronto-based Rio Algom (TSE: ROM), which
owns 25 per cent of Alumbrera, carried out the
work earlier in the year. The findings wiped out a
57 Mt block previously classified as ore reserves,
11 per cent of Alumbreras total.
466 Adelaide, SA, 19 - 22 September 2004
The admission goes against previous statements
made by Alumbrera spokespeople including
representatives of its 50 per cent-owner,
Brisbane-based mining company MIM Holdings
(ASX: MIM). Alumbrera previously said the
barren intrusion had been known about for some
time and taken into account in the mines open pit
plan.
The mines revised reserves stand at 460 Mt,
grading 0.53 per cent to 0.56 per cent copper and
0.65 g/t to 0.67 per cent copper (sic). Production
this year is estimated at 640 000 oz gold and
160 000 t copper.
Melbourne-based mining company North (ASX:
NBH) holds the remaining 25 per cent stake in
Alumbrera.(Business News Americas, 1999).
SUBSAMPLING
During 1999, Good Project Wrong Assays (1999), a seminar
to discuss sample preparation and assaying, was held in Sydney
as a forum at which attempts were made to strengthen the
relationship between the mineral industry and assay laboratories.
The result was a special volume (AusIMM-AIG, 1999).
Subsampling of diamond drillcore, which appears to be a
simple process, usually requires careful supervision by
experienced geologists. Core is usually halved, which introduces
unqualified, yet often significant, errors. Assaying of each half of
a section of core independently has been carried out very often;
confirming that significant variation is to be expected between
the two. Notwithstanding this, for economic reasons, as well as
to retain a complete reference library of the core, the practice of
halving core will continue.
The two most common methods are by diamond sawing and
by splitting. Lack of care in this process may vitiate the work to
that point, as seen in Figure 3.
FIG 3 - Improper halving of Nabarlek high-grade uranium ore leads
to misleading assays results.
Porphyry copper example A
Self-defeating action observed at a porphyry copper deposit
during the first phase of drilling was that the core halving
engineer put one half of the mechanically split core from each
length into a sample bag, to be sent for assay, the other half back
into the core tray and all of the fines formed during the process
into the sample bag. As the rock was splitting preferentially on
its pyrite-chalcopyrite veinlets, the fines could be expected to be
of significantly higher copper grade than the core.
The assays, however, were so much higher than the writers
visual estimate, that the trail was followed to the assay
laboratory. There it was learned that each sample had been
crushed, ground, the plus 100 mesh fraction of ground material
PACRIM 2004

sieved off and returned to the client because that is what we
always do with soil samples and the minus 100 mesh fraction
then assayed for copper.
Porphyry copper example B
To simplify and speed field operations and to save transport
costs, a senior executive geologist ordered a remote site senior
project geologist to subsample half NQ core, crushed to
nominally -6 mm, and send the subsamples to a capital city for
pulverisation and assay. Corporate policy, based on substantial
experience gained in similar projects, was never to subsample
such half cores at coarser comminution than -10 mesh (2.0 mm).
Some 20 000 samples had to be reconstituted and re-assayed,
considerably raising the cost of the project, delaying its
feasibility study and introducing yet another degree of
uncertainty. With hindsight even the half core would now be
regarded as inadequate. The deposit later went into production as
a major mine, with significant gold credits.
The project geologist opposed the instruction, was transferred
off the project and quit the company on principle. Head office
was quite disturbed at the costly poor management displayed,
dismissed the executive, his superior and his immediate
subordinate, who was by then controlling the project.
Porphyry copper example C
Blast hole cuttings piles in another very large open pit copper
mine on the Pacific Rim were noted by the author as being
sampled by a long discredited variation of a spear technique for
grade control only a few years ago (see Figure 4). Fifteen years
ago Pierre Gys Sampling Theory and Sampling Practice was
published (Pitard, 1989) in which the shortcomings of such a
method were discussed in detail.
FIG 4 - Spear sampling cannot give acceptable quality samples.
On this mine the ore mineral was dominantly chalcocite,
which is relatively soft, but quite brittle and has a specific gravity
of between 5.5 and 5.8. During drilling the chalcocite, commonly
found in fracture pattern veinlets, liberates quickly and breaks
disproportionately into grains finer than the host quartz and
quartz/kaolinite aggregates, which in this deposit have an in situ
rock specific gravity of between 1.9 - 2.0.
A metallurgical test was conducted on such ore a few years
earlier. Fresh concentrate of relatively low grade was divided into
three fractions -100 +200 mesh (-0.15 +0.075 mm), -200 +400
and -400 mesh (0.038 mm). The copper in concentrate ranged
from 28 per cent for the coarse to 45 per cent for the fine fraction
(Duyvesteyn, 1995). This illustrates the tendency of crushing and
grinding chalcocite ore to produce significant amounts of very
fine grains. During blast hole drilling, vibration ensures that the
finest liberated high-SG grains migrate toward the base of the
cuttings pile. This physical bias is further accentuated by the
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
SAMPLING AND ANALYSIS WITHOUT FEAR OR FAVOUR
disturbance of the spear (tube) being inserted. Also the spear is
unable to retrieve material from the base of the pile (due to the
geometry of an inclined tube being used to sample down to a
horizontal surface). The bias is further accentuated as the top of
each pile is made up of material from the two metres of each
hole drilled to below the bench height of 15 m. The net effect of
inherent features of the spear method and the SG contrast in the
authors experience can commonly result in a ten to 20 per cent
reduction in actual assay grade of bench samples relative to in
situ material.
Inspection of Jones type riffle splitting of the spear samples
was then undertaken by the writer. The blanking off of unequal
numbers of chutes at one end of a splitter (see Figure 5), and
their narrow widths that caught the largest chips, blocked the
flow of material, and the cavalier manner of the operators in
handling of the equipment ensured poor homogenisation. Pitard
(1989) suggests the chute width should be twice the maximum
chip dimension plus 5 mm and that there should often be a
minimum riffle width.
FIG 5 - Blanking off renders splitter self-defeating asymmetry of
the split is even worse!
Otherwise, as Pitard noted (1989), Fine particles may easily
clog the riffles and dry fine particles stick to the walls of the
riffle splitter. Riffles too narrow for the feed can result in
particles not flowing freely, or in extreme cases becoming
trapped in the splitter (see Figure 6).
Each final blast hole split of a few kilograms was then crushed
and passed through another riffle splitter which had too small a
capacity with consequent sideways chute overflows and bin
spillage where SG differentiation could cause further losses of
-400 mesh chalcocite (see Figure 7).
Use of a hopper built into a correctly sized splitter to prevent
chute overflow when properly used, would have limited the
losses.
The subsamples were then noted as being trucked to another
facility where they were pulverised in a LM5 (a swing mill or
puck-and-ring grinder), mat rolled and scooped to acquire a
250 - 300 g subsplit for analysis. The rolling ensured that a
significant portion of the chalcocite migrated to the mat surface
whence the relatively thick leading edge of the scoop was
incapable of lifting it, thereby potentially down grading the
chalcocite content by perhaps a further ten to 15 per cent (see
Figure 8).
The 250 - 300 g packets were then trucked to yet another site
to be opened and a 1 g sample scooped out from the
chalcocite-depleted top of each packet, after yet another
unwitting gravity segregation process. Those one gram
subsamples were then analysed by AA for Cu.
A 250 g subsample of this ore, grading 1.0 per cent Cu, might
have a volume of about 125 cc that contained about 0.5 cc of
chalcocite (approximately 3 g); so losing only about 0.05 cc of
467
C C BROOKS

FIG 6 - Ramming oversize through, compounding an unequal split, long-term damage as well.

fine-grained chalcocite from such a subsample, or segregating it

FIG 7 - Grossly incorrect feeding of a splitter, note overflow of
alternate chutes.
FIG 8 - Scoop-lip too thick and wrongly shaped to collect tiny free
gold, PGMs or in this example chalcocite grains, results in a
depleted grade assay.
to the base of a packet out of reach of a scoop, would lower the
grade as assayed by some ten per cent of value, or 0.1 per cent
Cu. Confirmation of part of the above was seen in one laboratory
handling this material, when a group of samples was left
overnight at -150 mesh, in metal bins. When a sample was
scooped from a bin next morning it revealed a thin layer of steely
blue fine chalcocite particles, too fine to be scooped, lining the
base of the bin, but indicating a significant downgrading of the
subsample going to assay.
The net result of these practices was that the samples were
consistently degraded from 30 per cent to 50 per cent of primary
in situ value and the chalcocite that was being left behind
cumulatively in and around crushing and grinding facilities
occasionally resulted in spectacular contamination giving up to 40
per cent grade overstatements. Using grade data that are
consistently understated by 40 per cent to 100 per cent of
laboratory-reported values has potentially profound consequences.
This problem initially had minimal impact when head grades
were four per cent Cu to seven per cent Cu and marginal grade
volumes mined were relatively small. As the edges of the deposit
were reached in the upper benches increasing volumes of
low-grade resource were being mined. Millable ore appeared to
be being sent to the leach pads and leachable ore to waste in
relatively large quantities. If, as appeared likely, the reported
grades were being grossly under reported that could lead to
over-stated mill recoveries. The metallurgists could be lulled into
not examining the plant when warranted.
The consequence of diverting millable ore to the leach pads
would be both net lower copper recovery and receipt of cash
years after what should have been sale of its contained copper in
a mill concentrate within months of mining. The waste-to-ore
ratio would also be negatively impacted by default sending
leachable ore to waste. Significant amounts of cash flow would
be forgone irretrievably, not just delayed.
When the field observations and their potential impacts on the
operation were presented to a senior staff member by the author
his computer estimation quickly demonstrated that the annual
cash flow was being depleted by tens of millions of $US.

468 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


The root cause of this was that a bold, but justifiable attitude of
rock-in-the-box at any cost was established during early
development. This had been adopted by senior and middle
management without question and maintained until long after it
should have been held. One presumes it was never considered at
board level.
As in several other operations, there has been a move here to
substituting Jones type riffle splitters with rotary splitters and to
closer training and supervision of those using subsampling
equipment.
Gold example
A gold mine for sale, then owned by a subsidiary of one of the
worlds biggest mining groups, provides another illustration of
subsampling problems. During the authors inspection of its mill,
claimed to achieve 87 per cent recovery, a metallurgist was called
to show subsample selection for assay of head feed samples.
These were of pulverised rock scooped from the top of a
mat-rolled pile of nominally -150 mesh material and taken with a
teaspoon (literally). The metallurgist was subsampling only the
most SG differentiated and depleted part of the sample.
Obviously the head sample, as assayed, was well below true head
grade, hence the over-statement of the mill recovery. Hellman
(1999) reported from one deposit that just the act of scooping
from a mound in a pulveriser could result in an average of 30 per
cent gold less than that assayed by cyanidation of the whole pulp
or by screen fire assaying of a large sample.
Alluvial gold example
Another example is of a known alluvial gold field being
re-sampled by churn drilling in the 1980s by a major mining
companys exploration arm. The damp to wet samples were
placed in plastic lined tea chests and trucked 1000 km to the
laboratory. When the assays were all virtually or truly zero the
option over the property was abandoned. A short time later the
author, while consulting to the company, asked whether the
plastic liners, which surely had all of the gold concentrated in the
residual film of sludge at the base and/or embedded into the
plastic, had been ashed and assayed because the material had
been effectively jigged for two days in transit. The liners had
been thrown away probably with all of the gold. Sometimes it
is not sampling or sample preparation nor the laboratory that is
the problem.
LABORATORY ANALYSIS
Selection of an analytical scheme appropriate to any operation
requires experience and expertise. As many published articles
and books show, too often such decisions are taken in ignorance
by people unqualified to decide. Perhaps some dare not reveal
their inability to make such judgments.
Matrix effects
A simple example is given in Brooks 2003 of gold ore being
reported at about half true value by a major contracting
laboratory, which consistently runs all its appropriate checks.
Both the sample preparation and the analytical technique
selected were inappropriate for material with high tellurium
levels and significant amounts of free gold. No amount of
replication, duplication, check assays or standards will reveal
such a management error. The laboratory had no idea that it was
asked to carry out work which would give misleading answers
and had no procedure in place to offer clients to guard against
such risks.
Subsampling
The impact of changes in laboratories sub sample size selection
is well documented in Eames (1999), and relevant graphs are
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
SAMPLING AND ANALYSIS WITHOUT FEAR OR FAVOUR
reproduced in The AusIMMs Field Geologists Manual,
(Berkman, 2001). Pertinent to the example, Eames table on his
page 81 shows clearly that a 10 g sample of less than 100 ppb Au
is too small to give a 15 per cent relative error at 95 per cent
confidence level. Pitard (eg 1989) and many others have also
published in detail on such matters. In particular Hoffman and
Dunn (2002) and Dominy et al (2002) have highlighted the need
to pay close attention to the relationships between sampling,
subsampling and assaying.
The following recent example of a laboratory managers
decision, without consultation, to summarily change the authors
INAA analysis request for gold from 30 g to 10 g aliquots shows
this. When the many hundreds of exploration samples were later
repeated at 30 g, several below detection of 5 ppb Au samples
yielded low, but measurable levels of gold. Generally there was
frequently good agreement but exceptions such as those
illustrated in Table 1 also occurred.
TABLE 1
Comparison of 10 and 30 g Aliquot INAA gold results.
Aliquot Gold (ppb)
10 g 520.0 331.0 257.0 50.9 47.9
30 g 70.5 75.0 52.8 194.0 275.0
Other elements analysed at the same time as the gold in each of
ten and 30 g, confirmed little likelihood of mixing or confusion of
samples. The variation can be attributed to fully liberated free
particulate gold, illustrated by one sample which, at 10 g reported
3 g/t Au compared with 30 g reporting 0.1 g/t Au, when a 1 g/t
result might otherwise be anticipated. Analytical testwork has also
shown that gold occurs in two physical forms in these samples.
One component of the work clearly indicated that gravity settling
occurred during transport within those 250 g samples which were
checked.
Appropriate method
Inappropriate analytical technique can have horrendous effects.
An example is given in Brooks (1999) of chromium values being
reported at three orders of level (ppm) below true values (parts
per thousand!) solely due to an incorrect digestion method being
employed. An unrelated party detected the error a few years after
the original sampling but the cause was identified by the author
many years later.
Equipment
Another unpleasant surprise was sprung on a new copper mine a
few years ago when the by-product gold values being reported at
the mine became suspect. The concentrate buyer seemed always
to be reporting gold below the mine value, but just within
splitting limits. The independence of the umpire assayers also
became suspect. The author set up pyrometric cones in the new
electric furnace used at the mine for fire assaying. Those
revealed a net 200C gradient across the furnace at the selected
operating temperature controlled by a centrally mounted
thermocouple. Samples at the front were too cool to fuse
properly while those at the rear were being volatilised by
overheating. The equipment itself was a major part of the
problem.
Discussion
Notwithstanding the above, in general, laboratory clients are
protected by laboratories managements being skilled people
who are very well aware of the need for continually monitoring
all of the steps of their processes by regular use of laboratory
in-house and commercially available standard and reference
469
C C BROOKS
material. When used with both client and laboratory in-house
and external round robins, blanks, duplicates and replicates, most
problem areas in most laboratories are quickly detected and
rectified. This does not exempt companies or individuals from
keeping close watch on all aspects within their responsibilities of
sampling, packaging, storage and transport prior to analysis.
OVERALL IMPACT OF POOR SAMPLING
AN EXAMPLE
Management of a major copper mine producing a high-grade
chalcocite concentrate relatively recently recognised the
presence of inconsistencies indicative of possibly serious
problems in its operation.
At the request of staff, several independent investigations were
carried out by senior experienced consultants, including this
author. As a result of one of his checks, to show the financial
impact of poor sample preparation, management was led through
the following steps.
1. Assume: +0.7 per cent Cu mined is to be milled;
-0.7 per cent +0.5 per cent Cu is to be
heap leached; and
-0.5 per cent Cu is to be sent to waste.
Copper produced in concentrate is sold within a few
months of production.
Copper produced by leaching is sold within several years of
having been mined.
2. Ore in an in-ground ore block has a SG of two and an in
situ grade of 0.9 per cent Cu.
3. Sample management practices deplete the chalcocite
mineral content of blasthole material by 33 per cent with
assay results of 0.6 per cent Cu for what is really 0.9 per
cent Cu in situ.
4. The 0.6 per cent Cu badged ore is therefore diverted to the
leach pad instead of the mill feed.
5. When the performance of the leach achieves the typical
70 per cent efficiency (0.63 per cent of the 0.9 per cent
primary in situ grade) an exemplary 95 per cent or better
recovery is reported but the unexpected result remains
unrecognised as indicating a problem. Mill performance is
similarly over-rated where the feed belt samples are
processed through the same or a similar sample management
system. The superintendent metallurgist is also likely to
measure his apparent success by company reward for a task
well done and miss the significance of the upgrade.
From this it is obvious that significant amounts of leachable
material of +0.5 per cent Cu grade but reported as 0.3 per
cent to 0.4 per cent Cu would be sent to the waste dumps,
resulting in an increased waste-to-ore ratio that would raise
the direct operating cost to produce copper and inevitably
modify the longer term mine design which would then
sterilise viable resources.
Medium-term reconciliation of net metal production, tail
losses, feed grades, bench-based predictions and resource
estimates should identify the existence of a problem that would
need to be resolved at the mine managers desk. Typically,
however, inconsistencies in performance data are resolved thus:
the mill blames the mine, the mine blames the geologists and
everyone is happy well sort of. It is ironic that often the mill
superintendent is responsible for analytical facilities but unaware
of the traps.
It is quite evident that weaknesses in pit floor drillhole
sampling, mill sampling practice and processing of samples can
become significant causes of mistrust between engineers,
470 Adelaide, SA, 19 - 22 September 2004
metallurgists and geologists in our example. Only a mine
manager, who proactively sets aside the cultural consequences of
vigorous pursuit of cost centre competition prior to, or to the
exclusion of understanding the causes of such problems, can ever
hope to deliver a technically sound resolution.
The consequences of poor sample handling have the potential
to derail project economics by causing mistrust between
members of the mine management team to the point where their
interpersonal politics lead to communication breaking down
often with denial of the technical evidence of a problem and the
opportunity for it to be addressed. An inadequately skilled mine
manager is thus potentially a commercial time bomb.
This model demonstrated the potential for significantly
increased unit production costs and direct loss of cash flow that
could be of sufficient magnitude as to make the difference
between very acceptable prosperity and a loss making
proposition that could lead to incorrect selection of mine size,
premature closure of an existing mine, or even worse, failure to
develop a viable deposit.
CONCLUSIONS
Many conflicts between geologists, mining engineers, primary
metallurgists and their managements have arisen through lack of
appreciation of the potential for major errors in these activities.
Indeed, serious losses, both in the millions of dollars per year of
operating costs and in misallocation of much larger amounts of
capital investment, have been identified during the past decade or
so as having been caused dominantly by ignorance of the
existence of traps in sampling and sample preparation, as well as
in assaying of samples of such deposits.
Most of those traps have been identified and quantified by
experienced operators, but the messages have not yet reached
enough senior people despite wide publication. As a result,
inadequately trained juniors, and even non-professional people
are still unwittingly involved in generating numerical data with
errors that mislead management through their lack of the
requisite skills to recognise the traps.
The mining world has a long history of similar errors to those
at Alumbrera, made by top management not having detailed
knowledge of specifics yet failing to ensure that someone in the
chain of command has such or, at least, listens to someone who
has. Usually the results are similar. There, one senior vice
president was fired and each of the three equity-owning
companies has been taken over. The same corporate malaise led
to substantial underestimation of the capital cost of putting the
mine into production. That may well have contributed to the
mess.
As Ron Manners is quoted as having said (Michael Weir,
2004): There is no excuse now for any company to have
second-rate directors on their boards. He referred to 1969 times
when: We had dentists, butchers and all your golfing mates on
the board to make up the numbers there was no expertise.
Today the problem seems more that small and big mining
companies boards have dominantly commercial, accounting and
legal expertise. With only such people there is nobody to assess
technical expertise of senior management level appointees. This
results in scenarios such as those discussed in this paper, which
might well be more damaging than the previous pervasive
ignorance at board level as misplaced overconfidence by
otherwise highly competent but mining field technically ignorant
directors has often resulted in gigantic financial losses. In
Australia not being an expert in a particular field is specifically
excluded as a legal defence for ducking corporate failure, the
author has been verbally advised, but trading off of poorly
performing assets due to inept management, for example of
assaying, cannot be consistent with duty of care to equity
holders.
PACRIM 2004

The material properties of different sets of samples require
discrete sample preparation protocols that will usually cost
significantly more than standard preparation (if truly
representative subsampling is to be achieved). The alternative is
to accept that true values may lie anywhere between 100 per cent
above and 50 per cent below those reported (Brooks, 2003).
Most primary metallurgists working on research or evaluation
projects insist on designing their own sample preparation and
analysis protocols to cope with the individual characteristics of
each material being analysed. They understand the risks of
delegating such work.
Geologists, mining and metallurgical engineers in exploration
and production are slowly realising that they, too, have to look
carefully at sample preparation protocols and analytical
techniques. Contract laboratory managers and work forces
generally have yet to grasp the importance of this spreading
understanding which normally should not be their responsibility.
Too many non-technical managers in exploration and mining
oppose the trend to recognising the importance of such protocols
or are oblivious of it.
The cost of preparing samples appropriately is usually
significantly more than that incurred by incorrect application of a
standard preparation, but a correct approach is commonly very
cost effective. The same holds true for the selection of fluxes for
fire assaying for gold.
During mineral exploration and early evaluation of a resource,
analysis is an investigative tool that will on occasion reveal the
need for more intense sampling, modified procedures, change of
drilling technology, assessment of additional elements and
determination of ore mineralogy. Application of rigid
uncompromising detailed budget constraints by executive
management may well be exceedingly foolhardy in such
circumstances.
Errors on the low side of true value caused by sampling and/or
sample preparation mistakes, faulty practices and/or analysis
may result in the following:
In exploration:
real anomalies not followed up;
low-grade anomalies sold off or abandoned; and
loss of confidence in the team and/or its techniques at both
management and field levels when and if errors revealed.
In confirmatory, resource or reserve work:
undervaluing a prospect inhibiting a prove-up stage;
undervaluing a resource, stopping or delaying study of
reserves; and
undervaluing a reserve, changing mining and/or
metallurgical approaches away from best corporate returns.
In production:
misallocation of tonnages to mill, leach, stockpile or waste;
conflicts between geologists, mining engineers and
metallurgists; and
installed plant has less capacity than needed.
Errors on the high side of true value caused by sampling
and/or sample preparation mistakes, faulty practices and/or
analysis may result in the following:
In exploration:
false anomalies followed up;
inflated stock market valuations; and
loss of confidence in the team and/or its techniques at both
management and field levels when the errors are revealed, as
they usually are.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
SAMPLING AND ANALYSIS WITHOUT FEAR OR FAVOUR
In confirmatory, resource or reserve work:
overvaluing a prospect, accelerating prove-up, or worse,
indicating that less prove-up than necessary will be sufficient;
overvaluing a resource causing misallocation of tonnages
into reserves; and
overvaluing a reserve, often causing huge misallocation of
development capital.
In production:
misallocation of tonnages to mill, leach, stockpile or waste;
installed plant has higher capacity than needed; and
conflicts between geologists, mining engineers, metallurgists,
management and financiers.
The most important part of the whole process is, as has often
been said, the original sampling. Scientific philosophy dictates
that samples should be taken to try to disprove a theory and if
they fail to disprove it, the theory is strengthened.
ACKNOWLEDGEMENTS
This paper was written at the request of consulting metallurgist
David Pollard, following encouragement to publish on this topic
over some years from Chris Oates of AngloAmerican, Chris
Benn of BHP Billiton and Peter J Rogers of Chavin Consulting.
It benefited greatly from comments, assistance and criticism by
geologist J Lindsay Curtis and from a review and severe editing
by an anonymous AusIMM reviewer, who is thanked for their
sterling efforts.
REFERENCES
AusIMM-AIG, 1999. Good Project Wrong Assays: Getting Sample
Preparation and Assaying Right! 122 p (The Australasian Institute of
Mining and Metallurgy and Australian Institute of Geoscientists).
Berkman, D A, (compiler), 2001. Field Geologists Manual, 395 p (The
Australasian Institute of Mining and Metallurgy: Melbourne).
Brooks, C C, 1999. Analysis? Which analysis, of what and why?,
Comunicaciones, 50:5-19 (Departamento de Geologa, Universidad de
Chile).
Brooks, C C, 2000. Drilling porphyry copper deposits, Business News
Americas, October.
Brooks, C C, 2003. The more they remain the same, Explore, 119:9-11.
Business News Americas, 1999. Alumbrera Spokesman Admits 57 Mt
Barren Surprise Argentina, Article.
Dominy, S C, Nopp, M A and Annels, A E, 2002. Errors and uncertainty
in mineral resource and ore reserve estimation: the importance of
getting it right, Exploration and Mining Geology, 11(1-4):77-98.
Duyvesteyn, W P C, 1995. The development of the Escondida leach
process and its implementation in the Escondida copper cathode
plant, in Proceedings Worlds Best Practice in Mining and
Processing Conference, pp 29-42 (The Australasian Institute of
Mining and Metallurgy: Melbourne).
Eames, J, 1999. Unrealistic expectations of assay results, in Good Project
Wrong Assays: Getting Sample Preparation and Assaying Right!
pp 79-96 (The Australasian Institute of Mining and Metallurgy and
Australian Institute of Geoscientists).
Hellman, P L, 1999. Issues concerning the quality of assay results, in
Good Project Wrong Assays: Getting Sample Preparation and
Assaying Right! pp 1-25 (The Australasian Institute of Mining and
Metallurgy and Australian Institute of Geoscientists).
Hoffman, E L, and Dunn, B, 2002. Sample preparation and bulk
analytical methods for PGE in, The Geology, Geochemistry,
Mineralogy and Mineral Beneficiation of Platinum-Group Elements,
Special Volume 54 (Ed: L J Cabri), pp 1-11 (Canadian Institute of
Mining, Metallurgy and Petroleum: Montreal).
Joseph, M E, 1962. Written communication to Broken Hill South Ltd.
King, H F, McMahon, D W and Bujtor G J, 1982. A guide to the
understanding of ore reserve estimation, Supplement to The
Australasian Institute of Mining and Metallurgy Proceedings, 281:21.
471
C C BROOKS

Pitard, F F, 1989. Pierre Gys Sampling Theory and Sampling Practice, 2 Weir, M, 2004. Croesus ready for $100m expansion, The West Australian,
vols, 461 p (CRC Press). February 9, p 31.
Sykes, T, 2003. Theyre mining it from Africa to Asia, The Weekend
Australian Financial Review, 9 - 10 August, p 14.

472 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

Assessing Ore Loss and Ore Dilution Quantifying the Effects
of Geological and Technical Uncertainties
P C Dare-Bryan1 and P A Dowd2
ABSTRACT
This paper describes an integrated approach to the evaluation of ore loss
and ore dilution via geostatistical simulation of in situ geology and grade
characteristics, simulation of blasting mechanisms and simulation of
loading from a blast muck pile. The approach is demonstrated on a case
study.
The sequential Gaussian method is used to simulate in situ bench
grades on the scale of the smallest selectable volume within a blast muck
pile. The simulated grades of these component volumes are interpreted as
the reality that would be available with exhaustive sampling. The
simulated bench is subjected to a blast simulator, based on the
SCRAMBLE model, which effectively moves each of the component
model volumes to its final resting place in the blast muck pile. Ore/waste
delineation and selective loading processes are applied to the simulated
muck pile to determine the degree of selectivity that can be achieved by
various sizes of loader and types of loading and to quantify ore dilution
and ore loss.
An in situ model representing the reality of knowing only the data and
information that are available from specific grade control drilling and
sampling grids is obtained by sampling the geostatistically simulated
model on a specified grid. The volumes comprising the in situ model are
then populated by estimates based only on the data and from the specified
grade control drilling and sampling grids. Different drilling and sampling
grids are used to generate different models, each reflecting the levels of
data and information available.
Selectivity is assessed as a function of the drilling and sampling grids
as well as the size and type of loader. Performance is assessed against the
ideal selectivity that could be achieved on the perfect knowledge model
comprising the simulated values of each component volume. Costs, prices
and financial criteria enable an optimal selection of the grade control
drilling grid, size of loader, type of loading and blast design.
INTRODUCTION
The critical component of most metalliferous open-pit mining
operations is ore selection, ie the minimisation of ore loss and
ore dilution during extraction. In general, extraction comprises
drilling, blasting and loading, all of which are planned and
designed on the basis of uncertain models of geology and grade.
Whilst selection processes vary according to the metal/mineral
mined and the scale and type of mining, in general they can be
described by the degree to which they depend on the blasting
stage or the loading stage. Selection by blasting rests on the
ability to delineate in situ ore and to design and implement
accurate and relatively small-scale blasts that yield separate ore
and waste piles that can be separately loaded. Selection by
loading replaces relatively small blasts by large pattern blasts, in
which the intention is to break the ground in situ; ore is then
selected from the broken rock by selective loading, often on the
basis of identifying indicative secondary features, such as
fractures or quartz veins (Dowd, 1995). In the former, the blast is
designed to heave the broken rock onto the bench so as to
provide a muck pile that can be loaded either selectively or in its
entirety. In the latter, the blast is designed to minimise heave and
maximise the shattering effect so that the broken rock falls back
as far as possible to its original in situ position.
1. Department of Mining and Mineral Engineering, University of
Leeds, Leeds, LS2 9JT, UK.
2. FAusIMM, Faculty of Engineering, Computer and Mathematical
Sciences, University of Adelaide, Adelaide SA 5005. E-mail:
peter.dowd@adelaide.edu.au
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
There are several widely used software packages that are used
to model the effects of various types of blast. There is also a vast
array of methods available for modelling the in situ
characteristics of the orebody that generally yield block grade
models, quantified geological models and geomechanical block
models. There has, however, been very little attempt to combine
these grade, geological and geomechanical models with blast
modelling to provide a link between the estimated in situ
characteristics of the orebody and the locations of the same
(displaced) characteristics following the blast. Doing so would
provide a powerful tool for evaluating different types of selection
and thereby enable planners to optimise the selection process in
terms of blast design, type and size of loading equipment,
maximisation of ore recovery and minimisation of ore loss and
dilution.
Models comprising estimated characteristics, such as grade,
geology and geomechanical properties, are, of course, based on
relatively sparse data, which limit the scale on which the models
can be provided and, for any given scale, underestimates the
variability of the characteristics. Geostatistical simulation can be
used to provide, on any required scale, models of geology, rock
characteristics and grade that faithfully reproduce the observed
variability of geological structures, geomechanical features,
grades and other characteristics. Simulation also provides a
means of incorporating a quantified risk analysis within the
planning and design procedures.
THE METHOD
In brief, the method comprises:
generation of an in situ model of the orebody comprising the
grade, geology, geomechanical properties and grade control
variables within sufficiently small volumes determined by
the smallest selectable volume within a blast muck pile;
definition of a blast volume comprising a large number of the
in situ model volumes, and subjecting it to a blast simulator,
which effectively moves each of the component model
volumes to its final resting place in the blast muck pile; and
application of selective loading processes to the simulated
muck pile to determine the degree of selectivity that can be
achieved by various sizes of loader and types of loading and
to quantify ore dilution and ore loss.
The in situ model, representing perfect knowledge at all relevant
scales, is obtained by geostatistical simulation. An in situ model
that represents the reality of knowing only the data and
information that are available from specific grade control drilling
and sampling grids can be obtained by sampling the
geostatistically simulated model on a specified grid. The
volumes comprising the in situ model are then populated by
estimates based only on the data and information corresponding
to the specified grade control drilling and sampling grids.
Different drilling and sampling grids can be used to generate
different models, each reflecting the levels of data and
information available. Selectivity can then be assessed as a
function of the drilling and sampling grids as well as the size and
type of loader. Performance is assessed against the ideal
selectivity that can be achieved on the perfect knowledge model
comprising the simulated values of each component volume.
473
P C DARE-BRYAN and P A DOWD
Applying costs, prices and financial criteria enables an optimal
selection of the grade control drilling grid, size of loader, type of
loading and even blast design.
Geostatistical simulation of orebody
characteristics
Values estimated from sparse data display significantly less
variability than the corresponding true values as illustrated by the
simple one-dimensional example shown in Figure 1.
Grade value
Location
Reality Samples Estimation
FIG 1 - One-dimensional example illustrating the difference
between true and estimated values (Journel and Huijbregts,
1978).
The solid line in Figure 1 may represent measurements of a
variable taken on contiguous cores along a drill-hole. If the only
measuremen ts that are known are those indicated by the circles,
the broken line would represent realistic estimates of the variable
at the unsampled locations. Although the estimates are unbiased
their variability is significantly different from that of the true
values. The estimated values have local accuracy (ie near the
data locations the estimates are close to reality), but they lack the
detail of the true values. Geostatistical simulation refers to
methods that sacrifice some of the accuracy for more detail by
generating sets of values that have the same statistics as the data
values, have the same spatial statistics (variograms) as the data
values and are identical to the data at the sampled locations.
Geostatistical simulation is essentially a generalisation of the
concept of Monte Carlo simulation to incorporate the spatial
characteristics of geological variables. There are many methods
available for generating geostatistical simulations (see, for
example, Dowd, 1992); the method used here is sequential
Gaussian simulation (Journel and Alabert, 1989, 1990).
Blast simulation
Most blast simulation software is based on a finite element
model, a discrete element model or a discrete block model.
Although, in principle, finite element and discrete element
models provide the most realistic means of blast simulation, the
computation times for all current implementations are prohibitive
for the necessary scale and detail of modelling, especially when
multiple runs are required to assess different grade control
drilling scenarios and strategies. For this reason a discrete block
modelling approach was preferred for the work reported here.
The discrete block model is based on the SCRAMBLE code
(Sophisticated CRA Model of Blasting with Explosives)
developed by CRA (now Rio Tinto plc) Advanced Technical
Development from the ICI SABREX code (Scientific Approach
to Blasting Rock with Explosives) (Harries and Hengst, 1977;
Jorgenson and Chung, 1987; Kirby et al, 1987; Chung and
Tidman, 1988; Mohanty et al, 1988). The code has been revised
to include, inter alia, a fragmentation model based on the Bond
Work Index.
474 Adelaide, SA, 19 - 22 September 2004
The blast simulation software uses a standard regular block
model as input to the blast simulator, which then moves each
block to its final position within the muck pile. Although the
block effectively remains intact in the muck pile it is assigned an
estimated degree of fragmentation. Movement and final position
are determined from models of the behaviour of explosive gases,
energy release, heave mechanics, fragmentation, throw and
velocity of movement as functions of, inter alia, bench height,
burden, hole spacing, hole diameter, rock density and rock
fracture density. An overview of the models used is given in the
Appendix.
This approach becomes more realistic as the block size
becomes smaller and approaches the average size of particles in
the muck pile. In principle, the block size can be made as small
as desired but in practice the size is limited by computing
constraints.
Simulating the loading process
The Floating Stope Optimiser (FSO) routine in the Datamine
mine planning software was used to simulate an optimised
selective loading process. The FSO procedure is similar to the
floating cone method of open pit optimisation and provides a
flexible means of locating optimal envelopes of block model
grades (Randall and Wheeler, 1998a, 1998b).
To apply the FSO to a selective mining operation the envelope
size is defined as the selective mining unit for the excavation of
the muck pile. The sub-cell size, which defines the grid spacing
at which the envelope is successively positioned throughout the
block model, is determined by the minimum digging width of the
excavator used.
When an excavator works through a muck pile the broken rock
continually recovers the natural angle of repose. Thus to recover
a pocket of ore near the bottom of a muck pile a cone of
material, projected up from the ore pocket, must be removed
with it. If it is difficult to distinguish between the material rolling
down over the ore and the ore itself, it can be assumed that all the
material will have to be removed as either ore or waste. To
incorporate this into the selection process a slope of 45 is
applied to the four vertical sides of the envelope cube from the
base of the envelope in the XY plane, generating an envelope
shape as shown in Figure 2.
FIG 2 - Envelope shape for Floating Stope Optimiser.
The output from the FSO flags all blocks as ore or waste.
These are then processed in Datamine to generate data on the
total tonnes mined, tonnes excavated as ore and waste, head
grade of ore and tonnes of metal in ore. Multiple runs are taken
for each muck pile over a range of cut-off grades to find the
optimum.
PACRIM 2004

Optimisation procedure
A blast design is applied to the complete geostatistically
simulated blast volume (the reality) and to the estimated block
models for the blast design. Once the block models have been
heaved to generate the corresponding muck piles, the muck pile
block models, with associated block grade values, are entered
into the FSO to evaluate the ore/waste excavation boundaries to
give the optimum head grade based on a selected cut-off grade
and selective mining unit size. The region of the bench that is to
be excavated as ore is evaluated on the basis of the total tonnes of
metal/mineral within that region minus the portion of
metal/mineral expected to be lost in the processing operation.
The 80 per cent passing size of the resulting muck pile (see
Appendix) is then used to adjust the standard cost per tonne
values for the down-stream processes of loading, hauling and
primary crushing. The total mining cost for the bench comprises
the drilling and blasting costs derived from the blast design, the
revised loading and hauling costs and the mining services costs,
all as a cost per tonne blasted. The total processing cost
comprises the adjusted primary crushing costs and the remaining
processing operations costs, which are expressed as a cost per
tonne processed.
The value of the bench is thus the value of the concentrate
output from the processing plant less the mining and processing
costs.
CASE STUDY
The case study is based on the Minas de Rio Tinto S.A.L. (MRT)
open pit copper mine at Rio Tinto, southern Spain, which is
typical of a low-grade operation in the later stages of its life. The
application described here is to the low-grade Cerro Colorado
stockwork mineralisation. Ore/waste delineation for selective
mining is particularly difficult because the head grades are near
the economic cut-off grade and there are no clear geological
controls on the mineralisation within the host stockwork.
The mining operation has been temporarily closed pending an
increase in the copper price. During operation the mine produced
concentrate with an average grade of 24 per cent copper.
Mining method
The operation at MRT used traditional drilling and blasting on
10 m and 12 m benches that were drilled with two Buycrus Eyre
45R rigs and one 60R rig drilling 250 mm holes to a depth of
11.2 m or 13.7 m depending on the bench height. A square blast
pattern was employed with burden and spacing dimensions
ranging from 5.5 m 6.5 m to 6.6 m 8.0 m. The holes were
charged with heavy ANFO because of water problems in the
FIG 3 - Experimental semi-variograms and two-structure spherical models for B4053 (left) and B4065 blasthole data.
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
ASSESSING ORE LOSS AND ORE DILUTION
lower benches. P&H 2100 BL electric face shovels and
Caterpillar 994 wheel loaders were used for loading and
Caterpillar 789 dump trucks used for hauling.
Two blasts, B4053 and B4056, were selected for this study.
Generating block models
Experimental semi-variograms were calculated from the
blasthole data using a conical search. As no significant
directional anisotropies were detected within the two blast
volumes all directional semi-variograms for each blast were
combined into a single omni-directional semi-variogram for
modelling purposes. For both blast volumes a two-structure,
spherical semi-variogram model was fitted to the experimental
semi-variograms as shown in Figure 3.
Sequential Gaussian simulation, with the blasthole grades as
conditioning data, was used to generate a realisation of the entire
bench on a block grid of 0.5 m 0.5 m 0.5 m, the grid
determined on the basis of blasting, selection and loading criteria.
The simulation provides a realisation of the grade distribution
throughout the bench on the scale required for the blast
simulation. For each specified blast design, new sample hole
data are taken from the simulation block model of the bench.
These sample data are then used to generate ordinary kriging
estimates of the block grades to produce an estimated block
grade model of the bench. The semi-variogram used for kriging
is the model fitted to the experimental semi-variogram of the
sample data taken from the simulation block model.
Blast modelling parameters
The simulated heaving action and muck pile generation were
adapted to replicate the muck piles generated by the actual blasts,
based on the data available for throw and the overall shape of the
muck pile profile. The blast pattern specifications for the two
blasts used in this study were:
Burden 6.5 m
Spacing 8.0 m
Bench height 12 m
Vertical blast hole length 13.7 m
Hole diameter 250 mm
Main explosive charge 540 kg ANFO
Initiation sequence S1
Inter-hole delay 50 ms
Inter-row delay 100 ms
475
P C DARE-BRYAN and P A DOWD
The geomechanical data used were:
Youngs modulus 750 kbars
Uniaxial compressive strength 1.2 kbars
Poisson ratio 0.25
Rock density 2.75 g.cm-3
The modelling was calibrated against the original blast designs
for the two benches using these input data and the muck pile
profiles from the field data.
Selection of ore/waste boundaries in muck piles
For the excavators used at MRT, with a bucket size of 13 m3, an
FSO envelope size of 8 m 8 m 8 m was used with a sub-cell
size of 2.7 m. More selective loading was also assessed using a
6 m 6 m 6 m envelope.
Cost of the blasting and selection processes
For a given blast design it is relatively straightforward to
calculate the costs associated with drilling and blasting, by
summing the constituent costs. However, the composition of the
muck pile produced by the blast directly affects the down-stream
processes of loading, hauling and primary crushing, and the
overall cost evaluation of a blast must include the costs of these
processes. It is not possible to quantify directly the effect of
different quality blasts on the down-stream processes and the
best common variable for comparisons is the degree of
fragmentation achieved by the blast. It is generally recognised
that the costs of the down-stream processes, including operating
and maintenance, decrease as fragmentation improves
(MacKenzie, 1966). A common practice is to use a functional
relationship, formulated through in-pit operational assessment, to
adjust the cost per unit weight worked as a function of the degree
of fragmentation. To simplify calculations all process costs are
calculated as cost per tonne worked.
Drilling costs
Drilling costs are expressed as a cost per metre drilled (DCm) for
the 250 mm hole diameter used in this study. The tonnage of
rock associated with each blasthole, taken as a standard for a
specific hole pattern, is given by Equation A8, divided by 1000
to give tonnes. Cost per tonne (DCt) is then:
DC m HL
DC t =
M
where:
HL is the hole length (m), including subdrill
Blasting costs
The initial costs are calculated for a single hole and are divided
into fixed costs per hole booster, detonator, surface connection
and manpower costs and the variable cost of the main charge
placed in the hole. The main charge costs (EXm) are calculated
from:
EX m = EX (Ah EC l ) e (2)
where:
EX is the cost of the explosive ($/kg)
2
Ah is the cross-sectional area of the hole (m )
ECl is the charge length of the explosive in the hole (m)
e is the density of the explosive used (g.cm-3)
476 Adelaide, SA, 19 - 22 September 2004
Loading costs
The loading costs for the original blast design are taken as
$0.14/tonne for a muck pile with 80 per cent passing size of 0.5 m.
Within reasonable limits, as passing size decreases loading costs
decrease due mainly to an increase in the ease of digging, which
leads to faster loading rates and reduced maintenance costs.
MacKenzie (1966) reports a linear relationship between cost per
unit loading and degree of fragmentation for Quebec Cartiers
16-D iron ore mine. Van Zeggeren and Chung (1975) found that
their data followed a square root relationship and Nielsen (1983)
used a variable exponent selected by the user. For this application,
with too few operational data to derive an appropriate relationship,
the linear equation:
C l = (D80 ) + ( SC l ) (3)
was used where:
Cl is the adjusted loading cost ($/tonne)
D80 is the calculated 80 per cent passing size (m) using
Equation A14
SCl is the standard loading cost ($0.14/tonne)
, are constants
The inclusion of the standard loading cost (SCl) in Equation 3
allows the loading cost relationship to be adjusted for different
loaders with different attributes.
Haulage costs
Haulage costs also decrease with muck pile particle size because the
truck is more completely filled, providing the ore density allows it.
The relationship used for haulage costs (Ch) in $/tonne is:
C h = e D 80 (4)
where:
is a constant
Primary crushing costs
Because variations in feed size to the primary crusher affect
power costs much more than general maintenance and plate
replacement costs, the Bond Index Equation A9 was used to
calculate crushing costs:
(1)
1 1
C cr = 241.4 (5)
16300 1 / 2 D80 1 / 2
where:
Ccr is the adjusted crushing cost ($/tonne)
is a constant
Costs unaffected by blasting practices
Costs incurred in producing a saleable product that are not
affected by blasting practices include mining services and the
entire mineral processing operation down-stream of the primary
crushing. These values, also expressed as $/tonne, are assumed to
remain constant.
Results
By way of example, Figure 4 shows colour-coded simulated
grades of sections of the 0.5 m 0.5 m 0.5 m blocks that
comprise bench B4056 and Figure 5a shows the muck pile
generated by applying the blast modelling process to this bench.
PACRIM 2004
 ASSESSING ORE LOSS AND ORE DILUTION

FIG 4 - Representations of the simulated in situ bench grades for B4056 showing colour-coded grade ranges on (A) horizontal planes and
(B) cross-sectional planes. Horizontal planes are top and bottom of 12 m bench and 6 m mid-plane. Vertical planes are extremities (0 m and
80 m) and intermediate planes at 26 m.

Figure 5b shows the muck pile that results from applying the
blast simulation to the same bench but with the component block
grades kriged from the simulated grades on the 6.5 m 8 m
drilling grid. The smoothing effect of kriging is clearly evident
when comparing Figures 5a and 5b. Figure 5a represents the
muck pile given complete information whereas Figure 5b is the
interpretation of the composition of the muck pile on the basis of
the data. Selection is planned and implemented on the basis of
Figure 5b but the volume selected will have the grade and
tonnage of the equivalent volume in Figure 5a.
Figure 6 shows the corresponding muck piles generated from
simulated and estimated block grade models for B4053. Figures
5 and 6 clearly show the significantly different spatial
distribution of grades in the two muck piles with consequent
implications for selection.
By way of example, when selection is applied, via the FSO, to
the two muck piles shown in Figure 6 the volumes selected are
those shown in Figures 7 and 8.
For each bench there are nine block models: the simulated
block grades, taken as reality, and eight models of estimated
block grades kriged from simulated values on various drilling
grids together with variations in other blast design parameters as
summarised in Table 1.
The grades of the blocks that comprise the two benches are
similar in terms of histograms but they differ significantly in
their spatial distributions within the respective benches. It is the
latter that has the major effect on the spatial distribution of the
grades in the muck pile and consequently on the ability to load
selectively.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 477

P C DARE-BRYAN and P A DOWD

FIG 5 - B4056: muck piles generated by blast design number one (A) from simulated bench grades and (B) from kriged bench grades.

Bench B4053 is sub-economic for some blast designs but must
still be blasted to allow continuing mine development. Having
blasted this bench, any losses are minimised by processing the
ore in the muck pile. Bench B4056 is economic for all blast
designs and is mined and processed in the normal manner.
The financial performances of each blast design against the
reality of the simulated block model are summarised in Figure
9 for B4053 and in Figure 10 for B4056. These figures show the
ideal, or maximum, bench values corresponding to the simulated
block grades together with the actual bench values achieved by
selecting from the muck piles generated from the estimated block
grades for the various blast designs.
Figures 11 and 12 show the tonnages of copper within the ore
selected from the muck pile generated from the simulated block
grades together with the actual tonnages recovered from the
muck piles generated from the estimated block grades for the
various blast designs.
Numbers on the horizontal axes of Figures 9 - 12 denote the
blast designs given in Table 1. Blast 1a (smaller FSO envelope) is
a smaller selection envelope applied to blast one, in which the
envelope corresponds to smaller-scale selection (6 m 6 m
6 m) using a wheel loader.
Note that in some cases more copper is recovered from the
muck pile generated from the estimated block grades than from
the muck pile generated from the simulated block grades (eg
blast designs 7 and 8 in Figure 10). This is, however, at the
expense of diluting the ore with additional waste, which reduces
profit (eg as indicated by the bench values for blasts seven and
eight in Figure 8).

478 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 ASSESSING ORE LOSS AND ORE DILUTION

FIG 6 - B4053: Muck piles generated by blast design number one (A) from simulated bench grades and (B) from kriged bench grades.

The differences between ideal selection and selection based on
estimated block grades are more significant for B4053 because
the economic grades are more widely dispersed through the
bench and the muck piles than they are for B4056. The
differences are large and critical for B4053 as planning on the
basis of the estimated block grades leads, more often than not, to
financial loss.
The real effects on the operation can be quantified by
comparing expected performance against the actual performance.
Figures 13 and 14 show, for each blast design and for selection
based on estimated block grades, the difference between the
estimated and actual copper content of the selected ore regions
together with the difference between the estimated and actual
financial values of the selected ore regions. It is these differences
between planned and actual performances that have the greatest
impact on the viability of the operation.
The results summarised in Figures 13 and 14 are functions of
the complex relationships among block grade values, heave
mechanics of the blasting process, the spatial distribution of ore
and waste blocks in the muck pile and the method of selecting
from the muck pile. The absolute values of the bars shown in
Figures 13 and 14 are the deviations from planned outcomes and
are measures of the ability to plan the operation to acceptable
levels of accuracy and of the consequences of not being able to
do so. The larger differences for B4056 (Figure 14) are a
function of the more distinct ore/waste boundaries in the
resulting muck pile, which in turn provide a greater propensity
for ore loss and ore dilution with small changes in the selection
volumes. By contrast, the greater dispersion of the ore
throughout the muck pile generated from B4053 offers less scope
for selectivity and less adverse consequences arising from
changes in the selection volumes.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 479

P C DARE-BRYAN and P A DOWD

FIG 7 - Muck piles for B4053. (A) muck pile generated from simulated block grades (reality); (B) muck pile generated from estimated block
grades using blast design one. Red indicates exposed selected ore and grey is non-selected broken rock.

SUMMARY AND CONCLUSIONS
This study demonstrates the effectiveness of an integrated
approach to quantifying the effects of geological and technical
uncertainty on ore loss and ore dilution and hence on
profitability. The methods presented here provide a means of
quantifying and managing the risk associated with mine planning
based on specific amounts of sample data and specified technical
designs. They also provide a means of optimising blast design to
maximise selective loading of ore.
Geostatistical simulation is used to quantify the effects of
estimated (smoothed) grade distributions on blast design and the
selection of ore regions within the resulting muck pile. Blast
simulation is applied to the in situ geostatistically simulated
grade distributions to generate muck piles from which ore
regions can be delineated and loaded at various scales.
Although a very specific blast modelling process has been
used in this study it could readily be replaced by any other type
of modelling either to provide a more realistic simulation of
heave mechanics and fragmentation or to simulate other types of
blasting and selection. Similarly, other types of geostatistical
simulation could be used and multiple variables, including
qualitative geological variables, could be simulated and
incorporated into the selection procedure (eg selecting
gold-bearing ore on the basis of observable quartz veins and
fracture networks in the muck pile). The methods and approach
used in this study do not limit the generality and practical
potential of the application.
A real-time, virtual reality version of the approach described
here could also be used to guide loader operators in making
optimal selections from muck piles. Real-time applications
would require very rapid capture of accurate survey and
locational data, which could readily be provided by GPS.

480 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 ASSESSING ORE LOSS AND ORE DILUTION

FIG 8 - Volumes of ore selected from muck pile for blast design one generated from (A) simulated block grades and from (B) estimated block
grades.

TABLE 1
Blast designs used in study for estimated block grades.
Blast Design Burden Spacing Powder Hole
design changes (m) (m) factor diameter
-1
(kg.tonne ) (m)
1 Control 6.5 8 0.31 0.25
2 Changing hole 6 7.5 0.31 0.23
3 diameter 7 9 0.31 0.27
4 8 9.5 0.31 0.30
5 Increasing 6 7.5 0.37 0.25
6 powder factor 5.5 7 0.43 0.25
7 Decreasing 7.5 9 0.25 0.25
8 powder factor 8.5 10.5 0.19 0.25

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 481

P C DARE-BRYAN and P A DOWD

Simulated Estimated
20000
10000
0

Bench Value ($)

-10000
-20000
-30000
-40000
-50000
-60000
-70000
-80000
1 2 3 4 5 6 7 8 1a
Control changing hole diam increasing PF decreasing PF smaller
FSO
envelope
blast design

FIG 9 - B4053 optimised bench values for blast designs.

Simulated Estimated
90000
80000
Value of bench ($)

70000
60000
50000
40000
30000
20000
10000
0
1 2 3 4 5 6 7 8 1a

Control changing hole diam increasing PF decreasing PF smaller

FSO
blast design envelope

FIG 10 - B4056 optimised bench values for blast designs.

Copper actually recovered Simulated copper in ore

380

340
Copper (tonnes)

300

260

220

180

140

100
1 2 3 4 5 6 7 8 1a

Control changing hole diam increasing PF decreasing PF smaller

FSO
envelope
Blast Design

FIG 11 - B4053 simulated (actual) copper in ore selected from muck pile and amounts recovered
on the basis of estimations from various blast designs.

482 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 ASSESSING ORE LOSS AND ORE DILUTION

Copper actually recovered Simulated copper in ore

360
340

Copper (tonnes)
320
300
280
260
240
220
200
1 2 3 4 5 6 7 8 1a

Control changing hole diam increasing PF decreasing PF smaller

FSO
envelope
Blast Design

FIG 12 - B4056 simulated (actual) copper in ore selected from muck pile and amounts recovered
on the basis of estimations from various blast designs.

expected recovered copper less amount recovered

expected bench value less real value
30 33000
25 28000

20 23000
18000
15
Copper (tonnes)

13000
10

Value ($)
8000
5
3000
0
-2000
-5
-7000
-10 -12000
-15 -17000
-20 -22000
1 2 3 4 5 6 7 8 1a

Control changing hole diam increasing PF decreasing PF smaller

FSO
envelope
Blast Design

FIG 13 - Differences between planned and actual performance for B4053.

expected recovered copper less amount recovered

expected bench value less real value
30 33000
25 28000
20 23000

15 18000
Copper (tonnes)

13000
10
Value ($)

8000
5
3000
0
-2000
-5 -7000
-10 -12000
-15 -17000
-20 -22000
1 2 3 4 5 6 7 8 1a

Control changing hole diam increasing PF decreasing PF smaller

FSO
envelope
Blast Design

FIG 14 - Differences between planned and actual performance for B4056.

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 483

P C DARE-BRYAN and P A DOWD

REFERENCES The gaseous detonation products flow into the fractured rock
mass at the local speed of sound until the gas vents through a free
Chung, S H and Tidman, J P, 1988. Effective modelling for cast blasting, face; at this stage a rarefaction wave travels back toward each
in Proceedings International Symposium for Mine Planning and blasthole decompressing the cracks. As the rarefaction wave
Equipment Selection, Singhal, pp 357-360 (A A Balkema).
travels through the rock the pressurised crack system imparts an
Dowd, P A, 1992. Recent developments in geostatistics, Computers and impulse, which heaves the broken rock mass out from the bench.
Geosciences, 17(10):1481-1500.
In generating the muck pile empirical routines are used to limit
Dowd, P A, 1995. Bjrkdal gold mining project, northern Sweden, Trans
Instn Min Metall, Section A: Mining Industry, 104:A149-A163. the angle of repose whilst producing a smooth surface and
adding swell factors.
Harries, G and Hengst, B, 1977. Use of a computer to describe blasting,
in Proceedings 15th APCOM Symposium, pp 317-324 (The
Australasian Institute of Mining and Metallurgy: Melbourne). Equation of state for explosive gases
Hustrulid, W, 1999. Blasting Principles for Open Pit Mining, Vol 1
General Design Concepts. (A A Balkema: Rotterdam, Netherlands).
The equation of state for the gaseous products of detonation is:
Jorgenson, G K and Chung, S H, 1987. Blast simulation surface and
E(1 + )
3
underground with the SABREX model, CanIMM Bulletin, 80:37-41.
p= (A1)
Journel, A G and Alabert, F, 1989. Non-gaussian data expansion in the 100(1 + 2)
earth sciences, Terra Nova, 1:123-134.
Journel, A G and Alabert, F, 1990. New method for reservoir mapping,
Journal of Petroleum Technology, February:212-218. where:
Journel, A G and Huijbregts, C, 1978. Mining geostatistics, 600 p p is the gas pressure in kbars
(Academic Press: New York).
Kirby I J, Harries, G and Tidman, J P, 1987. ICIs computer blasting is the gas density in g.cm-3
model SABREX the basic principles and capabilities, in
Proceedings 13th Conference on Explosives and Blasting Technique E is the available energy in J.g-1
(Ed: R D Boddorff), pp 184-198 (Society of Explosives Engineers).
and are dimensionless constants
Leiper, G A and Plessis, M P, 1987. Describing explosives in blasting
models, in Proceedings 2nd International Symposium on Rock The available energy E is the work done by the explosive gases
Fragmentation by Blasting, (Eds: W L Fourney and R D Dick), pp in expanding adiabatically from the density to ambient
462-474 (Society for Experimental Mathematics). conditions, and is obtained from:
MacKenzie, A S, 1966. Cost of explosives do you evaluate it properly?
Mining Congress Journal, pp 32-41. ()2 + 5 ( )2 + 5 1 (1 + 2 )
E
Mohanty, B, Tidman, J P and Jorgensson, G K, 1988. Advanced computer ln = + 1n (A2)
simulations the key to effective blast designs in open pit and Eo 4 4 8 (1 + 2)
underground mines, in Computer Applications in the Mineral
Industry (Eds: Fytas, Collins and Singhal), pp 41-48.
Nielsen, K, 1983. Optimisation of open pit bench blasting, in Proceedings
where:
1st International Symposium on Rock Fragmentation by Blasting, o is the initial gas density after detonation (equal to the
Lulea, Sweden, pp 653-664 (Society for Experimental Mechanics). explosive density)
Pryor, R N, Rhoden, H N and Villalon, M, 1972. Sampling of Cerro
Colorado, Rio Tinto, Spain, Trans Instn Min Metall, 81:A143-159. Eo is the initial available energy
Randall, M and Wheeler, A, 1988a. Balancing the books, Mining The values for Eo, and can be obtained from an ideal or
Magazine, pp 337-342. non-ideal detonation model. An ideal detonation model is
Randall, M and Wheeler, A, 1988b. Where did it go? Mining Magazine, adequate for the large diameter holes used in this study; more
pp 245-249. accurate data could be obtained from non-ideal models such as
Van Zeggeren, F and Chung, S H, 1975. A model for the prediction of CpeX (Leiper and Plessis, 1987).
fragmentation, patterns and costs in rock blasting, in Proceedings Equation A1 reduces to the ideal gas law for small gas
15th Symposium on Rock Mechanics, South Dakota, USA, pp
557-569.
densities and, together with Equation A2, allows available energy
and pressure to be generated as a function of their density during
the expansion process.
APPENDIX: BLAST MODELLING
The adapted version of the SCRAMBLE/SABREX blast Heave mechanics
modelling code used in this study is an energy-based approach All regions within the gas envelope have a common gas density
comprising two separate models: heave mechanics and and pressure. The leading edge of the envelope is regarded as the
fragmentation. The heave mechanics are based on the energy gas front, which is assumed to move at the local speed of sound
released from the adiabatic expansion of the explosive gases (m.s-1) given by:
following detonation. Fragmentation is based on the powder
1
factor (ratio of charge weight in kg to mass in tonnes of rock
100000 p 2
broken by the charge) converted to an energy equivalent via the c= (A3)
Bond Index.
The velocity of detonation for each blasthole is taken as
infinite and the wall is allowed to expand until it reaches a state where:
of equilibrium determined by the isotropic expansion
is the adiabatic exponent for the gases at pressure p
characteristics of the quasi-static gas pressure and the elastic
(kbar) and the density (g.cm-3), is given by:
resistance of the rock. The expanded blasthole sets up hoop
stresses in the surrounding rock, creating a system of radial
cracks that, because of tensile failure, spread away from the hole. 1 + (1 + )3 3
= + (A4)
The radial fractures, together with any pre-existing geological 1 + 2 1 +
discontinuities, define the damage created in the rock mass by
the blast.
and is derived from the equation of state given in Equation A1.

484 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004


To calculate the necessary density and pressure of the gas
within the envelope the volume of rock within the envelope is
assumed to be in a state of hydrostatic compression at pressure p.
The resultant reduction in the volume of rock is given by:
Vp
V = (A5)
G
where:
V is the initial volume (m3)
G is the bulk modulus
V is the volume increase in the envelope contributing to
the reduction in gas density and pressure.
Another small increase in volume is associated with the gas
pressure compressing the rock below and behind the blasthole.
As the gas expands with the moving gas front the local speed
of sound in Equation A3 falls and a time-stepping loop is used to
track the expansion of the gas. The time steps used are defined
by:
b + b
t = (A6)
c R
where:
b+b is the equilibrium blasthole radius
Equation A6 shows that, although the time steps can vary, the
corresponding spatial steps are constant and equal to the
equilibrium borehole radius.
The time-stepping procedure is:
1. calculate the initial local speed of sound from Equations A3
and A4 prior to the expansion of gas into the rock mass;
2. calculate the appropriate time step from Equation A6 and
generate the appropriate gas front profile;
3. calculate the increase in volume from Equation A5, and
then calculate the new gas pressure and density using
Equations A1 and A2;
4. re-calculate the local speed of sound using Equations A3
and A4; and
5. repeat the steps while keeping track of the total elapsed
time.
Venting of the explosive gas begins when the gas front meets a
free face. At that time the gas fronts retrace their original paths
and, during this period of contraction the gas density, pressure
and speed of sound are assumed to be constant within the volume
of the gas envelope. The respective constant values are those that
were calculated at the time of venting, while the pressure beyond
the gas fronts is assumed to be insignificant.
At the time of venting, the rock mass is assumed free to move,
reacting to a momentum impulse that is imparted on the rock
mass. The calculated impulse is based on the assumption that the
rock mass does not start to move until the gas fronts have
contracted. The total impulse imparted on the rock mass
(kg.ms-1) is given by:
t0
p(t )A(t)dt = M .v
8
10 v (A7)
tv
where:
tv is the time (s) at which gas venting takes place
PACRIM 2004 Adelaide, SA, 19 - 22 September 2004
ASSESSING ORE LOSS AND ORE DILUTION
to is the time (s) at which the contracting gas fronts reach
their blastholes
p(tv) is the gas pressure (kbar) in the gas envelope
A(t) is the area (m2) over which the pressure is applied
M is the mass (kg) associated with each blasthole
v is the velocity (ms-1) with which the rock mass is
heaved
t is the time (s)
To derive heave velocities from Equation A7 an expression for
M can be applied for a vertical free face to calculate the mass of
rock associated with each blasthole using:
M = B. S . H. R .1000 (A8)
where:
B is the burden (m)
S is the hole spacing (m)
H is the bench height (m)
is the rock density (g.cm-3)
In practical situations the highwall of a bench is not vertical
and the program has an input variable for face angle to calculate
the true mass of rock associated with the first row of holes.
The momentum impulse for each blasthole is resolved into the
vertical and horizontal directions on the basis of the areas
defined by the gas envelope. For the vertical impulse the area at
the base of the envelope is used in Equation A7. However, due to
the angled highwall, the front row has an inconsistent burden and
the area is taken as an average of the areas at the top and bottom
of the explosive column length.
Two impulses are computed in the horizontal direction. The
first is the section of rock between the toe of the bench and the
top of the explosive column, and the second impulse is the region
at the top of the bench where the blasthole is filled with
stemming material.
A similar averaging process is used to account for the effect of
the front row of holes in the calculation of the horizontal
impulse, which results in three horizontal heave velocities
defining the heave velocity profile. On subsequent rows the
effective free face is assumed to be vertical.
For the heave action the blocks comprising the block model
are treated sequentially within a time-stepping loop using a raster
pattern starting at the toe of the bench with priority given in
order to z, x, and then y. For each run through the time-stepping
loop all block positions and velocities are recalculated from
ballistic trajectory equations and the revised values are stored in
three-dimensional arrays; in-flight interactions with other blocks
are not modelled. Each block remains in the time-stepping loop
until it travels to a point in space at which, ahead or below it,
another three-dimensional array describing the mine floor has a
positive value, defining that volume of space as containing a
block.
When a block drops out of the time-stepping loop to form part
of the muck pile it immediately comes to rest on the ground and
becomes part of the array that defines the floor and the
developing muck pile. The input value for maximum angle of
repose ensures that if the defined angle is exceeded in the
generation of the muck pile then the block is moved down the
surface of the muck pile until it reaches a point of stability.
When all blocks have come to rest, swell is applied to the
muck pile by raising each block by a pre-defined factor
proportional to the change in vertical height the block underwent
in moving from the bench to the muck pile.
485
P C DARE-BRYAN and P A DOWD

Fragmentation 1 1 KB
Wi = K B 1/ 2 = (A12)
The Bond Index equation from comminution theory is used to 100
1/ 2
100 1 / 2
assess the effect of different blasting practices on the degree of
fragmentation resulting from a blast (Van Zeggeren and Chung, Substituting the calculated KB in Equation A12 gives:
1975 and Nielsen, 1983). The equation relating energy input to 241.4
degree of comminution is: Wi = = 24.1 kWh.tonne-1
100 1 / 2
1 1 From Equations A11 and A12 it is possible to calculate the
W = K B 1/ 2 1/ 2 (A9)
P F energy required to reduce material from an infinite size down to
the desired 80 per cent passing size as:
where: 1/ 2
100
W in kWh.tonne-1, is the energy input to a machine Wt = Wi (A13)
reducing particle size P
F is the feed size, measured in microns (10-6 m), and
defined as the mesh size of a screen that allows 80 per If it is assumed that the only factor that influences the degree
cent of the material to pass of fragmentation in blasting is the amount of energy imparted to
the rock mass and that the energy distribution and initiation
P is the product size in microns also at 80 per cent variables can be ignored, then Equation A13 should give a good
passing representation of the energy input from the explosive in a blast,
KB is a constant determined for a specific feed material based on the resulting fragmentation.
For the 6.5 m 8 m MRT blast designs the material in the
The constant KB is determined by rearranging the Equation
resulting muck piles had an 80 per cent passing size of
A10 to give:
approximately 0.5 m. From Equation A13 the energy imparted
by the explosive is:
P1/ 2F 1/ 2
K B = W 1/ 2 1/ 2
(A10) 1/ 2
F P 100
Wt = = 0.34 kWh.tonne-1 = 1.23 MJ.tonne-1
5 10
5

and the amount of energy required to reduce a known feed size to
a given product size is measured. For MRT the amount of energy
needed to reduce the secondary crushed product from -19 mm to
a final product size of -210 microns was, on average over a
two-month period, 16.10 kWh tonne-1. As the Bond Index works
on 80 per cent passing size the feed and product sizes are taken
as 16300 microns (16.3 mm) and 180 microns respectively.
Substituting these values into Equation A10 gives:
180 1 / 2 16300 1 / 2
K B = 1610
. 1/ 2
= 241.4 kWh-micron.tonne-1
16300 180
1/ 2
Equation A9 can also be rearranged to calculate the energy
required to reduce an infinite feed size (F = ) down to any
product size P. This is referred to as the total energy (Wt) and is
given by:
1 1 K
Wt + K B 1 / 2 1 / 2 = 1B/ 2 (A11)
P P
Based on Equation A11 the Bond Work Index (Wi) is the
amount of energy required to reduce an infinite feed size down to
an 80 per cent passing size of 100 microns. This is used as a
common basis of comparison across different materials and
processes and is given by:
The energy supplied by the explosive acting on the rock mass
can be derived from the known powder factor (PF) at 0.31
kg.tonne-1 for the blasts and the energy contained in the
explosive used. The energy for the heavy ANFO used, with
specific density 1.2 g.cm-3, is 4.5 MJ.kg-1. The explosive energy
per tonne is therefore:
PF Explosive Energy = 0.31 4.5 = 1.40 MJ.tonne-1
This value of 1.40 MJ.tonne-1 compares favourably with the
value of 1.23 MJ.tonne-1 derived using the Bond Index for
comminution (Hustrulid, 1999). If it is assumed that the
difference in values is due to slight differences in the efficiencies
of the two processes then it is reasonable to reconcile the two
values by applying a factor () that is appropriate over a range of
energies.
By rearranging Equation A13 and applying the correction
factor (Van Zeggeren and Chung, 1975) the equation for
product size from the powder factor used in the blast design is:
2
W
P = i (A14)
Wt
where:
Wt is the energy equivalent of the powder factor

486 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 Authors Index
A D Hodgkison J 417
Aitken O C 3 Dare-Bryan P C 473 Hollings P 13
Amos Q G 3 de Caritat P 219, 273 Hooper B 381
Anand R R 29 Dominy S C 291 Horn C M 453
Anderson C 221 Dowd P A 473 Huntington J F 223, 283
Arif J 93
Arundell M C 207
E I
Ashley P M 381
Eggleton R A 273 Isles D 221
Elliot J 303
B Elliott P J 307
J
Baker T 93 Enders M S 27
Johansen G F 291, 317
Banks M J 95 Erceg M 143
Bastrakov E 223 Erskine W D 447
Basuki Dwi Priyono M D 359 Evans K G 447 K
Bell P D 105 Keeling J 223
Bellio M G 447 Kidd R P 323
F
Belperio A 115 Kirste D 219, 273
Fan K 143
Berman M 223
Faure K 127
Berthelsen R 339 L
Fouet T 417
Binns R A 161
Freeman H 115 Lawrence M J 45
Black A 443
Frikken P H 13 Leach T M 143, 153, 303
Boyd G W 211
Leveille R A 27
Brathwaite R L 127
Loayza C E 105
Brooks C C 465 G
Lowder G G 333
Buenavista A 173 Gatehouse S G 207
Lowe I 53
Burhanuddin S 161 Gee R D 29
Lynch J E 385
Goldfarb R J 189
Gmez J G 105
C M
Gonzalez J 173
Cannell J 13
Goode A D T 199 MacKenzie D J 135
Clemens T 443
Mair J L 189
Clode C 93
Marsh E E 189
Close S E 9 H
Mason P 223, 283
Connor P 223 Hai L V 95
Mauger A 223
Cooke D R 13 Hananto N D 161
McCarthy P L 57
Corbett G 15 Harlan B 93
McConachy T F 161
Coward D 223 Harman P 221
McCuaig T C 417
Craw D 135 Hart C J R 189
McQueen K G 231
Cristofani P 25 Heithersay P 223

PACRIM 2004 Adelaide, SA, 19 - 22 September 2004 487

McRae A 231 Q Thirnbeck M R 391
Meares R M D 333 Trench A 87
Quynh N N 95, 347
Middleton C 173 Trumm D 443
Miller I B 453 Turvey D 435
Miller J 417 R Tyrwhitt D S 401
Miskelly N 63 Richards T H 359
Moliere D R 447 Ritchie T J 251
U
Moreno G 173 Roach B 381
Utomo E P 161
Morrison B 69 Robinson J R 323
Mueller S H 189 Rohrlach B 173
Munro S 339 Rombach C S 189 V
Murfitt R H 95, 347 Vickers M D 333

S
O Sale M J 239 W
Osman A H 429 Saynor M J 447 Wake B A 333
OSullivan K N 199 Schodde R C 367 Walshe J L 13
Scott K M 257, 273, 283 Watanabe K 263
Setijadji L D 263 Wesson B 403
P
Shaw R 71 Whitbourn L 223
Parr J M 161
Shywolup W 347 White M J 333
Partington G A 239
Sirinawin T 347 Widdup H 417
Perdanakusumah A 93
Skirrow R 223
Permana H 161
Smith R J 77
Perry K W 459 Y
Subang L 173
Petrie B 135 Yang K 283
Switzer C K 381
Phillips R 223 Yeats C J 161
Pinto R M 105
Priadi B T
161 Z
Prihananto A 345 Tan K P 273
Zulkarnain I 161
Pringle I 303 Teale G S 385

488 Adelaide, SA, 19 - 22 September 2004 PACRIM 2004

 MONOGRAPH SERIES
1 Detrital Heavy Minerals in Natural Accumulates George Baker 1962
2. Research in Chemical and Extraction Metallurgy Ed: J T Woodcock, 1967
A E Jenkins
and G M Willis
3. Broken Hill Mines - 1968 Ed: M Kadmanovich 1968
and J T Woodcock
4. Economic Geology of New Zealand Ed: G J Williams 1974
5. Economic Geology of Australia and Papua New Guinea - 1 Metals Ed: C L Knight 1975
6 Economic Geology of Australia and Papua New Guinea - 2 Coal Ed: D M Traves 1975
and D King
7. Economic Geology of Australia and Papua New Guinea - 3 Petroleum Ed: R B Leslie 1976
H J Evans
and C L Knight
8. Economic Geology of Australia and Papua New Guinea - 4 Industrial Ed: C L Knight 1976
Minerals and Rocks
9. Field Geologists Manual
1st Edition Ed: D A Berkman 1976
2nd Edition and W Ryall 1982
3rd Edition Ed: D A Berkman 1989
4th Edition Ed: D A Berkman 2001
10 Mining and Metallurgical Practices in Australasia Ed: J T Woodcock 1980
(the Sir Maurice Mawby Memorial Volume)
11. Victorias Brown Coal - A Huge Fortune in Chancery Ed: J T Woodcock 1984
(the Sir Willis Connolly Memorial Volume)
12. Australasian Coal Mining Practice
1st Edition Ed: C H Martin 1986
2nd Edition Ed: C H Martin 1993
and A J Hargraves
13. Mineral Deposits of New Zealand Ed: Dr D Kear 1989
14. Geology of the Mineral Deposits of Australia and Papua New Guinea Ed: F E Hughes 1990
15. The Rocks Speak H King 1989
16. Hidden Gold - The Central Norseman Story J D Campbell 1990
17. Geological Aspects of the Discovery of Some Important K R Glasson 1990
Mineral Deposits in Australia and J H Rattigan
18. Down Under - Mineral Heritage in Australasia Sir Arvi Parbo 1992
19. Australasian Mining and Metallurgy Ed: J T Woodcock 1993
(the Sir Maurice Mawby Memorial Volume) and K Hamilton
20. Cost Estimation Handbook for the Australian Mining Industry Ed: M Noakes 1993
and T Lanz
21. History of Coal Mining in Australia Ed: A J Hargraves, 1993
(the Con Martin Memorial Volume) R J Kininmonth,
C H Martin
and S M C Saywell
22. Geology of Australian and Papua New Guinean Mineral Deposits Ed: D Berkman 1998
and D Mackenzie
23. Mineral Resource and Ore Reserve Estimation Ed: A C Edwards 2001
The AusIMM Guide to Good Practice

Copies of all books currently in print can be obtained from

The Institute office - Tel (03) 9662 3166

Key: Out of print

 PUBLICATIONS OF THE AUSTRALASIAN INSTITUTE
OF MINING AND METALLURGY

CONFERENCE, SYMPOSIUM AND MISCELLANEOUS

1972 1/72 SI * Project Evaluation and Management, Melbourne
2/72 * Annual Conference, Newcastle
1973 1/73 S2 * Subsidence, Illawarra
2/73 S3 * Mine Filling, North West Queensland
3/73 S4 * Transportation, Sydney
4/73 S8 * Mine Fires, Southern Queensland
5/73 * Annual Conference, Western Australia
1974 1/74 S5 * Support in Pillar Extraction, Illawarra
2/74 S6 * Recent Technical and Social Advances in the North Australian Minerals Industry,
North West Queensland
3/74 S7 * Pellets and Granules, Newcastle and District
4/74 * Annual Conference, Southern and Central Queensland
1975 1/75 S9 * People and the Mining Industry - The Future, Broken Hill
2/75 S 10 * Occupational Safety in Mines, Southern Queensland
3/75 S 11 * Australian Black Coal, lllawarra
4/75 * Annual Conference, South Australia
1976 1/76 S13 * Landscaping and Land Use Planning as Related to Mining Operations, Adelaide
2/76 * Design and Construction of Tunnels and Shafts, Melbourne - 2 nd Australian
Tunnelling Conference
3/76 S 14 * Thick Seam Mining by Underground Methods, Central Queensland
4/76 S 15 * Sampling Practices in the Mineral Industries, Melbourne
5/76 * AnnualConference, lllawara
1977 1 /77 S 16 * Apcom 77, Brisbane
2/77 S 18 * Coal Borehole Evaluation, Southern Queensland
3/77 S 17 * Underground Operators Conference, Broken Hill
4/77 * Annual Conference, Hobart
1978 1 /78 S 19 * Mill Operators Conference, North West Queensland
2/78 S20 * Rock Breaking Equipment and Techniques, Melbourne
3/78 * International Resource Management, Canberra
4/78 * Annual Conference, Townsville
1979 1/79 S21 * Utilisation of Steelplant Slags, lllawarra
2/79 S22 * Estimation and Statement of Mineral Reserves, Sydney
3/79 * Annual Conference, Perth
1980 1/80 * Annual Conference, New Zealand
2/80 S23 * Australia/Japan Extractive Metallurgy Symposium, Sydney
3/80 S/24 * Occurrence, Prediction and Control of Outbursts in Coal Mines, Southern
Queensland
4/80 S25 * Management in the Mining Industry, Melbourne

Copies of all books currently in print can be obtained from The Institute office
Tel (03) 9662 3166 or Fax (03) 9662 3662

* = Out of print
The S numbers in the third column refer to an older identifying number for Symposia, the numbers preceeding the S number signify the new
publication ordering number.
1981 I/81 S26 * International Blast Furnace Hearth and Raceway Symposium, Newcastle
2/81 * Fourth Australian Tunnelling Conference
3/81 S27 * Ignitions, Explosions and Fires, Wollongong
4/81 * Annual Conference, Sydney
5/81 S28 * Strip Mining 45 Metres and Beyond, Central Queensland
1982 1/82 S29 * Off Highway Truck Haulage Conference, Newman
2/82 S30 * Mill Operators Conference, North West Queensland
3/82 S31 * Underground Operators Conference, West Coast Tasmania
4/82 * Annual Conference, Melbourne
5/82 S32 * Carbon-ln-Pulp Technology for the Extraction of Gold, Perth and Kalgoorlie,
(Reprinted 1988)
6/82 S33 * Seam Gas Drainage with Particular Reference to the Working Seam, Wollongong
1983 1/83 S34 * Computers in Mining, Southern Queensland
2/83 * Annual Conference, Broken Hill
3/83 S35 * Project Development Symposium, Sydney
4/83 S37 * Ventilation of Coal Mines, Wollongong
5/83 S40 * Principles of Mineral Flotation (The Wark Symposium), Adelaide
1984 1/84 S36 * Metallurgy Symposium, Melbourne
2/84 S38 * Coal and Mineral Sizing, Wollongong
3/84 * Annual Conference, Darwin
4/84 S39 Gold Mining, Metallurgy and Geology, Kalgoorlie
1985 1/85 S41 * Smelting and Refining Operators Symposium, North Queensland
2/85 S42 * Underground Operators Conference, Kalgoorlie
3/85 * Annual Conference, Brisbane
4/85 S43 * Scientific and Technological Developments in Extractive Metallurgy (G K
Williams Memorial Volume), Melbourne
1986 1/86 * l3th Congress The Council of Mining and Metallurgical Institutions, Singapore, 6
Volumes
2/86 S44 * Selective, Open Pit Gold Mining Seminar, Perth
3/86 S45 * Ground Movement and Control Related to Coal Mining, Wollongong
4/86 S46 * Australia: A World Source of Illmenite, Rutile, Monazite and Zircon Conference,
Perth
5/86 S47 * Second Project Development Symposium, Sydney
6/86 S48 * Large Open Pit Mining Conference, Newman
7/86 S49 * Education and Research for the Mineral Industry for the Future, Melbourne
8/86 * The AuslMM 10 Year lndex
1987 1/87 * Vl Australian Tunnelling Conference: Bore or Blast, Melbourne
2/87 S50 * Risk and Survival Seminar, Canberra
3/87 * Annual Conference, Newcastle: Coal Power 87
4/87 S51 * Research and Development in Extractive Metallurgy, Adelaide
5/87 * Leslie Bradford Golden Jubilee Oration
6/87 S52 * Mining and Environment: A Professional Approach, Brisbane
7/87 * Pacrim 87, Gold Coast, Queensland
8/87 S53 * Dense Medium Operators Conference, Brisbane

Copies of all books currently in print can be obtained from The Institute office
Tel (03) 9662 3166 or Fax (03) 9662 3662

* = Out of print
The S numbers in the third column refer to an older identifying number for Symposia, the numbers preceeding the S number signify the new
publication ordering number.
 9/87 S54 * Equipment in the Minerals Industry: Exploration Mining and Processing
Conference, Kalgoorlie
10/87 S55 * Resources and Reserves, Sydney
11/87 * South Australias Mining Heritage
1988 1/88 S56 * 21st Century Higher Production Coal Mining Systems Symposium, Wollongong
2/88 * The Second International Conference on Prospecting in Arid Terrain, Perth
3/88 S57 * Third Mill Operators Conference, Cobar
4/88 S58 * Underground Operators Conference, Mount Isa
5/88 * Fourth International Mine Ventilation Congress, Brisbane, (Proceedings and
Addendum volume)
6/88 * Annual Conference, Sydney: Minerals and Exploration at the Crossroads: The
International Outreach
7/88 S59 * Second AuslMM Mineral Heritage Seminar, Sydney
8/88 S60 * Economics and Practice of Heap Leaching in Gold Mining Workshop, Cairns
9/88 * Third International Mine Water Congress, Melbourne
10/88 S61 * Explosives in Mining Workshop, Melbourne
1989 1/89 * Mineralogy and Petrology, Sydney, February
2/89 * Second Large Open Pit Mining Conference, Latrobe Valley Vic
3/89 * NQ Gold 89 Conference, Townsville Qld
4/89 * Annual Conference, Perth-Kalgoorlie: Education, Training and Professional
Development; Industrial Minerals; Project Development/Processing
5/89 * Mineral Fuel Alternatives and the Greenhouse Effect, July 1989
6/89 Non-ferrous Smelting Symposium: 100 Years of Smelting and Refining
Operations in Port Pirie, SA September 1989
7/89 * Dewatering Technology and Practice Conference, Brisbane October 1989
89 MINVAL 89, Mining and Petroleum Valuation 1989, Sydney September 1989
1990 1 /90 * Ore Reserve Estimates - The Impact on Miners and Financiers, Melbourne, March
1990
2/90 Annual Conference, The Mineral Industry in New Zealand, Rotorua New
Zealand, March 1990
3/90 Pacific Rim Congress, Gold Coast Qld, May 1990
4/90 * Mining Industry Capital and Operating Cost Estimation Conference, Sydney,
June 1990
5/90 * Third International Symposium on Rock Fragmentation by Blasting, Brisbane,
August 1990
6/90 * Sir Edgeworth David Memorial Oration, May 1990
7/90 * Mine Geologists Conference, Mount Isa, October 1990
1991 1/91 * Fourth Mill Operators Conference, Burnie Tas, March 1991
2/91 * World Gold 91, Cairns Qld, April 1991
3/91 Mining Industry Optimisation Conference, Sydney, June l991
4/91 * PNG Geology, Exploration and Mining Conference, Rabaul, June 1991
5/91 Qld Coal Symposium, Brisbane, August 1991
6/91 * Reliability Production and Control in Coal Mines, Wollongong, September 1991
7/91 * Fifth AuslMM Extractive Metallurgy Conference, Perth, October 1991

Copies of all books currently in print can be obtained from The Institute office
Tel (03) 9662 3166 or Fax (03) 9662 3662

* = Out of print
The S numbers in the third column refer to an older identifying number for Symposia, the numbers preceeding the S number signify the new
publication ordering number.
1992 1/92 * Enviromine Australia, Sydney NSW, March 1992
2/92 The AuslMM Annual Conference, The State-of-the-Art - A Product of 100 Years
of Learning, Broken Hill NSW, May 1992
3/92 Energy, Economics and Environment Gippsland Basin Symposium, Melbourne,
June 1992
4/92 Arnold Black Mineral Heritage Oration
5/92 The Man from ASARCO: a life and times of Julius Kruttschnitt
6/92 5th Underground Operators Conference, Ballarat, July 1992
7/92 * 11th International Conference on Ground Control in Mining, Wollongong, July
1992
8/92 * Third Large Open Pit Mining Conference, Mackay, August 1992
9/92 * Extractive Metallurgy of Gold and Base Metals Conference, Kalgoorlie, October
1992
10/92 * Sampling Practices in the Minerals Industry, Mount Isa, November 1992
11/92 * Rehabilitate Victoria, Latrobe Valley, November 1992
1993 1/93 Mining People - A Century
2/93 The AuslMM Centenary Conference, Adelaide, March 1993
3/93 XVIII International Mineral Processing Congress, Sydney, May 1993
4/93 * Narrow Vein Mining Seminar, Bendigo, June 1993
5/93 * International Mining Geology Conference, Kalgoorlie, July 1993
6/93 Vlll Australian Tunnelling Conference, Sydney, August 1993
7/93 * World Zinc 93 - International Symposium, Hobart, October 1993
1994 1/94 1994 AuslMM Student Conference, Brisbane, April 1994
2/94 * PNG Geology, Exploration and Mining Conference, Lae, PNG, June 1994
3/94 No Two The Same by Bert Mason
4/94 Sixth Extractive Metallurgy Conference, Brisbane July 1994
5/94 * 1994 AuslMM Annual Conference, Darwin, August 1994
6/94 * 4th Large Open Pit Mining Conference, Perth, September 1994
7/94 * Recent Trends in Heap Leaching, Bendigo, September 1994
8/94 * Maintenance in the Mining and Metallurgical Industries,Wollongong, October
1994
9/94 * Fifth Mill Operators Conference, Roxby Downs, October 1994
10/94 Mineral Valuation Methodologies 1994, Sydney, October 1994
11/94 Victorian Mining Week Conference, Melbourne, October 1994
1995 1/95 1995 AuslMM Annual Conference, Newcastle, March 1995
2/95 Sir Maurice Mawby Memorial Oration
3/95 Worlds Best Practice in Mining and Processing Conference, Sydney, May 1995
4/95 APCOM XXV 1995 Conference, Brisbane, July 1995
5/95 Mineral Valuation Methodologies 1994, Sydney, October 1994 (revised)
6/95 * EXPLO 95 Conference, Brisbane, September 1995
7/95 Underground Operators Conference, Kalgoorlie, November 1995
8/95 * Young Professionals Conference, Mt Isa, October 1995
9/95 * PACRIM 95 Congress, Auckland, New Zealand, November 1995
10/95 * Ethics, Liability and the Technical Expert, Sydney, December 1995

Copies of all books currently in print can be obtained from The Institute office
Tel (03) 9662 3166 or Fax (03) 9662 3662

* = Out of print
The S numbers in the third column refer to an older identifying number for Symposia, the numbers preceeding the S number signify the new
publication ordering number.
1996 1/96 * 1996 AusIMM Annual Conference, Perth, March 1996
1a/96 * 1996 AusIMM Annual Conference Supplementary Volume, Perth, March 1996
2/96 * Ethics, Liability and the Technical Expert, Sydney, March 1996
3/96 * Entrepreneurs and Partners, Sydney, July 1996
4/96 * Contract Operators Conference, Kalgoorlie, October 1996
5/96 Asia/Pacific Mining Communications Summit, Singapore, November 1996 -
Withdrawn
6/96 * Nickel 96, Kalgoorlie, November 1996
1997 1/97 1997 AusIMM Annual Conference, Ballarat, March 1997
2/97 * World Gold 97 Conference, Singapore, September 1997
3/97 Sixth Mill Operators Conference, Madang, PNG, October 1997
4/97 Gem 97, Madang, PNG, October 1997
5/97 * Contract Operators Conference, Brisbane, Qld, October 1997
6/97 Third International Mining Geology Conference, Launceston, Tas, November
1997
7/97 Mindev 97 - The International Conference on Mine Project Development,
Sydney, November 1997
8/97 1997 AusIMM Travelling Technology Forum, Singleton, NSW, March 1997
1998 1/98 * MINEFILL 98 - The Sixth International Symposium on Mining with Backfill,
Brisbane, Qld, April 1998
2/98 * AusIMM98 - The Mining Cycle, Mount Isa, Qld, April 1998
3/98 * Seventh Underground Operators Conference, Townsville, Qld, June/July 1998
4/98 Mine to Mill Conference, Brisbane, Qld, October 1998
5/98 Third Regional APCOM - Computer Applications in the Minerals Industries
International Symposium, Kalgoorlie, WA, December 1998
1999 1/99 10th Australian Tunnelling Conference, Melbourne, Vic, March 1999
1a/99 10th Australian Tunnelling Conference Keynote Addresses and AsiaPacific
Forum, Melbourne, Vic, March 1999
2/99 Students and Young Professionals Conference, Perth, WA, July 1999 (N/A)
3/99 ICARISM 99 Conference, Perth, WA, September 1999
4/99 PACRIM 99 Congress, Bali, October 1999
5/99 EXPLO 99 Conference, Kalgoorlie, WA, November 1999
2000 1/2000 Southern Africa - Australia Mineral Sector Synergies Symposium, Canberra,
ACT, March 2000
2/2000 After 2000 - The Future of Mining, Sydney, NSW, April 2000
3/2000 4th International Mining Geology Conference, Coolum, Qld, May 2000
4/2000 Young Leaders 2000, Sydney, NSW, July 2000
5/2000 MINPREX 2000, Melbourne, Vic, September 2000
6/2000 Seventh Mill Operators Conference, Kalgoorlie, WA, October 2000
7/2000 MassMin 2000, Brisbane, Qld, October - November 2000
2001 1/2001 Strategic Mine Planning 2001, Perth, WA, March 2001
2/2001 AusIMM Youth Congress 2001, Brisbane, Qld, May 2001
3/2001 International Heavy Minerals Conference, Fremantle, WA, June 2001
4/2001 EXPLO 2001, Hunter Valley, NSW, October 2001
5/2001 Mineral Asset Valuation Issues 2001, Sydney, NSW, October 2001

Copies of all books currently in print can be obtained from The Institute office
Tel (03) 9662 3166 or Fax (03) 9662 3662

* = Out of print
The S numbers in the third column refer to an older identifying number for Symposia, the numbers preceeding the S number signify the new
publication ordering number.
2002 1/2002 Young Leaders 2002, Kalgoorlie, WA, March 2002
2/2002 Metallurgical Plant Design and Operating Strategies, Sydney, NSW, April 2002
3/2002 CMMI Congress, Cairns, Qld, May 2002
4/2002 Green Processing Conference, Cairns, Qld, May 2002
5/2002 Eighth Underground Operators Conference, Townsville, Qld, July 2002
6/2002 AusIMM 2002 Conference, 150 Years of Mining, Auckland, New Zealand,
September 2002
7/2002 Iron Ore Conference, Perth, WA, September 2002
8/2002 Value Tracking Symposium, Brisbane, Qld, October 2002
2003 1/2003 Twelfth International Symposium on Mine Planning and Equipment Selection,
Kalgoorlie, WA, April 2003
2/2003 Young Leaders 2003, Brisbane, Qld, April 2003
3/2003 Sixth International Conference Acid Rock Drainage, Cairns, Qld, July 2003
4/2003 Eighth Mill Operators Conference, Townsville, Qld, July 2003
5/2003 Mining Risk Management Conference, Sydney, NSW, September 2003
6/2003 Water in Mining 2003, Brisbane, Qld, October 2003
7/2003 Fifth Large Open Pit Conference, Kalgoorlie, WA, November 2003
8/2003 Fifth International Mining Geology Conference, Bendigo, Vic, November 2003
2004 1/2004 New Leaders 2004, Ballarat, Vic, April 2004
2/2004 Second International Green Processing Conference, Fremantle, WA, May 2004
3/2004 EXPLO 2004, Perth, WA, July 2004
4/2004 Metallurgical Plant Design and Operating Strategies, Perth, WA, September 2004
5/2004 PACRIM 2004 Congress, Adelaide, SA, September 2004

Copies of all books currently in print can be obtained from The Institute office
Tel (03) 9662 3166 or Fax (03) 9662 3662

* = Out of print
The S numbers in the third column refer to an older identifying number for Symposia, the numbers preceeding the S number signify the new
publication ordering number.
 SPECTRUM SERIES
1. Making the Mount Isa Mine, 1923 - 1933 Don Berkman 1996
2. History of Drilling Graham McGogggan 1996
3. The Cobar Mineral Field - A 1996 Perspective Warren Cook 1996
Andrew Ford
Julian McDermott
Peter Standish
Craig Stegman and
Therese Stegman
4. Towards 2000 - Resource to Reserve Inputs Seminar - Melbourne, Vic 1997

5. Towards 2000 - National Conference on Ironmaking Resources and Reserves 1997

Estimation, Perth, WA
6. Towards 2000 - The Resource Database Towards 2000 - Wollongong, NSW 1997
7. Towards 2000 - Ore Reserves and Finance - Sydney, NSW 1998
8. Towards 2000 - Assessment of Reserves in Low Rank Coals - Morwell, Vic 1997
9. Towards 2000 - Ore Reserve Reconciliation Workshop - Darwin, NT 1997
10. Towards 2000 - Gold and Nickel Ore Reserve Estimation Practice Seminar 1998
11. Towards 2000 - Resource/Reserves Estimation Practice in the Central West 1999
New South Wales Mining Industry, Cobar, NSW
12. Field Guide for Geoscientists and Technicians 2004

Copies of all books currently in print can be obtained from The Institute office
Tel (03) 9662 3166 or Fax (03) 9662 3662