Escolar Documentos
Profissional Documentos
Cultura Documentos
Edited by
Publication arrangements by
Commission of the European Communities
Directorate-General Scientific and Technical Information and
Information Management, Luxembourg
EUR 6862
Copyright © 1980 ECSC, EEC, EAEC, Brussels and Luxembourg
Softcover reprint of the hardcover 1st edition 1980
LEGAL NOTICE
Neither the Commission of the European Communities nor any person acting on behalf of the
Commission is responsible for the use which might be made of the following information.
INTRODUCTION ix
OPENING SESSION;
E.R. OXBURGH, S.W. RICHARDSON, S.M. WRIGHT, M.Q.W. JONES, S.R. PENNEY,
S.A. WATSON and J.R. BLOOMER / Heat Flow Pattern of the United Kingdom 447
J. WHEILDON, M.F. FRANCIS, J.R.L. ELLIS and A. THOMAS-BETTS /
Exploration and Interpretation of the SW England Geothermal
~~~ 456
R. GABLE / Terrestrial Heat Flow in France 466
G. VASSEUR (Groupe FLUXCHAF) / A Critical Study of Heat Flow Data in
France
474
N. BALLING, J.I. KRISTIANSEN, K.D. POULSEN and S. SAXOV / Geothermal
Trends of Denmark 485
W. VAN DALFSEN / The Shallow Subsurface Temperature Field in The
Netherlands 496
M. FANELLI, A. ROSSI, M. SALOMONE and L. TAFFI / Temperature and
Heat Flow Patterns of Italy 506
G. FRAS, J.M. THILLIER and P. JOUANNA / Harmonic Pick-up:
Preliminary Mathematical Model for Interpretation 516
B. CROSNIER, J.L. PORTALES and P. JOUANNA / The Perfecting of a
Harmonic Pick-up for the Simultaneous Detection of the Thickness
and Extent of a Fracture Crossed by a Boring 521
J. BEHRENS, B. ROTERS and H. VILLINGER / In Situ Determination of
Thermal Conductivity in Cased Drill Holes 525
L. CEPPATELLI and G.C. FERRARA / Design and Testing of Down Hole
Probes for Operation in Deep and Hot Environments 535
M.G. BERTAUX, G. BIENFAIT, J. JOLIVET and G. VASSEUR / Direct Heat
Flow Measurement Probe Project 545
G. MUSMANN and W. KESSELS / An In-Situ Thermal Conductivity Probe
556
SESSION V - GEOPHYSICS I - ACTIVE AND PASSIVE SEISMICS
-vi-
J.M. LEJEUNE, N. BADINIER, V. BARTHES, A. CLOT, J.M. COUDERT, A. DESPLAN,
H. FABRIS, O. GOYENECHE, Ph. MAGET, J. ROJAS, P. SCHERR and D. TOURNAYE /
Technical and Economic Feasibility of Low Enthalpy Geothermal Projects
in the E.E.C. 810
-vii -
H.D. FUHRBERG, F. BRINKMANN and K. SCHOBER / Erfahrungen mit Frac-
behandlungen in Tiefen, Geringpermeablen Gaslagerstatten 994
A.S. BATCHELOR, C.M. PEARSON and N.P. HALLADAY / The Enhancement of
the Permeability of Granite by Explosive and Hydraulic Frac-
turing 1009
F.H. CORNET / Analysis of Hydraulic Fracture Propagation a Field
Experimentation 1032
O. KAPPELMEYER and F. RUMMEL / Investigations on an Artificially
Created Frac in a Shallow and Low Permeable Environment 1044
-~-
INTRODUCTION
Four years have passed since the decision of the Council of Ministers
to launch the first Community Energy Research and Development Programme,
dealing with geothermal energy, solar energy, hydrogen production and utili-
sation, energy conservation and energy systems analysis. A seminar on geo-
thermal energy was held two years ago to report on work in progress at the
half-way stage of the four-year programme.
It is hoped that these proceedings will prove helpful to all those who
work in the field of geothermal energy and in particular will provide guid-
ance to the Commission of the European Communities in implementing its Second
Energy R&D Programme which has recently got under way.
-u-
EXHIBITION
EXHIBITION HALL
-x-
Conference Chairman A. STRUB
Session Chairmen:
Keynote Speakers
H. TAZIEFF
M. FYTIKAS
C. OTTE
J.P. MUNIER-JOLAIN
R.M. POTTER
F. TONANI
Th. KREY
H. ALONSO ESPINOSA
M. FARABOLINI
H.D. FUHRBERG
-xi -
-
._--.-- .. - - -'.'-
~.-
..... - .... _-
-- ----.--
~:
- xili-
I
~.
- Opening Address
A STRUB, Commission of the European Communities
Directorate-General for Research, Science and
Education
- Opening Address
C. CORVI, Chairman of the ACPM on Geothermal Energy
- Keynote speech
H. TAZIEFF, Directeur de recherche au CNRS, President
de l'Institut National de Volcanologie
- 1-
OPENING ADDRESS
A. STRUB
Commission of the European Communities
Directorate-General for Research, Science and Education
Let me use this opportunity to make a few remarks which might contribute
to put this event into the right context.
But with all this, we must not forget to prepare ourselves for the longer
term future.
-3-
This leads us to the topic of energy research and development. It is only
by the means of science and research that we can prepare the technologies
of tomorrow. Those technologies which will have to come into play in a
far, post-oil future, much as thermonuclear fusion, and also those which
can yield some contribution to our energy supply already before the turn
of the century, like solar and geothermal energy.
The R&D efforts undertaken towards this goal in the EC Member countries
are considerable, and - fortunately - still increasing. We may honestly
say that the European Communities playa significant role in this overall
effort, be it by its research activities carried out in their own Joint
Research Center (mainly Ispra/ltaly), be it by their so-called "indirect
action programmes" in the frame of which EC is stimulating work to be
carried out by research bodies in the nine Member countries. To do so, EC
is not only providing for financial support but also supplying the interna-
tional platform for information exchange and for coordination of research.
This seminar is dealing with the geothermal energy part of the first EC
Energy Research ad Development Programme. This programme was started in
July 1975 and ended in June 1979. It contained the following sub-programmes
- energy conservation
- hydrogen production and use
- solar energy
- geothermal energy
- energy systems modelling.
-4-
A budget of 59 million EUA (of which 13 million for geothermal energy) has
been allocated to this purpose and some 700 R&D contracts have been
concluded in consequence.
As already said, this First Programme is terminated now and we are presently
summing up its results and presenting them to a wider public.
In the meanwhile, the Communities' Second Energy R&D Programme has been
launched, containing the same five sub-programmes as the first programme
and involving a total budget of 105 million EUA over 4 years (mid-1979 to
mid 1983).
I will stop now and give the floor to Dr. Corvi, Chairman of our Advisory
Committee on Geothermal Energy Research. I hope you will draw much benefit
of this meeting but also enjoy in general your stay in this hospitable
city of Strabourg.
-5-
OPENING ADDRESS
Dr. C. COR VI
Chairman of the ACPM on Geothermal Energy
As you know, the Advisory Committee is made up of experts from the nine
Member States of the Community. It gives advice and help to the Commission
on the best way of implementing the programme, the results of which we shall
be hearing about and discussing over the next three days.
We on the Committee have been closely and actively invoLved in both working
out and implementing the programme. We are therefore fully conversant with
all aspects of the work and the progress made on each.
However, because of our numerous other activities, we have had little chance
to become directly acquainted with the individual contractors who carry out
the research - research which often our Committee has discussed at some
length before commissioning.
-7-
research programme and use this essential information for our next task, the
second Community programme, which has just started.
May I ~ay that we are doing all we can to enable the Commission, thanks to
the assistance of and two-way cooperation with a large number of research
laboratories and establishments, to carry out its task of coordinating,
intensifying and even stimulating research into the energy sources of
tomorrow. Geothermal energy will thus be fully able to play its part in
helping to ease our energy supply problems, at least in those places where
nature has provided the requisite conditions.
-8-
GEOTHERMAL RESEARCH IN THE EUROPEAN COMMUNITY
INTRODUCTION
Exploration successes
The purpose of this Seminar is to provide an assessment of geothermal
activity in the Community, whose vitality is confirmed by your presence
alone, if, in fact, there is any need for such confirmation. There are
other encouraging signs as well, particularly in exploration. Recent
borehole successes with our Research, Development and Demonstration
programmes have revealed major geothermal resources in France, Belgium,
the United Kingdom and Italy which will increase the European Community's
energy reserves.
We now know - from research and from fieLd expLoration - that many tem-
perature charts are faLsified, as it were, by the favoured, generaLLy
high, structuraL position of the measuring points, by convection, the
variety of instruments, measuring conditions and procedures and the inter-
poLation ruLes empLoyed which are more arithmeticaL than physicaL. Never
mind, the atLas is a unique initiative, the first of a series drawn up
on a cooperative basis at Community LeveL and does not "wipe away"
nationaL and regionaL geothermaL specificities, anomaLies and particuLari-
ties (quite the reverse, in fact). We intend to continue with,and diversify
this exempLary achievement in our second R&D programme.
-11-
(iv) Nor we should forget the decision which has now been finalized to
carry out a preliminary exploratory deep drilling of the Rhine Graben at
Strasbourg-Cronenbourg, less than five kilometers from the Conference
building. It is no mean feat to have arrived at an agreement between the
fifteen promoters of the project, who come from three oil companies, one
coal-mining firm, two heating groups, gas and electricity companies and
utilities, heating engineers and distributors and the local authorities,
all partners whose skills are as complementary as their interests are
divergent.
Considerable hope was placed on the Sasso borehole whose objective (the
reconnaissance of deep fractions in the Larderello dry steam field at
depths of 4500 to 5000 m) was particularly ambitious and without precedent
in geothermal exploration. The available indices, the presence of a con-
tinuous and energic seismic reflector throughout the zone and gas geo-
chemistry indicating a deep origin with a source temperature of 350°C were
all favourable. However, hostile conditions (drilling with a total loss
of circulation; cementation of the tubing while negotiating the upper
field; corrosion of the downhole and wellhead equipment, tubulars and drill
stems; high temperatures (380°C); frequent fishing jobs and instrumenta-
-12-
tion) have recently lead the operator to abandon the well after two years
and 4200 metres of accumulated problems. In addition to the temperatures,
the data collected show a productive capacity down as far as 3800 m and
good injectivity in the formation below 3000 m. The experience gained on
the Sasso well will indeed benefit to the deep exploration of geothermal
fields in future and the development of suitable drilling equipment and
techniques.
It is true that geology is not an exact science, and the profession acknow-
ledges the right to errors.
A number of factors both outside and inside the European Community temper
the optimism that is felt; in particular there are the still too modest
role played by the Community in world geothermics (research, production,
budgets) and the slowness which has been observed in the development of
electrical and other direct uses for geothermal heat, where its potential
is proven, operationaL procedures exist and geothermaL experience, if not
the appropriate legisLation, is considerable.
-13-
date an output of 7100 MWt, of which 50 MWt, or less than 1%, was produced
by the Community (France and Italy). By way of comparison, the USSR, Hun-
gary and Iceland produce 5000, 1500 and 500 MWt respectively.
We are very pleased to note the presence here of ENEL, union Oil of Cali-
fornia (Dr Otte) and the Mexican Federal Electricity Commission (Dr Alonso);
between them these three concerns account for 75% of installed geothermal
generating capacity and probably 90% of the reserves.
The objective in the French Plan is more modest : a saving of 1 Mtoe per
year is the ceiLing which has been set for geothermaL energy by 1985; this
represents 650 000 equivalent dwellings (or 3500 ha of green houses) and
will require 300 doubLets at an average rate of 60 (e.g. 120 wells) per
year, the permanent use of at least fifteen drilling rigs (capable of rea-
ching depths between 1500 and 3500 m) and an annuaL investment of
FF 1000 milLion valued as at January 1980 which will give an overall pay-
back time often in excess of ten years. We are also familiar with the pro-
blems of hitting onshore drilling rigs when not even thirty such units
are available in the Community at the moment for these tasks, all prospec-
ting schemes considered.
- 14-
capacity of 500 MWe in 1985, increasing to 700 MWe by 1993. Leaving aside
the reconditionning of existing welLs and instaLLations required to main-
tain capacity at its current LeveL (400 MWe), this increase wiLL invoLve
using at Least ten driLLing rigs and making an excLusiveLy mining in-
vestment of Lit. 40 000 miLLion per year vaLued as at January 1980.
- 15-
THE ACHIEVEMENTS OF THE FIRST R&D PROGRAMME (1975-1979)
The first R&D programme, which is now drawing to a close, had as its
objectives:
(i) to draw up an initial inventory of geothermal resources in the
Community;
(ii) to develop the technology and procedures required by exploration
for and the exploitation and management of these resources;
(i i i) to disseminate the resulting experience and information to all
the Member States.
The programme which was adopted in 1975, to run for four years, was di-
vided into five projects
(i) the acquisition and collection of basic geothermal data;
(i i) the improvement of exploration methods
(i i i) hot water sources
(i v) steam sources
(v) hot dry rocks.
This is probably the area where our programme has had its greatest impact.
I have already mentioned that the Commission, thanks to the magnificent
contribution of Dr. Haenel, has published a Community Atlas of Subsurface
Temperatures. This reference work, which collates the various national
and Community efforts and provides the profession with a suitable geo-
thermal data base, is enough on its own to justify our programme.
Indisputably, the Community initiative has helped to stimulate national
projects for collecting information on subsurface temperature gradients
and heat flow, but it must also be admitted that the quality of the data
-16 -
is often very variable, since they have largely been inherited from oil
exploration, the purpose of which, we should remember, is to locate oil
and gas and not to study thermal conduction in geological formations.
The truth is that these measurements are often heterogeneous(originating
from different loggings, bottom hole temperatures or tests carried out
with often poorly calibrated instruments when the well is not in thermal
equilibrium) and are more indicative of trends, where they are not ac-
tually wrong, than of actual formation temperatures. One might refer,
for comparison purposes, to the interesting statistical studies which
have been carried out in the Paris Basin, Aquitaine, the Po Valley and
more recently, in the Rhine Graben. I was only half joking when I said
that the temperature charts produced so far were inaccurate and that,
at the very least, they should be taken with a pinch of salt. The measu-
rements are open to doubt and the high structural position of the mea-
suring points falsifies interpolation, as anticlines are known to be hot
areas which concentrate heat flow lines, whereas synclines, especially
if they are subsiding are colder. Finally, there is the question of
convection. Important differences have been observed between forecast
and actual temperatures, amounting in some cases to tens of degrees,
especially as geothermal engineers have the curious habit of drilling in
structural lows, unlike their oil cousins who search for "traps" and po-
rosity on structural tops.
- 17-
by the subterranean reLief and regionaL variations caused by deep struc-
tures, as was shown by the heat fLow determination campaigns conducted
under the Community programme.
Thus, the strong heat fLow anomaLy which trends south-west, north east
through France aLong a hercynian path, coincides with a rise in the Moho-
rovicic discontinuity. The anomaLy at Touraine, which was for Long unex-
pLained, wouLd appear to be associated with a Large Permian graben, whiLe
the anomaLy at Landau in the Rhine VaLLey is to be expLained by deep con-
vection within the basement fractures kept open by distensive tectonics,
which appear to spread to the underLying triassic reservoir formations.
AdmittedLy, such an anomaLy couLd aLso be expLained by a straightforward
conduction modeL and it is true that there is no unique soLution to the
probLem. In the basement zones there is often a Link between high heat
fLow and radiogenesis, as has been demonstrated in Cornish granite.
Two recent decisions are LikeLy to improve our understanding of the me-
chanisms which govern subsurface temperature distribution. The first
concerns metroLogy and aims to standardize heat fLow measurement in the
Community and in particuLar, the appLication of topographic and paLeocLi-
matic corrections, to devise a standard for caLibrating the apparatus
which measures thermaL conductivity and to define the thermaL characte-
ristics of the various rocks under fieLd conditions, especiaLLy at great
depth. The second tries to reconstitute the temperature pattern within
the Aquitaine Basin, which is a weLL known sedimentation region, whiLe
making due aLLowance for convection, structuraL and LithoLogicaL condi-
tions and heat fLow distribution.
- 18-
aLLow this) on the Lines of the recent report on triassic reservoirs
in the Paris Basin. AccordingLy, the Commission wouLd Like to undertake
a structuraL report of the Rhine VaLLey, based principaLLy on the
reinterpretation of existing seismic sections together with research on
a post-jurassic tectonic modeL and an account of porosity patterns.
Seismic interpretation wiLL pLay an essentiaL roLe, just as it did recen-
tLy in the expLoratory driLling for oiL at Binnenweg, south of Strasbourg.
The project for testing reservoirs of geothermaL interest which have been
negotiated by petroLeum and mining exploration crews has not yieLded the
expected resuLts. It has in fact been difficult to reconcile the main and
the secondary objectives in one and the same project and to modify, even
sLightLy, a petroLeum or mining research programme, especialLy when it
is a matter of testing a reservoir formation beneath a Large open-hoLe sec-
tion during the actual driLLing procedure. Furthermore, the quaLity of in-
formation obtained very often represents an insufficient return on invest-
ment. However, reservoir data, which are more representative because they
resuLt from productio~ testing on oil-production welLs which were initial-
Lyat thermaL equiLibrium have been obtained in Nctthern Ireland, EngLard,
the Rhine Graben (at Landau and FeLdkirch) and Northern Italy.
It is quite possibLe that, in future, cooperative projects or joint
ventures, wiLL be formed for muLtipurpose expLoration which wiLL make it
possibLe to devise adapted expLoration and driLling programmes. Such
operations would be made easier if, as we beLieve, certain expLoration
programmes receive in one way or another a significant amount of direct or
- 19-
indirect pubLic assistance. More geothermaL engineers wouLd aLso have to
acquire the technological skills needed for this type of well servicing
and operations.
DownhoLe instrumentation
DownhoLe Logging which, with "Tot co" or singLe-shot manoeuvre, wilL give
an instantaneous temperature and recording for the mud during driLLing
wiLL be the subject of deveLopment research. CoupLed with the continuous
monitoring of the mud temperatures on the surface, this kind of Logging
- 20-
will make it possible (through adapted modelling with the well treated
as a heat-exchanger) to predict the formation gradients during drilling
and detect abnormal temperatures, thus providing information which will
benefit the entire drilling business.
PROSPECTING METHODOLOGY
Geophysical methods
Generally speaking, the hopes that were based initiaLLy on the deveLop-
ment of geophysical prospecting methods specifically for geothermal pur-
poses were iLLusory, if not naive. There are three main areas of consi-
deration: structure, reservoir formation and the properties of the soa-
king fluid. There is aLso the geological context of the system which
can be sub-divided into two classes: sedimentary and volcanic (with so-
me overlapping areas of course), these affect the prospecting objective
which will take account of structural and geometric factors in the first
case, and anomalies in the second. They also affect the choice of method
which may be complicated by logistic considerations and by the penetra-
tion of the signal in the overlying formation. As to interpretation mo-
dels, it is thought that, knowing geothermal landscapes, as we do, they
should rarely be tabular with infinite extent of layers. Thus it is a
colLection of seismic, electricaL, gravimetric, magnetic and eLectro-
- 21-
magnetic methods rather than anyone particular method which will be
used for prospecting. It is not surprising, therefore, that the most
convincing results in our programme were recorded in specific integrated
projects at Mont Dore, Vesuvius, the Phegraen Fields and the Eiffel.
The Mont Dore project is exemplary for more than one reason. Firstly,
there was the exploration strategy, whereby Mont Dore was selected from
the whole Massif Central as a priority objective by virtue of its central
volcanic nature, which is more favourable for geothermal energy than the
younger, though fissural, volcanism of the Chaine des Puys. Then there
was the variety of conventional and exotic geophysical methods used which
made it possible to define the caldera, describe its geometry, if not its
filling, and to localize, within the caldera, a target which was both
anomalous and structural. The intention was that the target should be
recognizable for drilling the following autumn. Geophysical prospecting
is also being carried out on the slopes of Vesuvius to locate the top of
a fractured limestone reservoir and this too will culminate in a drilling
operation.
Few methods can rival seismic reflection and, to a slightly lesser ex-
tent, direct-current electrical prospecting. With the exception of gravi-
metries and magnetism, most of the other methods are stiLL LargeLy expe-
rimentaL, especially the electro-magnetic ones which exploit naturaL
sources such as magnetoteLLurics much favoured by our contractors. The
latter's penetrating power, sensitivity to deep resistivity-contrasts cou-
pled with its fLexibiLity are major advantages where high-enthalpy geo-
thermal energy is concerned. However, the wide variation in signal
strength, noise, the unreliability of the model, when there are pro-
nounced lateraL discontinuities, and the channeling of telluric currents
in certain sedimentary fillings can cause considerable problems of inter-
pretation, notably at low and very low frequencies. The method known as
differential magnetic soundingCwhich, incidentally, has made it possible
to characterize the pseudo-continous telluric concentration in the
Pyrenees and the Rhine Valley) has prospecting potential on a regional
scale, since it enables very extensive conduction anomalies to be loca-
ted. The latter should also correspond to geothermal phenomena, for there
to be a specifically geothermal advantage, but this is not always the case.
-22-
ExperimentaL seismoLogy can have impLications on geothermaL prospecting
as has been shown by the seismic transmission method which uses Moho
reflected (PMP) waves at criticaL incidence to "ilLuminate" from beLow
and so identify three dimensionaL abnormaL, preferabLy hot, bodies.
This method definiteLy heLped to assess the geometry of the Mont Dore
caLdera. AdditionaL horizontaL scanning by Line and fan shooting couLd
Locate in the caLdera a highLy contrasted Low veLocity area attributed
to an extensiveLy fractured basement which couLd favour hydrothermaL
convection. Both methods can offer major advantages in mountainous areas
in conjunction with EM methods to which they bring some objectivity
(the information is in the signaL rather than in the head).
The use of seismic noise is certainLy one of the most controversiaL me-
thods, for it is difficuLt to distinguish,in the frequency bands now
used, between noise of cLimatic or human origian and noise from a geo-
thermaL source or a source which is presumed to be geothermaL in the
absence of a cLearLy recognizabLe signature. Noise directivity and
aneLastic attenuation studies have not succeeded in removing this basic
ambiguity from deep earth uproar.
- 23-
One of the criticisms of our first programme was that the projects and
the geophysical themes were too widespread, and that very few, paradoxi-
cally, were situated in recognized hot zones, when one of the aims of
the programme was precisely the sensitivity of detection methods to high
temperatures. In one project, which I shall not name, a borehole more
than 3000 metres deep was scheduled to precede the geophysical field
tests which is contrary to the usual procedure and to common sense.
Our second programme will therefore investigate at the Travale test site,
directly above a known steam field which is already being exploited, all
those experimental methods which we consider have still not been properly
evaluated, even though they have been the subject of previous research.
Geochemical methods
-24-
has been particuLarLy convincing and has demonstrated the reLiabiLity of
Low-reactivity aLkaLine geothermometers in a granite environment.
Mention shouLd aLso be made here of the interesting work being carried
out in the PhLegraen FieLds on detecting diffuse geothermaL fLuid Leaks
in near-surface water and the continuous monitoring of free and mud-
trapped gas when boreholes are being sunk, about which Mr Tonani will be
taLking to us Later on.
- 25-
FinaLLy, at forefront of dry geothermaL technoLogy, a German team has
been studying the Liberation of energy of a chemicaL origin from the cir-
cuLation of a deep exchanger such as hot dry rocks. The kinetics of dis-
soLution in an acidic medium (acceLerated in the Laboratory) have shown
that the extra reactionaL energy gained was LargeLy offset by the snags
encountered - precipitation, pLugging, eLimination of surface deposits
and subsidence. The addition of NaCL may partiaLLy neutraLize these dis-
advantages by sLowing down the kinetics of dissoLution.
-26-
tions must be taken, however, in faLL-off tests with reoard to the thermaL
conditions induced by the injection of coLd water.
Low enthaLpy, which is of interest to aLL the Member States, has been
-27-
better covered in our programme than high enthaLpy, work of which is onLy
being ~one in ItaLy at the moment, aLthough some is pLanned to start soon
in Greece as weLL.
UntiL a short whiLe ago, the onLy use that was seen for Low-enthaLpy
geothermaL energy was the heating of Low-rent housing bLocks, and this
- 28-
confined the process to the upper Jurassic (Dogger) in the Paris area, a
modeL which has served as the basis, moreover, for French LegisLation on
assistance for geothermaL energy schemes. However, the need has quickLy
aris~n to diversify the modeL by i"cLuding the heating of existing pre-
mises as weLL as new dweLLings and extending the use of geothermaL energy
to agricuLture and industry, which wouLd mean, in view of new economic
constraints, that research must be done on other reservoir conditions and
heating processes.
Two courses are possibLe heat pumps or heating processes suited to Low
geothermaL temperatures.
Monitoring the triaL period when the two geothermaL doubLets at CreiL were
used for heating has provided usefuL information. Three heat pumps in se-
ries in a counter current arrangement have raised the coverage of require-
ments by geothermaL ~nergy from 40% with a singLe heat-exchanger to
64% (with heat pumps). The energy efficiency is indisputabLe, but the
economic and financiaL return are much Less certain: the investment is
onLy paid back after thirty years with a 9% discount rate. The operation
-29-
of heat pumps is continuaLLy beset with operationaL probLems, moreover,
as their coefficient of performance is stiLL inadequate and, where eLec-
tric heat pumps are concerned, depends on the scaLe of eLectricity char-
ges; the pumps themseLves are LiabLe to be switched off at peak times
(four hours a day at CreiL during the winter).
With this concept of geothermaL energy artificiaLLy derived from hot dry
rocks (HDRs), we are deaLing with an area of geothermaL futuroLogy which
has LatterLy become the subject of controversy and emotion.
- 30-
table and it is possible technologically and economically to fracture
impermeable basement rocks at depths of five to six kilometres.
One can easily understand why, give or take a few details, the artificial
geothermal energy model quickly became popular with energy planners eve-
rywhere. It made possible a release from site constraints, which are so
important, and it also raised the potential contribution of geothermal
energy from 5% of the world's energy demand to nearly 20%. From occupying
a marginal position, geothermal energy now became a serious source of
supply, with the HDR process even constituting for some people the only
long-term future for this form of energy. In other words, an engineering
outlook had superseded the mining perspective in which had hitherto al-
ways prevailed in the prospecting of geological media.
What, then, roughly is the average performance required of the HDR heat-
exchanger in order to guarantee a minimum production of 50 MWt over thir-
ty years? The answer looks like this
Heat exchange surface : 1-2 km 2
Stimulated volume : 0.2-0.4 km 3
Coefficient of recovery of heat in place 1-2%
Mass flow rate 100 kg/sec.
Maximum wellhead pressure: 100 bars.
That very briefly is the size of the problem; at the moment it is not
certain that a satisfactory soLution can be found, despite the firm faith
that technoLogy can achieve anything and it is onLy a matter of putting
up the money.
Leaving aside the economic aspects, what are the technological means
availabLe?
The absolute maximum depth so far obtained in drilling has just been re-
corded in the KoLa peninsular in Northern Karelia (9700 m, with a down-
hole temperature of 160°C !). It ~ possible therefore to drilL down to
target depths.
Among the secondary modes of stimuLating the rock which are appLied after
hydrauLic fracturing, mention shouLd be made of Leaching and thermal frac-
turing. The first expLoits the particuLar texture of granite: seLective
dissolution of the silicates resuLts in a seLf-propped fracture which
obviates the need to circuLate a heat transfer fluid at high pressure.
The second results from cooLing the rock and is a method whereby, theo-
reticaLLy, a fracturing network perpendicular to the hydraulic fracture
-32-
is deveLoped as a resuLt of the tensiLe stresses due to contraction of
the rock voLume. ThermaL cooLing is currentLy used for the stimuLation
of geothermaL weLLs in IceLand and New ZeaLand in rocks which however
exhibit some porosity.
The strategy adopted by the Commission has been a simpLe one. On the prin-
cipLe that a muLtipath connection between the weLLs must be estabLished,
it was essentiaL :
So much then for the theory. But experience has confirmed the third the-
sis (muLtipLe fracturing).
I must just say another word about the second project, namely that at the
Camborne School of Mines in Cornwall. The thinking behind this is based
on :
-~-
The probLem, then, is threefold:
(i) Can the process which has been given an expLoratory test
near the surface be applied effectiveLy at greater depths?
The probLem is firstLy one of rock mechanics and the onLy
solution is to carry out an experiment with a more represen-
tative state of stress, i.e. at a minimum depth of 2000 m
where temperatures, although substantially high, will still
be manageable;
(ii) There is the risk that the variations in the hydraulic con-
ductivity of the fractured volume of rock might induce pre-
ferential fLow paths which will restrict the effective heat
transfer area. Leaching couLd pLaya vaLuabLe roLe here to
beat channeLLing.
Thus, no sooner is one probLem apparently soLved than others arise which
were not anticipated at, the outset. That is a measure of how fascinated
we are by stimuLating, dynamic research of this type. Even if it does not
resuLt in the hoped-for goal - the recovery of energy everywhere from
heat in pLace - such research wiLL probably have a beneficiaL effect with
-~-
regard to the stimulation of dry geothermal wells or the extraction of
energy from shallow thermal anomalies mainly in volcanic zones. And
talking about volcanoes, why not envisage the possibility one day of re-
covering energy from the molten magma - this would provide the ultimate
challenge for geothermal energy, but I shall leave Mr Tazieff to talk
about that prospect in greater detail.
It has been our pleasure to work with you over the last four years and
we are ready to continue our cooperation in both research and development
projects which are bold and imaginative.
Geothel"'lllal SHes
Locatlon H-Ip
o.r tne Selected
I
flesearch,
Deyelopment.
.nd
DelllQnstrat.!on
I pt"O!irUM
...00000000
- 36-
KEYNOTE SPEECH
by
Haroun TAZIEFF *
Although I am certain you know more about the history of geothermal energy
than I do, may I begin by giving a brief historical review of this indus-
try as seen through the eyes of a vulcanologist.
Exactly half a century has been lost - from the time exploitation began at
Larderello to the start-up of production at Warakei - and think how many
thousands of millions that would represent in any currency - as a re-
sult of the sometimes innocent and sometimes deliberate refusal to see
the relationship between vulcanicity and high temperature geothermal de-
posits.
-37-
The second reason is that it was mainly the oilmen - since Larderello is
in an area of sedimentary structure - that carried out both the prospec-
ting w9rk before drilling and the drilling itself. Now it happens that
oilmen are generally and inevitably just as little informed about volca-
nic geology as the vulcanologists are about petroleum geology. This led
to somewhat negative results. And since failure acts as a brake on deve-
lopment, it is not surprising that little effort has since been made.
Larderello was considered a freak of nature, something exceptional and
hence too unusual to merit costly prospecting.
A third reason was the extremely low price of oil and a fourth reason,
the relative ease of building thermal power stations at any location.
A geothermal power station on the other hand, can be located only at the
site where productive drilling is possible: why then hunt everywhere
for hypothetical hot water when it's so easy to build an oil-fired power
station anywhere one wanted with fuel costs (then) so low?
It was not until 1955 in New ZeaLand and then in the United States that
high enthaLpy geothermaL energy moved into its second stage of develop-
ment, a development which was then to embrace the whoLe worLd. Digressing
for a moment - since this has nothing to do with volcanic systems apart
from the rifts with which volcanic activity is always associated, I wouLd
- 38-
like to pay tribute to Belgium for having built in 1948 a small-scale
geothermal power plant at Kitwaba in Katanga; it was thanks to the intelli-
gence of a good thermodynamics engineer, Mr. A. Rollet, that enough elec-
trical current was produced to exploit a cassiterite deposit with a heat
source of only 91°C. Unfortunately, the essential economic requirements
of this situation were not mirrored elsewhere, particularly when oil was
costing only one dollar per barrel ••• Naturally, the geothermal KWh pro-
duced at Larderello in the first instance and then at Wairakei and by
the geysers of Iceland was much cheaper than the KWh produced from an
oil-burning power plant. But what's the use of tiring oneself out sear-
ching for, inventing and adapting techniques when cheap oil was abundantly
avai lable ?
Turning to the present day, high enthalpy geothermal energy is now deve-
loping throughout the world: folLowing Italy, New Zealand and the United
States, Japan, the Soviet Union, Mexico and Iceland have built geothermal
plants and new deposits have been discovered everywhere from Guadaloupe
to Indonesia and from Chili to Djibouti. Europe, however, is not very
fortunate in this respect and pLate tectonics explain why no great riches
in the form of high temperature deposits are likely to be found in our
continent beyond the frontiers of the plate of which Europe is a part.
At the frontiers, however, the geothermal riches are immense thanks to the
volcanic activity in Iceland and in the Mediterranean.
- 39-
and geochemicaL prospecting stage through to the construction of power
pLants via the intermediate stages of driLLing, pipeLines, exchangers,
etc ••• and it is this technoLogy and the associated knowhow which Europe
can export, as Japan has done; this country not onLy carries out prospec-
ting work for various deveLopping countries but aLso exports turbines,
generators and other geothermaL equipment to aLL countries which need it,
the most industriaLLy deveLoped of which incLude the United States, ItaLy,
New ZeaLand, IceLand, etc •••
Coming back to Europe, there are quite a few high temperature deposits
to be discovered and exploited in Italy, Greece and IceLand of course,
France, the FederaL RepubLic of Germany and very LikeLy Hungary, Spain
and possibLy IreLand; in each case, the deposits are the resuLt of reLa-
tively recent voLcanic activity. But Europe has enough Low enthaLpy geo-
thermaL energy to free itseLf from the major part of its burden of expen-
ditures on fueL and uranium used directLy or indirectLy as electricity
for heating purposes. The heating of dweLLings, administrative and commer-
ciaL premises, pubLic buiLdings and in agricuLture, stock-rearing and
some industriaL appLications represents a vast caLorie consumption for
Europe. To produce these caLories within one's own frontiers for direct
use, without any energy transformation - i.e. without costLy intermediate
instaLLations and an unnecessary and vast enthaLpy production as ine-
vitabLy required by eLectricaL heating - is a simpLe matter if the normaL
heat generated inside the earth is used. No need for the vitaL thermal
anomaLies to Locate the very hot underground water needed to produce elec-
tricity : the normaL average gradient of 30 0 per kiLometre and the normaL
heat fLow are sufficient. After too many years of deLiberate setbacks,
-40-
France is now setting the exampLe and is deveLoping a heating programme
which uses water at temperatures of between 600 to 90 0 found in the ma-
jority of sedimentary basins and at economicaLLy accessibLe depths,
i.e. 1500 to 2000 metres. Some 15 000 dweLLings are ~Lready heated by
thermaL water and the number should soon increase to 40 000.
Here again, the prospects for exporting technology and equipment both to
the Third World and to other industrialized countries are immense. Even
to hot countries, paradoxically, since geothermal hot water can also be
used to produce frigories as well as calories. And it is thanks to a
Swedish, i.e. European, process (ElectroLux) that this production of coLd
has been made possible. Not only can this cold contribute to the general
well-being in tropical regions but can also - and above all - be used to
conserve perishable foodstuffs such as fish, bananas, meat, etc •••
This provides Europe with an outstanding opportunity not only to improve
the alarming food supply situation in the Third World but also to re-esta-
blish Third World confidence in its former colonizers and to improve the
present relationships between these countries and Europe.
-41-
an installed capacity of 50 000 MWe foreseen for the year 2000. Japan is
already the main exporter of geothermal turbines. I am often considered
as being an over optimistic geothermal supporter but I acknowledge I could
hardly believe this goal was realistic in spite of a favourable geodyna-
mic context. But my last visit to Japan has convinced me that this coon-
try is about to achieve this ambitious objective. Europe should benefit
from this challenge which will offer many opportunities in saving energy
imports and favouring export of geothermal know how, technologies and
goods.
I often feel like a medecine doctor who has to decode numerous indices
in order to establish a diagnostic. 50 being myself a doctor experienced
in earth illness (or health who knows ?) my answer will be : there is li-
ttle future offered to geothermal energy in the exploitation of volcanic
kinetic and moreover thermal activity. The reason is quite simple : this
type of energy is too irregular and random and the fluids are so agressi-
ve (high temperature and hyper acid gases mainly) that it is not worth
attempting its domestication.
-42-
SESSION I: Regional exploration - Reservoir assessment
-43-
GEOTHERMAL SYNTHESIS OF THE UPPER RHINE-GRABEN
F. MUNCK * and
Prof.Dr. K. SAUER **
F. WALGENWITZ * Dr. R. TIETZE **
Abstract
In the alsatian and badean part of the upper Rhine graben, four
water bearing formations display characteristics suitable for geothermal
energy, which are the Rauracian-Sequanian formations, the Great-Oolite,
the upper Muschelkalk and the Bundsandstein.
The shape of these aquifers has been obtained with the help of more than
300 basic bore-logs, of data from geophysical investigations, of geolo-
gical maps and reports, and is represented on maps with their depth,
thickness, lithology, temperature and salinity, at scale 1/250.000.
The maps of temperature distribution as well as of mineral content of
the flowing fluids have been produced from data obtained in oil and
thermal water wells. All other available data referring to porosity,
permeability and piezometric head are given in tables. The Bundsandstein
offers in the northern part of the investigated area, due to its depth
and its thickness, the best geothermal reservoir conditions. In the South
on the opposite, the Great Oolite and, in part the upper Muschelkalk, are
the formations with the main geothermal potential.
-45-
I INTRODUCTION
The study presented below has ben performed within the frame work of
the European Community programme decided upon in 1975, in order to achieve
an inventory of the existing areas of potential geothermal interest.
The upper Rhine graben area, from Basel to Karlsruhe is in fact a
geological unit which led the Geologisches Landesamt Baden-Wurttemberg and
the Service Geologique Regional (BRGM) to present a german-french joint
venture for the acquisition, collation and interpretation of existing data.
The upper Rhine Graben belongs to the middle european rift system
where the geothermal gradient is usually higher than normal, in possible
relation with the tectonic activity which began 45 million years ago.
This intense tectonic activity causes movements of the existing faults in
the sedimentary formations and it results in a complex parcelling, mostly
unknown and conditioning the feeding and the exploitation of the aquiferous
reservoirs. The exploitation of geothermal energy would be of evident
interest in an area with a very high density of energy consumers,indus-
trial and urban. It was therefore primordial, for both countries, borde-
ring the upper Rhine river, to achieve the collation of all existing data
in order to state precisely the knowledge of the available geothermal
potential and to underline the gaps as well as the means for improvement.
The study has been performed in close cooperation and the results
are presented in bilingual and sometimes complementary german-french text
and maps, in order to respect local and specific geological aspects.
-46-
II RESULTS
The numerous available data from various sources and origin, such as
governmental files, specific studies and regional synthesis, have been
carefully compiled in order to present a coherent synthesis covering the
whole area.
3. Selected favourable areas: Compared with the Basin of Paris, where the
general shape and caracteristics of the deep aquifers have been proven by
several geothermal bore holes, in the upper Rhine graben two areas seem
to be of evident interest in respect to their size, temperature and loca-
tion as to the concentration of energy consumers :
a) - in the southern part, the Great Oolite formation, at a depth of 1500-
2000 meters similar to that of the Dogger in the Basin of Paris, where the
temperature lies between 80 - 110°C, i.e. 20°C higher than over there.
Recent experimentation for injection and an oil exploration bore hole
drilled in 1978 have proven the permeability of the formation. The favoura-
ble area extends :
- 47-
- on the left side of the Rhine, over 300 km 2 , covering the Alsatian
potash basin and the Graben of Dannemarie, between the Western limit
of the Rhine Graben and the horst of Mulhouse.
III CONCLUSION
The collation and synthesis of existing data represent the first
necessary step of the inventory of geothermal resources. The preliminary
results, cleared up and locally favourable, led to propose to set up the
financial, legal and technical means in order to take the best profit from
all new available data. This is primordial to precise the results in order
to guide the exploration and the exploitation towards the most favourable
areas.
-48-
IV BIBLIOGRAPHY
The final report delivered to the EC contents :
- 140 bibliographic references covering geology, geophysics, geothermics
and drilling
- a complete listing of 366 bore holes from which basic logs are available
- 16 graphs showing temperature distribution and correction
- 41 diagrams showing chemical caracteristics of thermal waters
- a complete listing and a contour map of all recent deep investigation
geophysical surveys.
-49-
NEOTECTONICS AND THE GEOTHERMAL ANOMALIES IN THE RHINEGRABEN
Summary
Positive heat flow anomalies in the Rhinegraben rift system
are mainly found in its central segment. This part of the
~raben is (a) subjected to a compressive shear strain, and
lb) a thick intercalation of a Triassic high-porosity sand-
stone exists there, between basement and graben fill. The
compressive shear strain (a) causes open fractures in the base-
ment with hydrothermal convection. The sandstone body (b)
enables a horizontal dilution of the hot water, ascending from
the deep fractures, with meteoric water from the recharge area
on the graben flanks. In the overlying graben fill, local
peaks are observed which superimpose a general geothermal high.
Such peaks are mainly controlled by active faults which are
superposed on high temperature reservoirs of the deep-seated
fissure systems. For geothermal prospection the mapping of
Quaternary active faults on the graben floor is recommended.
Numerical calculations of the temperature effect of hydro-
thermal convection show that the maximum temperature anomaly
and the time span of balancing out the temperature depend on
(a) the discharge, and (b) the maximum depth of fissures in
which convection occurs. The lateral extension of a geothermal
anomaly is controlled by the period of water flow. Some nume-
rical results are presented.
- 50-
1. GEOLOGICAL OBSERVATIONS
Since MUllerried (1921) it is known, that the tectonic
history of the Rhinegraben evolved as a 2-stages rifting.
From mid-Eocene to lower Miocene times an extensional rift
valley has been active, with external wedge-block subsidence
and shoulder uplift, and internal tilt-block rotations.
Interrupted by a period of an extinct rift valley, over about
12 my, rifting reawoke again during Pliocene time. This is the
still active stage of rifting which is controlled by a sini-
stral shear strain. Fault-plane solutions of earthquakes
(Bonjer and Fuchs, 1979) and in-situ stress determinations
(lIlies, Baumann and Hoffers, in press) had confirmed a SE-NW
directed horizontal component of maximum compressive stress,
oblique to the graben axis.
The primarily extensional rifting had conditioned a slight
zigzag configuration of the graben system. When secondarily
remodelled into a sinistral shear zone (Fig. 1), a compressive
shear strain became implicated for its central segment. Exten-
sional shear, on the other hand, came about in the northern
segment and in parts of the southernmost segment, too. Con-
sequently, the central segment came under dragging, uplift
and erosion, whilst subsidence and sedimentation occured at
those segments that were under extensional shear (lIlies and
Greiner, 1979). The variation of the regional stress and
strain conditions in time had implicated some horizontal shift
of hydrothermal activities. Mineral-bearing fissures of Mio-
cene age on the graben flanks are mostly free of present-day
hydrothermal activity. In the same way, the eastern master
fault of the central segment, primarily a dip-slip fault,
became reactivated as a reverse or oblique fault (111ies and
Greiner, 197B), and the hydrothermal circulation on it is now
blocked up by compression. The active hydrothermal peaks, on
the other hand, are connected with those fractures which are
under extensional motion as controlled by the recent time
stress conditions. By different methods and models Buntebarth
(1979) and TeichmUller and TeichmUller (1979) have shown, that
-51-
the active geothermal activity of the graben was in existence
not before Plio-Pleistocene times.
The central segment, between about Strasburg and Heidel-
berg (Fig. 1), is the very part of the graben where the high-
est heat flow anomalies have been determined (Wohlenberg and
Hanel, 1978). In addition to the compressive shear strain,
this segment is characterized by a thick intercalation of Bun-
ter Sandstone, between basement and sediment fill. The isopach
lines of this high-porosity sandstone (Boigk and Schoneich,
1974) reveal a thickness maximum which is likewise congruent
with this segment. The sandstone allows horizontal water cir-
culation at a depth between 2000 and 4000 m, what may explain
some deviatory geothermal behavior of that segment (TeichmUl-
ler and TeichmUller, 1979).
The compressive shear motion along the central segment of
the graben has been mainly taken over by the pre-existing
pattern of 1 st generation extensional faulting. Discrete 1 st
order shear planes are only locally observed. Mostly the
given normal faults had been reactivated as 2nd order shear
planes or Riedel shears, scattered aver the whale graben floor.
Relevant neotectonic faults or morphological scarplets appear
often en echelo~ arranged and sigmoidal shaped, like the fea-
ther joints, first described by Hans Cloos in 1928. Present-
day faulting is additionally indicated by a moderate seismic
activity along some of these shear elements. At one feature,
mapped near the village of Rittershoffen (Fig. 2), the maxi-
mum geothermal gradient of the Rhinegraben has been found.
In one bore-hole a temperature of 184 °c has been measured
at a depth of 1555 m only (Marie, 1952). At another active
fault scarp on that map (Fig. 2) a warm spring was found.
In accord with the mechanism of feather jointing, the ob-
served features exhibit an extensional character. Mini-grabens
and local horsts are exposed in near-surface outcrops (lIlies
and Greiner, 1978). Their trend is often about 170 0 , fitting
the direction appropriate for the release of the given stress
direction by 2nd order shear. The unconsolidated sediment-fill
of the graben reacted by the formation of extensional collapse
features like the observed mini-grabens.
- 52-
Provided that the rocks are consolidated, to condition a
tectonically competent behavior, open fissures will be formed.
This will be the case in the Bunter Sandstone and in the Her-
cynian basement at the socle of the graben. Consequent hydro-
thermal convection will be put in action with a maximum tem-
perature as related to the depth of the fissures. Mineral
veins will be formed as observed at many places on the graben
shoulders. For geothermal prospecting mainly those fissures
are interesting where the rate of horizontal shear will be
higher than the rate of mineralization.
The boundary conditions as mentioned above and other avail-
able geological and geophysical data are the basis for this
generalized geothermal section across the central segment of
the graben (Fig. 3). Shear controlled fissures with hydro-
thermal convection are observed in the basement; horizontal
circulation may be seen in the lower-temperature reservoir
of the Bunter Sandstone, which is in connection with the re-
charge in the outcrop area of the graben flanks. The Bunter
is overlain by the younger Mesozoic cover and a pile of Ceno-
zoic fill sediments. These fine-grained and often ductile rocks
are causing a more conductive heat transfer. The more equal-
ized configuration of the geo-isotherms is superimposed by
local peaks on top of the fissure systems in the basement.
Active normal faults in the graben fill, which can be often
mapped on the surface, are tracing about the shear controlled
fractures in the basement underneath.
2. MODELLING
In order to evaluate the temperature effects of hydrother-
mal convection, a simple two-dimensional model is proposed
(cf. Fig. 4), which is based on the data given by Delisle
(1978).
At the point A of Fig. 4 water of a constant discharge Q
is entering the system with a constant temperature, descending
vertically in a fracture to point B, from here on horizontally
to point C, then rising with the double discharge 2 Q (due to
reasons of symmetry) to point 0, flowing back towards point A.
A similar model has been used by Lowell (1975). Fig. 4 shows
- 53-
the relevant material constants. In the model proposed on Fig.
4, temperature regime is controlled by Fourier's equations
.1L aT
of C) t
At the interface rock-water conservation of energy requires,
provided that K of rock is finite in the direction perpendi-
cular to the flow, and equal zero parallel to it, and provided
that the temperature of the fluid remains constant across the
fracture (Carslaw and Jaeger, 1959, 396), e.g. for the flow
from 0 to A (Fig. 4):
K
( aa~T+ c> r-) =M'c'(aT
at
a~
TABLE
Z
max 2 Q ~ Tmax tT
max
km kg cm -1 year -1 °c years
6 100 16 200,000
6 400 51 200,000
6 1,000 79 60,000
6 2,000 82 20,000
9.5 100 24 1,000,000
9.5 400 90 900,000
- 55-
~ SUbSidence] of the
graben
~ uplift floor
~ fault zone of the
nft system
[ZJ Jura told belt
~
~Alps
aI
km 50
I
SfEINCElO
\ .
o km 5
'e.~iM.Y .
/ historic:: earthquake
- 57-
~\
.". .0' ('~
4~ ESE
o
1
2 ,
; ()
~
t) o
~ Middle TriassIc to Quaternary, mainly r---"I high porosity sandstone r-:-l Hercynian basement,
00
'" L-....iJ marls with conductive heat transfer ~ Bunter (Lower TriassIc) ~ predominantly granillc
/ fau lt zone V open fissure with hydrothermal convection ,_,,' geoisotherm, schematically
20 t c = O· 2
9" 2.6
included
I
j
hydrothermaL
a I convection
J . . - - - - ...... ----.J
C 8
:r
!"z
I
TX
I
I
plane of
symmetry
........
....... roc]
km
-z
- 59-
REFERENCES
lIlies, J.H. and Greiner, G., 1978. Rhinegraben and the Alpine
system. Geol. Soc. Amer. Bull., 89: 770-782.
-60-
Lowell, R.P., 1975. Circulation in fractures, hot springs, and
convective heat transport on mid-ocean ridge crests. Geo-
phys. J. R. astr. Soc., 40: 351-365.
-61-
DETERMINATION or THERMAL CONDUCTIVITY AND DlrrUSIVITY or
RHINEGRABEN-SEDIMENTS WITH A NEW RING SOURCE DEVICE
G. SAT TEL
ABSTRACT
A new method was developed to measure the temperature field on
the surface of a sediment specimen. This surface was heated by
a ring source. The time dependent temperature values in the
centre of the heated ring give the volume dependent specific
heat, thermal conductivity and thermal diffusivity. It is not
difficult to prepare the specimens because it ist only neces-
sary to have a smooth but not polished surface. Measurements
with variable temperatures and variable hydraulic uniaxial
pressures up to 400 bars are possible. The time for one measure-
ment is not longer than one minute. For this reason this device
makes large series of thermal measurements possible. More than
800 determinations of thermal parameters of Rhinegraben-sedi-
ments have been carried out with these results:
The thermal conductivity and the volume dependent specific heat
increase with depth. That means, that the thermal conductivity
is a function of density, which also increasffiwith depth. These
results can explain the negative slope of temperature gradients
messured in boreholes near the Landau oilfield and the product
of temperature gradient and thermal conductivity gives a con-
stant heat flux of 130 mW/m 2 •
- 62-
1. INTRODUCTION
As a part of an EG research programme thermal conductivity and
diffusivity determinations were carried out by the Geophysical
Institute of Karlsruhe University. They are important for the
exploration and interpretation of the abnormal subsurface tem-
perature field of the Rhinegraben.
The specimens were sediment cores of the landau oilfield, most-
ly clay and marl, many of which han broken into pieces due to
drynesse, pressure, release and long-time storing.
The rest of the compact sediment material was partly breaking
into pieces when handled mechanically. Therefore stationary
conductivity measurements, which require considerable prepara-
tions of specimens, are not possible. for this reason the time
dependent ring-source measurement equipment was developed.With
the help of this probe a fairly representative value of ther-
mal conductivity of the core can be measured in a minimum of
time and with simple specimen preparation.
-63-
4 1 / KI
DL--:1D~-':':::D-'±D--:':
••~-4'''~I'/
Graph 1:
Probe arrangement, temperature-time function and differentiated
temperature-time function. ~ Thermal diffusivity
SC Volume dependent specific heat
K : Heat conductivity
Error sources are mainly the heat capacity of the heater and
thermocouple. Partial heat flow resistances of the heater are
balanced by the integrating effect of the ring-shaped source-
points. When dealing with specimens with layer structure it is
of great importance to consider the different heat flow perpen-
dicularly and parallel to the layers; this problem occurs es-
pecially when measuring marls and clays. The theoretical rela-
tion between the maximum of the differentiated temperature
function and thermal parameters is no longer valid, because it
is derived from a solution of the heat conductivity equation
for a homogeneous medium. (graph 2)
- 64-
Graph 2:
Isotherms of a ring source. Only the positive quadrant
(x,y,O, area of dotted frame) is shown.
Right: layers perpendicular to the plane of the ring source
(x,z - plane)
Middle: Homogeneous
left: layers parallel to the plane of the ring source
(x,z - plane)
In these cases, models of the ring source with layered speci-
mens were calculated with finite difference programmes. They
gave the following results which lead to the determination of
the anisotropie of thermal conductivity parallel and perpendi-
cular to bedding:
a) The experimentally determined K and ~ are roughly the arith-
metic mean of K and ~ of the layers, if the layers are
parallel to the source plane.
b) The experimentally determined ~ is roughly the minimum
of the layers. The value determined for K lies between
minimum K and the mean of K of the layers, for layering
perpendicular to the source plane.
- 65 -
o
2
3
. '• • @
. @
. . ..
'0 .....,.... ..
"r;
,]
"
IS
OS
J7
'9
20
Graph 4:
Volume dependent specific heat and density as function of depth
+ marls and clays; • sandstones; Gl chalky sandstones;
for this reason, the thermal conductivity also increases with
depth. The slope of the conductivity - depth function decreases
with depth, because the effect of compaction OCCODBS mainly in
the upper hundred metres. (graph 5)
The thermal diffusivity also increases with depth (graph 6),
but much less than the conductivity (graph 5), so that the
ratio of conductivity and diffusivity , the volume dependent
specific heat also increases slightly with depth. (graph 4).
CONCLUSION
Influence of saturation, pressure and temperature:
The resaturation of cores gives roughly 30% higher thermal con-
ductivity. But this procedure is difficult, because most of the
sediment cores begin to desintegrate, due to swelling of clay
minerals and salt. The ensuing destruction of pore structure
is only avoidable if pressure of SO - 100 bar is applied to the
speciman.
New meaeurements with pressures up to 400 bars showed that the
thermal conductivity increases only slightly with pressures.
Up to 400 bar, this effect lies in the error range of +/- 10~
of conductivity measuremant.
Increasing temperature gives lower conductivity. But this effect
is very small for clays, marls and sandstones compared with
rocksal t.
- 66-
The ratio of conductivity parallel and perpendicular to the
layering was found to be 1.05 -1.20 in most cases.
1~
,
, '.....
'.'
!Jr.!
~-~_Rho
1,5 2 1,5
Rho
1.5 1.5 3
l;!;r!
I
KofO oJ
'<l!1
J
I., 2 1.' J
Rho
t;!;r!
Graph 3:
Thermal conductivity as function of density
upper graph: marls ans clays
middle graph: marly sandstones
o chalky sandstones
lower graph: chalks and dolomites
The increasing density is due to compaction of the sediment
with greater depth. (graph 4)
- 67 -
1 3 0 ~1 0,' QS aI 1.1 I.' Xo,O·'
1 3
-K'K,J "' I¥,-IJ
, I~/
1
.... -
3
~ -:
8
9
10
11
,1
13
"
15
16
to
17
10 Iml
18
19
20
z·v l
1m/
Graph 5: Gra~
Thermal conductivity of clays Thermal diffusivity of clays
and marls and marls
(upper abscissa: dry core) • ~ parallel to bedding
(lower abscissa: resaturated) + X perpendicular to bedding
• K parallel to bedding
+ K perpendicular to bedding
- 68-
... -..
_ I'\DW
..... -
~RA1\R! " - <XNIUCTMTY
M "",,,--I I'""-~ ~'-1
10 • ••
T a
•z " •
~l
.. ...
D
'" '"
zH zH
Graph 7
Temperature log, gradient log, thermal conductivity log and
heat flow log of a typical Landau oil well.
- 69-
NUMERICAL CALCULATIONS OF STRONG TEMPERATURE GRADIENTS
AT INTERFACES OF HORIZONTALLY STRATIFIED SEDIMENTS
G. J. SCHROTH
Abstract
Measurements of temperature gradients carried out in oil
wells immediately after interruption of production frequently
show negative values. Such negative values are in contradicti-
on to a stationary temperature field. The temperature field is
calculated with the help of an Explicit Finite Difference
scheme for a simulated production process during and after
production. The gradients formed at the interfaces of sediment
layers,which are identified with that of the thermal diffusi-
vities, are called strong gradients. Their dependence on the
duration of the production process and on the contrast of the
thermal diffusivities, and also their growth and disappearance
after production stops are shown and discussed.
-70-
1 • INTRODUCTION
-71-
the gradient log of the production well LA-82 36 hours after
the end of production. LA-89 (on the right) has been inopera-
tive for six years. The peaks at 100 m depth indicates the dip-
ping of the tool from the air into the oil or water.
computations with a fine resolution lead to the following
results:
- The strong gradients increase with time at interfaces between
layers of different thermal diffusivities when the cooling
process is started.
- A short heating time generates bigger strong gradients than
a long time. The boundary condition at the radius r=R metres
is +20 0 C when the production process is started. R is the out-
side diameter of the casing. (Fig.4)
The appearance of the maximal value of the strong gradient is
established quicker after a short heating time. (Fig.5)
- The maximal value is dependent on the contrast of thermal
diffusivities between the two layers. (Fig.6)
- A negative gradient is induced if a sediment with a good ther-
mal diffusivity is coverd by a sediment with bad thermal dif-
fusivity.
- If there is only a small difference in the thermal diffusivi-
ties the maximal value appears earlier than when the diffe-
rence is large.
- It is assumed that there is no interaction between two strong
gradients i.e. the thickness of each layer is great enough.
If the upper and lower limits of the contrast of the thermal
diffusivities are given, one can determine the time at which
the gradients at the interfaces will be maximal for a chosen
heating time.
The results suggest a more detailed experiment in order to
detect the strong gradients and to get more information about
thermal diffusivities and their variation with depth.
-72-
a) b) Grodent (Kim)
lIDO
:~:.r~i~
700
IlOO
:~~rQ!l'~
!lOCI
:~:.~t~
400
~:ct~r~t~
600.
lOO
~~:.r~t~
20D
~:t~ro~lt~
100
~~~~~
Parameter
Radius (m)
- 73-
.. - -.
~ ~.
-,f
a) b)
d) c)
LA-82. LA-89.
Gradient <-C/m) Gradient ("C/m)
0.0 0.2 0.0 0.2
o o
100
200 Q...200
."
"'0
:r300
,... ""
~400
...
(II ...~400
(II
~500 3500
600
700
-74 -
200
dT
dZ
r*J
100
• T~)
I
~.-T-.--.
60
50 r--/
/
/
/.
."
/ ----f--
.".,.
/L K/K, • t<..j.... :
'
:3
/
v·
40 1--
i
- - --
Fig.5: The time when the maximal value of the strong gradient
appears as a function of the heating time. The maximal
value appears earlier for a short heating time than for
a long time. The contrast of both the thermal conducti-
vity and the thermal diffusivity is 3:1.
- 75-
\\
\
300
~
c;-w.....t
eC/....:J
.
200
\.
\.
'."
100
o
~
I " ............./.,...." -,-
.I I•
.-'
o 2
LITERATURE
Beck, A. E., Jaeger, J. C. and Newstead, G.: The measurement of
the thermal conductivities of rocks by observations in
boreholes. Australien Journal of Physics, Vol.9 286-269
Carslaw, H. S. and Jaeger, J. C.: Conduction of heat in solids,
Oxford 1959 (-. ed.)
Kappelmayer, O. and Hanel, R.: Geothermic with special referen-
ce to application, Geopublication Associates 1974, Gebrli-
der Borntrager, Berlin-Stuttgart
Sattel, G.: Aufbau einer MeBapparatur zur Bestimmung thermi-
scher Parameter von Gesteinen und Messungen an tertiaren
Sedimenten. Diplomarbeit am Geophysikalischen Institut
Karlsruhe
-76 -
GEOTHERMAL ANOMALIES IN THE RHINEGRABEN SEDIMENTS AND THEIR EXPLA-
NATION BY UPRISING DEEP GROUNDWATER FROM THE CRYSTALLINE BASEMENT
1
Institut fur Geophysik, ETH Zurich,
CH-8093 Zurich, Switzerland
2
Geophysikalisches Institut, Universitat
Karlsruhe,
0-7500 Karlsruhe, Hertzstrasse 16,
Federal Republic of Germany
Abstract
Geothermal anomalies within the Rhinegraben (Germany) are interpre-
ted postulating the rise of deep groundwater along faults in the base-
ment. The model described in this paper is related to the well-known geo-
thermal anomaly of Landau/Pfalz. Relationships are investigated between
a) the temperature in the centre of the anomaly, b) its lateral extension,
c) the water flux, d) the original depth of the water, and e) the age of
the anomaly. The numerical approach takes into account two processes:
Thermal conduction and convection in a porous medium. The estimated age
of the anomaly is about 100,000 y. The required water flux is very small,
and the source depth of the water flow is calculated to be about 6000 m.
The presented model of water flow can also be used to interpret the
salinity anomaly in the Landau field. An observed zone of fresh water
(pore water in the sedimentary cover) coincides with the central part of
the thermal anomaly. Using a salinity model corresponding to the hydro-
thermal model the extension of the fresh water zone can also be simula-
ted.
-77-
1. GENERALITIES
The sedimentary cover in the Rhinegraben zone extending from Frank-
furt to Basel is geothermally characterized by high subsurface tempera-
tures (1, 2, 3, 4, 8, 11, 12, 13, 14, 15, 18). Another problem is the
existence of a deep-seated geothermal anomaly in the upper mantle under
the Rhinegraben (7, 16) which is not the subject of this paper. The aim
of this paper is to find an explanation for the observed geothermal ano-
malies in the graben fill. A well-known significant anomaly is located
in the oil-field of Landau/Pfalz (Fig. 1). OUr considerations are restric-
ted to this special case, but they can be applied to other prominent ano-
malies within the Rhinegraben zone as well (e.g. Soultz/Pechelbroun,
Stockstadt). The following investigations are based on new continuous
temperature logs in oil wells of the Landau field. The logged depths
range from 800 to 1300 m. Further details about these measurements are
described in another paper (9). The results of the measurements are sum-
marized in Fig. lb. High temperatures exist in this area, with values of
about 1000C at 1000 m depth in the centre of the anomaly. This corres-
ponds to a heat flow of about 120 mW/m2.
-78-
_+-___- - 45°
__- - - - 5 5 °
b oI
2
!
KM
Fi.g l.
a) Location of Landau
b) Isotherms and fault pattern of the Landau field at 500 m below earth
surface. The numbers indicate temperature values in °C. The most im-
portant fault is the W-fault
- 79-
ding quantitative relations between:
e) the thermal water flow, i.e. the quantity of water transported per
unit time.
2. THEORETICAL ASSUMPTIONS
We consider a two-dimensional case, assuming an isothermal surface
condition and an undisturbed temperature field before the forming of the
anomaly.
We distinguish two regions: The sedimentary cover, and below this,
the crystalline basement, taking into account that permeable zones exist
in both regions. The most important point of our model calculation is
that two processes of heat transport take place: thermal conduction and,
at the same time, convection caused by moving waters within the permeable
zones.
In the permeable zones of the graben fill the flow of water is go-
verned by the Darcy's law. No accurate description on the special mecha-
nism of the permeability of a fault is available. We avoid this diffi-
culty by assuming that within the fault zone the definition of a Darcy
velocity is still approximately valid. Using this assumption, the two
regions (sediments and crystalline) can be handled with the same mathe-
mati cal formulae.
In our model the Darcy velocity field is a given input quantity and
is assumed to be constant, starting at an initial time t = O. We use a
basic equation, the applicability of which has already been successfully
tested for shallow aquifers (17). The equation is
y ~ grad T + oT = D V2 T (1)
ot
-~-
where
T temperature,
t time,
+
u vector of the Darcy filtration velocity
y ratio of the heat capacities of water and rock
D "effective thermal diffusivity" in water permeable zones, which
includes the effect of thermal dispersion.
3. HYDROTHERMAL RESULTS
In Figure 2 we assume that the whole graben fill is permeable and
that the flow of water is radially distributed.
The sedimentary cover has a constant thickness of 2000 m.
The fault in the crystalline basement (more precisely, the narrow
zone in which the thermal water can rise) is assumed to be perpendicular
to the earth's surface and to be 200 m wide.
2
The quantity S (flux of deep groundwater) is assumed to be 50 m /y
3
(Le. 50 m /y per 1 m along the fault perpendicular to the plane of the
figure) •
A decisive parameter in our calculation is the original depth Zo of
the thermal water. To build up a thermal anomaly as observed near Landau,
we must assume an origin from a depth of about 6000 m, if the initial
temperature distribution was normal (33 deg/km).
The result of the calculation is shown in Figure 2 in form of iso-
therms. After 100,000 years a state is reached which is very similar to
the observed temperature field in Landau. For a comparison of the model
results with the observations, we can consider the temperature at
z = 1000 m in the centre of the anomaly, on the other hand the decrease
- 81-
6 I. 2 o 2 I. 6km
! ! ! I ! !
o .}
I.
z [km]
temperature [OCl
at x = 0 and z = 1 km 5= 100m2/y
150
5= 50m 2/y
100
5= 30m2/y
50 time
o 20 1.0 60 80 100.103 [yl
- 83-
anomalies could be correspondingly interpreted with a minor source depth
Zo or smaller fluxes S. In any case a thermal anomaly would be connected
with a water-bearing fault in the basement. With other words we can ge-
neralize the model for the case of Landau and assume that hydrothermal
processes of this type are active in the whole Rhinegraben.
This supposition leads to the problem of the hydraulic system and
of its causes. Such a hydraulic system probably includes a region which
exceeds the proper graben zone.
+ aC
v grad C + at = D* V2 C (2)
where
D* = "effective diffusivity" in permeable zones (including dispersion
effect)
C concentration
t time
+
v Darcy velocity
-84-
salinity [gil)
• BNS
•
0 eM
120 •
0
MS
Mes I
100
PS
KE •
11
80
1 •
I
50
1.0
20
I X
0 W
I
NW
I
0 2 3 km SE
3I oI
2I 3km
I
n
o
20
ACKNOWLEDGEMENTS
This study was carried out under contract with the Commission of
the European Communities (Project No. 321/79/2 EGO). We thank Winters-
hall A.G. for granting permission to carry ou~ the measurements and
especially Dr. F. Doebl for providing all necessary informations.
-86-
REFERENCES
(9) Parini, M., Werner, D.: Die geothermische Anomalie des Erdoelfeldes
Landau/Pfalz, in preparation, 1980.
- 87-
(10) Schad, A.: Das Erdoe1fe1d Landau., Abh. Geo1. Landesamt Baden-Wftr-
temberg, 4, 81-101, 1962.
(11) Teichmue11er, M.: Die Diagenese der koh1igen Substanzen in den Ge-
steinen des Tertiars und Mesozoikums des mittleren Oberrheingrabens.
Fortschr. Geo1. Rhein1. Westfa1en 27, 19-49, 1979.
(14) Werner, D., Doebl, F.: Eine geothermische Karte des Rheingraben-
untergrundes. In: Approaches to taphrogenesis, J.H. lIlies & K. Fuchs,
eds., pp. 182-191, Stuttgart: Schweizerbart 1974.
(15) Werner, D., Fuchs, K.: Exploration and interpretation of the geo-
thermal anomaly in the Rhinegraben. In: Seminar on geothermal ener-
gy, Vol.I, pp. 189-206, Commission of the European Communities,
Brussels, 1977.
(17) Werner, D., Kley, W.: Problems of heat storage in aquifers. J. Hy-
drology, 34, 35-43, 1977.
-88-
CONTRIBUTION OF MAGNETISM AND GRAVIMETRY TO THE KNOWLEDGE
OF THE ANTEPERMIAN BASEMENT IN THE RHINEGRABEN.
APPLICATIONS TO GEOTHERMY.
Abstract.
-89-
1- INTRODUCTION.
The purpose of our work was to show that magnetism in the upper
crust can be used for geothermal prospecting. The first possible applica-
tion is the determination of the Curie Point Surface. In the Rastatt
area, no clear correlation between the large negative anomaly and an
eventual uplift of the Curie surface could be evolved. The second appli-
cation is to use magnetism, as well as gravimetry in order to drawn a
structural map of the basement under the mesozoic and cenozoic sediments.
Topography, faults, petrographic variations of the basement and conse-
quently variations of the heat production by radioactivity and of the
thermal conductivities have a great importance in geothermy. A gravimetric
map was already available, but magnetic data were poor and a detailed
magnetic map was requested.
Two methods have been used. The first consists ~n the quantitative
interpretation of the magnetic map. Computation of the magnetic structure
depths exhibit mainly the top of the visean and devonian formations.
The second method uses reflection seismic data. The deepest inter-
face which can be followed with a rather good approximation is the basis
of the Buntsandstein. So a map of depths of the Trias basis has been drawn
for the major part of the upper Rhinegraben (6,7). The differences of
depths between the Trias basis and the magnetic structures give an approxi-
mat~d thickness of the pprmo-rarboniferous basins.
- 91-
6- INTERPRETATION OF THE MAGNETIC AND GRAVIMETRIC DATA.
" .
". ">:: eVHL"::
....... ? ' .. : ",.)
" ",'
j ~ .-~.~ .. ~ ./ I~'~
.: / .... . .. --:~·'··tt1!R
\~~~\ .ai--~;·~ ~\~ .~.~:;:)
..... ' -~-.~ '- .. ,
--.'
···:.
"
'
, ""
" .'
10 km
- 94-
0 ' lI(oUISfilutU
IJ " ,/
o 1- J "
F:S:\
~ Gil "
ffi!illjl ' ~"
[~ .
[J I
o "
" ,
'"'-" .,
=:1:.
' ,,
,,
-".....
I~,·",A
. i1l1iil'
•••.•• •
. _ . -
I....
~~
~
C-:::_--:":-:.
~1 0,.
, ~t
Fig. 3 - Interpretation map of the magnetic and gravimetric properties of the basement.
1: Siluro-ordovician schists - 2: Devono-dinantian formations - in black: basic volcanism - 3: gneisses -
4,5,6,7,8: granites and grandiorites - 9: high intensity basement rocks (metamorphised schists, volcanic rocks)
10: intermediary density rocks (paeozo!c sediments, diorites, gneisses) - 11: low density rocks - 12: magnetic
rocks - 13 : strong magnetic rocks - 14 : magnetic discontinuities - 15 : boreholes which have reached the
basement.
Fig. 4 - Extract of Fig. 3 and isotherms at 500 m depth from Lauer (14).
- 96-
identify the granites under the sediments (fig.3) and by the way
indicates the zones of possible geothermal anomalies. Correlations with
the neighbouring basement outcrops, would allows us to extrapolate the
location of the radioactive granites and so to define the prospection
zone.
An other application of the knowledge of the basement is the eva-
luation of the thermal conductivities. The temperature distribution in
the sediments will be different above a granitic horst and above a
carboniferous or/and devonian basin because of the differences of thermal
conductivity.
REFERENCES.
-98 -
PRELIMINARY ASSESSMENT OF THE GEOTHERMAL
SUIlDDary
The most likely souroes of water at temperatures and flow rates suit-
able for geothermal development are in Permo-Triassic sandstones in deep
sedimentary basins. The distribution and properties of these rocks, to-
gether with the quality of the water they contain, have been studied and
their potential for geothermal energy development examined. The results
are sufficiently encouraging for test drilling at selected sites.
The potential for geothermal power from hot crystalline rock has also
been assessed from existing geophysical, geological and geochemical data.
Heat flow data are limited but several granite batholiths have been iden-
tified as preliminary targets, mainly on the basis of their large volume
(derived from gravity data) and their above average radiogenic heat
production.
- 99-
1. INTRODUCTION
In such a stable area the most likely source of water suitable for
geothermal energy development is in deep sedimentary basins where advan-
tage can be taken of permeable rocks containing water at temperatures high
enough for economic development.
- 100-
from the Permian to the Cretaceous but the Lower Permian and the Triassic
sandstones are the principal aquifers with geothermal potential.
The most favourable aquifer in the Permian is the Yellow Sands and
Basal Breccia of the Lower Permian. This formation thickens towards the
east and is better developed and more persistent below the east coast
south of the Humber, but even so the thickness there is only 30 to 60 m.
The sequence mainly comprises poorly cemented dune sands wit.h a porosity
between 20 and 25%. The permeability is not known although the nature of
the deposit and its mode of origin suggests values of several hundred
millidarcies.
- 101-
been made of the depth and thickness of the sandstones at the base of the
Permo-Triassic.
Heat flow is low in the Cheshire Basin and the mean temperature gra-
dient is only 20oC/km but, nevertheless, this is sufficient to give tem-
peratures in the sandstone of 60 to 90°C. Measured "bottom hole" temper-
atures in boreholes 2300 and 3500 m deep were 59 and aooc respectively,
and these are likely to be lower than equilibrium temperatures.
- 103-
The porosity of the sandstones is believed to be high, probably
greater than 20% in many areas. {The average geotnermal gradient, indicated
mainly by "bottom hole" temperature measurements, is only 23 0 C/km. This
o
implies that temperatures of 60 C or above can only be expected over a re-
latively small area where the aquifer reaches depths of more than 2100m.
However, there is evidence that the average temperature gradient. in the
overlying Keuper Marl is greater than that in the sandstones, so the thick-
ness of the marl probably affects the temperature of water in the sand-
stones.
-104 -
tests gave a bulk permeability of 72 mD and the transmissivity is, there-
fore, 15 Dm. An underlying Permian sandstone, only 45 m thick, had a simi-
lar permeability and a transmissivity of 3 Dm. Waters from both aquifers
were sodium-chloride brines with a total mineralisation of some 120 gil.
3.1 Introduction
The lack of heat flow data for crystalline rocks has meant that the
heat flow-heat production relationship is not well defined for much of the
UK and possible areas of high background heat flow cannot be recognised.
The studies to date (5) have therefore concentrated on identifying large
crystalline bodies enriched in heat produc~g elements.
In the absence of heat flow data, selection has been based on:
-105-
and the age, origin and geophysical expression of the igneous
bodies.
The most recent activity was of Tertiary age and resulted in the
emplacement of large basic plutons at the major volcanic centres. These
should have cooled to equilibrium temperature by the present time and the
basic rocks should be relatively depleted in radiogenic elements. Above
average heat flow is unlikely to be observed at individual Tertiary
centres although a more regional effect persisting from the Tertiary
period is a possibility.
-106-
O'litClOop oj
I"lruslon,
Fi gure 1
Nesozoic basins
and intrusions with
hot dry rock
pot ential
4. CONCLUS IONS
- 107-
this may be off-set by high permeabilities and the possibility of yields
from underlying Carboniferous sandstones.
The best prospects for applying the hot dry rock concept would appear
to be in Caledonian and Hercynian granites which have high radioactive
heat production, particularly if this is enhanced by the insulating effect
of poorly conductive sedimentary cover.
REFERENCES
-108-
GEOTHERMAL POTENTIAL OF LATE QUATERNARY EAST EIFEL VOLCANIC FIELD
Abstract
The Laacher See Volcano (Eifel), which erupted 10 500 years B.P.,
is studied in order to develop methods by which volume and heat content
of non-erupted,young,high level,cooling magma bodies can be inferred from
analysis of erupted magma fractions. Approximately 1.6 km 3 of phonolitic
to tephritic magma was ejected in Plinian and phreatomagmatic eruptions.
The erupted products indicate a strongly layered magma chamber with pheno-
crysts ranging from 5 vol %at the top (san,amph,hau) to about 50 vol %
at the bottom (pl,san,cpx,amph,sph,phl,mt,ap,hau,ol). Whole rock compo-
sitions range from Si02 = 60 and MgO = 0.01 in the first to Si02 = 55 and
MgO = 5.5 in the last erupted magma. U and Th concentrations show a roOf-
ward increase from 2 and 10 up to 35 and 100 ppm. Two compositionally
distinct magmatic components with about 1.5 and 5.5 per cent MgO, both
containing varying amounts of olivine (F080), were erupted from the lower
magma levels and suggest incomplete magma mixing between evolved phonolitic
and more mafic magma, which was introduced into the chamber. Xenolith in-
vestigations and geometrical constraints suggest a cylindrical magma body
(diameter 1 km, height 2 km) within Devonian slates and graywackes. Phyl-
lites and mica-schists within the last erupted tephra place the bottom of
the magma body at a depth of 4-5 km.
-109 -
INTRODUCTION
Volcanological, petrological, geochemical and geophysical methods
must be developed to locate high level magma chambers. The Laacher See
Volcano was chosen as a testing ground for developing such methods be-
cause of its young age (10 500 years old), large volume and excellent
and abundant outcrops. Because the erupted rocks are highly differen-
tiated to intermediate in chemical composition (phonolitic to tephritic),
more primitive parent magma (basanite) must still exist as a cooling
igneous body at an unspecified depth. After eruption of the top part of
the magma column, this more primitive magma could have (a) been stationary
and crystallizing;(b) moved upwards toward the surface or (c) increased
in volume by subsequent magma influx from the mantle into the crust.
The total heat of the Laacher See magma column can be partioned into
(1) heat given off by the differentiated magma to the wall rocks
including heat of crystallization along the lower, older part
of the erupted magma column. Most heat of this magma however
was spent in the form of explosive energy during the eruption,
(2) heat within the cooling mafic parent igneous body - probably
now crystalline but hot - plus heat given off to wall rocks
since the first emplacement of the magma column into the upper
crust,
(3) an unspecified amount of heat in case of later influx of new
magma.
To solve these problems, several different approaches are employed:
-110 -
Shape and depth of erupted upper part of magma column are evaluated by
studying the types of country rock erupted. Two main boundaries were found:
between the Devonian HunsrUck and Siegen formation in the upper third and
between Siegen and phyllite near the base of the magma column. The thick-
ness of the folded Siegen formations is estimated to be about 4 km in
this area, placing the base of the erupted magma column at 4 to 5 km below
the surface and the top at about 2 km. This, and theoretical considerations
suggest the magma body to have been cylindrical. The question is open
whether the more primitive parent magma followed directly beneath.
It appears that the erupted magma column - prior to eruption - was not
emplaced as a unit but grew upwards during the evolution as evidenced by
- absence of contact - metamorphosed country rock during the
first phase of the eruption but increase of such rocks during the
later phases. This is best explained by assuming the lower
part of the erupted column to be older and the upper-most
part to have been emplaced just prior to eruption. The lower
magma was also hotter but this is thought to be less impor-
tant,
- increase in abundance of coarse-grained subvolcanic rocks
(slowly crystallized equivalent of erupted magma) during the
eruption probably indicates longer residence and thus
cooling time of lower part of erupted magma column compared
to upper one.
These findings are important because they substantiate our theory that the
magma chamber developed dynamically at high crustal levels with only the
upper part having been erupted and that the non-erupted cooling primitive
parent magma also resides in the upper crust.
The erupted part of the magma column contained three major layers showing
striking differences in number and total volume of mineral phases with
abrupt gradients, ranging from less than 5 volume per cent total pheno-
crysts (6 phases) in the uppermost part to about 50 %and 10 mineral phases
in the last erupted batch. Major and trace elements show corresponding
changes. Presently, phenocrystic mineral phases are analyzed to be
used as geothermometers. Mineral, bulk rock and glass analyses are being
modelled by computer methods in order to specify the differentiation pro-
cess by which the magmas developed from each other. Both sets of data will
- 111-
be used to extrapolate non-erupted magma volume and thus heat reservoir.
Most, if not all, near surface magma reservoirs or columns are chemically
and mineralogically zoned. An understanding of the processes leading to
such zonations is essential for estimating composition,volume and thus
heat content of the - commonly non - erupted lower part of such columns.
The Laacher See-area (East Eifel) is unusually well suited for studying
such processes because more primitive (basanite) and intermediate (teph-
rite) magmas were erupted in abundance around the Laacher See area and
several are mineralogically and chemically zoned in an systematic way.
These are being analyzed in detail at present as analog systems for the
non-erupted part of the Laacher See Volcano.
The Laacher See Volcano is the youngest of 3 similar magma systems
the two other ones occurring about 3 km to the west; Wehr, active about
300 000 and Rieden, active about 650 000 years ago. Erosion has removed
the upper part of Rieden volcano exposing a thick caldera tuff filling
and several differentiated intrusions, some coarse-grained. All 3 phono-
litic caldera volcanoes (Laacher See, Wehr, Rieden) are basically similar
in structure and eruptive processes and rocks to the Jemez mountains in
New Mexico where the pilot hot dry rock project is carried out by the
Los Alamos group. We are now mapping and analyzing the Rieden volcano in
detail, especially the high level intrusions. Preliminary data suggest
that Rieden volcano may be very suitable for a hot dry rock project in
Central Europe and a number of shallow exploratory drill holes are planned
to define the subsurface lateral extent of the coarse-grained intrusive
igneous bodies more precisely.
- 112-
GEOTHERMICS IN HAINAUT (BELGIUM)
-113 -
Even without a difference in topographic level between the
discharge and the recharge areas one can imagine the instability of such
a thermo syphon and the existence of a permanent circulation. In our case
this topographic level difference does exist; it amounts to one hundred
meter and its effect comes in addition to the temperature difference
which influences the density and the viscosity.
R.A. Freeze and P.A. Witherspoon (1966) have analysed a hydro-
dynamic circulation model suitable to our case; adding to this model,
the relationships established by R.W. Stallman (1960) dealing with
heat conduction (from bottom to top) and convection would allow to get
the steady state isotherm positions.
Like all geothermal systems also this one has its own limita-
tions. Intuitively one knows that there exists a treshold production rate
above which temperatures would decrease. The recoverable heat energy
essentially depends on the heat exchanging surface available and the
convection speed. These elements are linked to the size of the system.
In our area, the discharge zone has at least a length of 30 km. One would
like to keep the aquifer in a steady state quantitatively analogous to
the one existing today.
The slight mineralization of the geothermal water does not
require to reinject it and besides the natural recharge keeps up the
hydrostatic pressure of the aquifer.
An increase in the water circulation speed would undoubtedly
increase the dissolution of the water soluble anhydrite in which the
karst has developed and therefore trigger off small pseudoseismic
activities similar to those felt at several times in the past in the
Raine valley.
Finally it should be noticed that the economy in drilling for
the karst at its northern boundary where the aquifer occurs at shallower
depth is probably illusory. Indeed the cementing of the pores and
fissures in the dissolution zone of the limestone is not always complete
and therefore a decrease of the pressure at the northern boundary area
of the karstic limestone could introduce surface waters in the production
well. Experiments underway at the Douvrain drill hole should clarify
that point.
- 114-
Conclusions : All observations made until now suggest a meteoric origin
for the hot waters in the Hainaut province. These waters circulate in an
area where the normal geothermal gradient is influenced by an important
convective heat transfer.
It is sure that the Carboniferous limestone reservoir can
already be used now at St. Ghislain, albeit in a limited way. However
the question is wether the reserVOlr extends in the whole western part
of the Hainaut provi~ce. The solution of this question is the aim of
future research. If the answer turns out to be positive, a new and
original type of geothermal exploitation could develop in that area.
Fig. 1.
a Schema de c I rc ulahon
dans Ie calcaire carbonifere du
III
<11-
.... Cl> 2
::J=
__ Cl>
or.
"-,~
'<11 ~
E '"C
<11 ~
I- -
A B c o
b. Temperatures suivant Ie parcours A B (0 du schema a
1. en I'absence de mouvements de convection
2 avec convection .
Sans source thermale, les deux aires hachurees auralent meme surface .
- 115 -
GEOTHERMAL EVALUATION OF THE TRIASSIC RESERVOIRS
A.N. KOROTCHANSKY
Summary
In the PARIS Basin, the area extension of the Triassic outcrops and
subcrops is about 12,500 square kilometers.
-116 -
1. INTRODUcrION
2. GEOLOGICAL CHARACTERISTICS
On the structural point of view, the PARIS Basin is very calm, for
it is an autogeosyncline (as defined by Mac Kay in 1942). The average
dip does not run above 2°.
The Triassic outcrops and subs crops extend over an area of about
12,500 square kilometers.
3. RESERVOIRS
-117 -
The characteristics of these reservoir-units are given in the following
table :
up to over 400
in Lorraine
Total thickness
(meters)
up to 175 up to 190 ----------------.
up to 100 else-
where
up to over 100
in Lorraine
Thickness of
net sands (meters)
20 to 50 20 to 45 -----------------
up to 20 else-
where
Over 50 in
Transmissibility up to a minimum up to a minimum
Lorraine
(darcys.meters) of 10 of 15
-----------------
Up to 20 else-
where
Salinity
(ppm)
Oto 150,000 9,000 to 300,000 o to 300,000
Temperature
(DC) 30 to 100 30 to 110 25 to 110
-1lS -
Areas of immediate interest
for development of geothermal energy
from Triassic reservoirs
ENGLISH
CHANNEL -~ __
(
/
/
/
/
)
(
)
o 50 100km
1=' = _..c'= ....-"'"
- 119-
of the geothermal water between 70 and 900C depending on the selected
reservoir-unit and location.
-120-
THE EVALUATION OF A PRELIMINARY GEOTHERMAL MODEL FOR THE
PHLEGRAEAN FIELDS VOLCANIC AREA (ITALY)
Summary
The results of volcanological and petrochemical studies, MT
surveys, passive seismic studies and a new gravity and magnetic
interpretation relatively to the Phlegraean Fields area, Italy,
are presented.
The evaluation of the geothermal model of the area and a
preliminary assessment of the local geothermal resources are
presented.
1. INTRODUCTION
This report deals with the results obtained by a multidisci-
plinary research carried out in thePhelgraean Fields geothermal
area near Naples, Italy (Fig. 1). The research was preceded by
the analysis and synthesis of all the available data about the
Neapolitan volcanic area, having the aim to indicate:the zones,
within this area, where there are evidences suggesting the pos-
sible existence of important geothermal anomalies linked to
shallow depth hot magmatic bodies;and the most appropriate stu-
dies to be developed for the assessment of their geothermal po-
tentiality (1, 2).
The conclusion of the former research indicated the conve-
nience of going on in this field both from a methodological
and an applicated point of view. The area which appeared to be
the most favourable one was the Phlegraean Fields and therefore
a number of geophysical and volcanological researches were pla~
ned there, having the aim of reconstructing the geometry of the
main geological structures and the evaluation of the local ther
mal model.
In particular, the research was aimed both at the evaluation
of the origin of the temperature anomaly of this area and at
the definition of the geological and geophysical parameters of
- 121-
lE GE NO
10 2ez:a 3_ 4~
5~ 6W 7tmm sC3
9 - 10- --
,
-
I~
HI.A Q
[?r
PROCI DA '
,-
. ,r- - Gelfed,
o
~
--'-__----'2p Km
CAPRI
D .
- 122-
preliminary assessment is feasible and it will be discussed in
the following together with the main results obtained by the dif
ferent methodologies.
A number of laboratory and field works are still under way
and the conclusion of the research is planned for summer 1980.
-123 -
-Na 2O 10 K 20
r 14 rCaO
8~
.A.
.~.
.y t •
8~ ••• •
• A
6. .1 0 •
.A. • 6
• '"
6~ c.. •
• 6
• •
6 • 6 •
6
• c. .A..
.A.
• • lie
• • 6
4 f- • 6•• J ••
.A. c. ! •
" 4 • 6
• • •
•
.A. c. '" '" C. 6
.A..A.
6
• .......•
2~
c. 6
Do 2
2
t 1\
30 50 70 30 50 70 30 50 70 90
D. 1. D.1. D.I.
• Phleg raean Fiel d s l avas
c. Ro ccamo nfina potas sic series ( G ,~i ara e t al ., 19 79) (10 )
p '.l egr aean Fields xenol it,les ( Gl"liara et al
., 19 79)(1 0 )
'"
Fig. 3. Best fitting model of fractioned crystallization process of Phlegraean Fields magma.
phic studies. The result show that such phases equilibrated at
temperatures of about 950 o C, excluding therefore, also on the
basis of isotopic data, late interaction with low temperature
fluids.
From a chemical point of view, all products result to be
slightly undersaturated as they show generally a small amount
of normative nepheline which tends to increase with the diffe-
rentiation index. Normative nepheline, in fact, reaches in the
alkaline trachytes a content of 19%. Phenocrysts and microcrysts
of leucite are not present in the medium to high evoluted terms.
This indicates,as shown by the Quartz-Nepheline-Potash
feldspar diagram,that the crystallization of the trachytic ma~
rna occurred at a water pressure of 1 kbar at least. The major
elements distribution shows variations along a continuous trend
which is compatible with shallow depth fractionation processes
of the phenocrysts and microcrysts phases detected in the sam-
ples (Fig. 2).
In order to make at least a preliminary control of this hy-
pothesis a fractional crystallization model was computed, using
as a IIstartingfl magma, the less fractionated rock found in the
area (trachybasalt P.F.20). As a result a model of compositio-
nal variation was obtained which indicates that the last term
is represented by the alkaline trachyte. The phases which are
segregated during the different stage and their quantity were
computed by means of a program which minimize the differences
between computed and observed compositions of thefltargetflmagma
(3). The obtained results are reported in the scheme of Fig.3.
For each step the sum of the squares of the residual oxides
~R2) is reported, which is generally very low indicating that
the computed mass balances are significantly good.
Therefore, also these data support the hypothesis that the
products of the Phlegraean Fields are the results of a shallow
depth fractional crystallization process starting from a basic
magma assumed to have a composition similar to the less diffe-
rentiated products (trachybasalts).
Following this model, trachytic products, which constitute
most of the outcropping volcanics, represent about 20% to 40%
of the original magma volume.
3. GEOPHYSICAL STUDIES
a) Magneto-telluric Soundings
Following a former campaign made in cooperation with the
Institute of Geophysics and Meteorology of the Braunschweig
University, RFD, (4), 14 MT soundings were carried out. The
location of the soundings, reported in Fig. 4, was cheosen so
to complete the former survey. 14 DC soundings were also carr-
ied out in the same locations. The MT instrumentation consists
- 126-
7
CAMPI FLEGREI
7
. ". "
... .. . . .
GOLFO
01 POZZUOLI
o M T soundings
o carried out by Hunsche et a I 1980
o,
I Km .
I
I P~EAMi1 1 OF 5- -I ir:2 - - -
-j 3 - - - - T 4- - - - -tj- - -
I [> I : ~!HI PASS 11-1 FILTER II r SHORT I', r AM P A DC
, , -
.... i~ __DATA
t • . TO
7 • B •
PREAMP. B RECORDER
, I
. FILTER CONTROL r--r--
I CONTROL : 1 CONTROL
r-p~EA-MP.- GAIN I ICLOCK SAMPLE INTERVAL FILTER BAND
i CO~OL CHARS I l...:IC:...::O::...:..N.:....:T::...:..R=O=-=L~(-=-H:.;...A=R'-"'S:.;...._1-1_ _ _ _ _ _ _ _ _----1
I , - 8---,---- +~~ DATA BUS~_ _ _---<-....J
~ ~
5I 3I L'_ _ _
_ _ _ _8 ____
1- DA T A-S-E:T~C·T--& CONTROC-SOS
I
9! -~
'ID MT. SISTEM:
A DC i CONTROL BLOCK DIAGRAM
CONTROL Ie HARS. Fig. 5
~([Jm) resistivity thickness • • ••
17 Ohm.m 80 m • 0• 0 0
10 2 • MT 9
1.3 100
•0 0
300 2100
•0 (,
3000
250
1800
1000000
•
•
0
0
•
•0
0
10 1
0
•
•
• •
• 00 •
• • ••
0
10
10- 1
10 0 10 1 T (Sec)
~(Qm)
3
10
resistivity thickness 0
0 Ohm,m 40 0 m •
40 350
100
... MT10
4
800 2000 •
4000 1000000
•
•
0
2
10 •
•• o
•
....
•
I •
10 ~------~-----------------r-----------------.--------
S,L
2.B
'2
BOO { \'\j~
;;}J - - --- ~ -
-:--. -
-- _. _. -- .-: 3000 ::' 250 5000
3 3000
5000 :
4000
K
... ---- -...
"'"' . ~
-; ~,
~ .--
• l.oeo l ne'work
• O n 'lIe ne,wOtk
•
6 Well
<00
SG
•
MASSICO
o 20km
I !
• Seismic stations
baving its top at a depth varing from few hundred meters in the
centre of the area to about 1 km. This demagnetization effect
within the volcanic layers was ascribed to thermal alteration.
In fact, a caolinization process may cause a strong reduction
of the volume percentage of ferrimagnetic elements with forma-
tion of pyrite. Moreover, temperature higher than 150°C may
cause a remarkable decrease of thermoremanent magnetization if
the magnetite present in the rocks of Phlegraean Fields has a
high ulvospinel content. (5). Moreover, the nature of the Phle-
graean Fields basement can not be even inferred from magnetic
data as, if we take into account the thermal and chemical dema-
gnetization effect due to the very high temperatures at depth,
a lavic basement acts, like a carbonatic one, as a non-magnetic
layer. Anyway, from a geothermal point of view this basement,
whatever its nature be, may represent a main target for the
search of a high permeable reservoir.
A comprehensive schematic model of the Phlegraean Fields
structural pattern, as deduced by the interpretation of gravity
and magnetic data, is shown in Fig. 11 (5).
- 133-
/
/
_/
~
~
~etfo
.' ~ /
/ ._<V
\ ...
...
'. I
I
" I
w
.... / '. .
I I. 01 NISIOA '.
......
OEL ~~~~lko::; f}
0·· · ;
\ \
,
\
/ ~500. ---r \
\
/
\
.I Fig. 10. Basement bathy~etry in the
I
I Phlegraean Fields as deduced by sei
I l. MISENO srnic reflection survey (on sea) and
/ gravity data (on land) (aft'e r Nunzia
ta, C. and Rapolla, A., 1980)(5). -
/
I
.....
8ACOLI
Km
sw ~ POZZUOLI
o ............... . . .
3 (! -= 2.7 (g 5 -..--.--~-
/";:/
~/-..,.. ....,....-("!TJ-=-2-.-5-C.. ,~:-;-~-.,-J....,
~--:
20~ U
Fig. 11. Comprehensive schematic model of the Phlegraean
Fields structural pattern as deduced by gravity and magnetic
data (5).
CONCLUSIONS
The model which at this stage of the research can be drawn
for the Phlegraean area consists of:
a) A main heat source, which is represented by a magmatic stock
occurring in the upper crust where normal faults and block t i l t
ing have created traps for the magma. This magma has differen-
tiated along the main trend: trachybasalt-Iatite-trachyte-alk~
line trachyte. These last terms represent about 20%-40% of the
parent magma. From a rough estimate of the outpoured trachyte
products the total magma volume can be estimated to be about
200-300 km 3 • The depth of this body can be evaluated on the b~
sis of the petrochemical and geophysical data. The former ind~
cate, on the basis of the trachyte paragenesys and of the pre-
sence in the undersaturated trachytes of alkaline feldspars
only, a depth higher than 3 km (PH20~ 1 kbar); the latter show,
on one side, that the depth of such a magma body is greater
than approximately 5-7 km (absence of compatible gravity anom~
ly) and, on the other, the presence at shallow depth of magma-
tic masses of limited extension which could represent intrusions
along feeding channels (ground deformation, conductive anoma-
lies (?»).
b) A main potential geothermal reservoir, which is represented
by a rigid sequence laying at a depth of about 2.5-3 km and up-
rising in the Miseno-Averno area at about 2.2 km. The nature of
this basement (carbonatic or lavic) could not be inferred nei-
ther from gravity, nor from magnetic, nor from seismic data.
Anyway, the general setting of the whole Campanian area favours
the hypothesis of a basement of carbonatic nature.
c) A cover, which is represented by a thick volcanic sediment~
ry sequence. Lighter sediments are present in the central part
- 135-
of the area. Moreover, this cover is locally fractured and high
permeability levels are also present. Both cases can represent
secondary but interesting productive targets. Most of the vol-
canic sedimentary sequence shows a remarkable demagnetization
most probably due to thermal alteration effects.
On the basis of this model a preliminary assessment of the
geothermal resources of the Phlegraean Fields and of the other
zones of the Neapolitan volcanic area was made, using the volu
me method as proposed by Muffler and Cataldi (1977)(8). The d;
tails of this assessment are discussed in a former work (9). -
The whole Neapolitan area was schematically divided into a
number of zones: Roccamonfina, Volturno plain, Qualiano-Parete,
Phbegraean Fields (subdivided into 3 subzones), Naples-Caserta,
Vesuvius and Ischia (Fig. 1). For each zone a geological and
thermal model was assumed on the basis of the available geolo-
gical and thermal data.
In particular the three subzones of the PhlegTaean Fields
are:
a) The Miseno peninsula and the Mofete area (34 km 2 )
b) Solfatara, Agnano and the bay of Pozzuoli (62 km2 )
c) the remaining part of the Phlegraean Fields (71 km 2 )
The schematic geological model utilized for the three zones,
which is mainly based on gravity and seismic reflection data,
and the thermal model based on available temperature data, are
reported in Table I together with the figures for the most re-
levant parameters appearing in the computation. The obtained
result are reported in Table II and Fig. 12. An evaluation of
of the reliability of this result is very difficult. In fact,
the figures that have been assumed for the different parameters
entering the computation are often obtained as an extrapolation
of shallow depth measurements or are deduced by comparison with
other well known geothermal areas. Anyway, we attempted such
an evaluation for the zone a) of the Phlegraean Fields and only
for the Base Resource assessment with the assumption that the
different parameters are independent each other. It results
that the obtained final figures are affected by an error of ±50%
(9). This figure demonstrates how approximate is this assessment
at the present stage of the knowledge. Anyway, the obtained re-
sults confirm the remarkable importance of the geothermal ener-
gy in the Neapolitan volcanic areas. It should be stressed that
the noticeable amount of geothermal energy available in this
area for electrical production seems to be mainly concentrated
in the Phlegraean Fields areas and that a remarkable amount is
also available for non-electric uses.
-136 -
Fig. 12. Neapolitan volcanic area. Geothermal energy reserve:
total and for electrical energy production (underlined values).
Values expressed in GWyt •
-\37-
Table I
Parameters utilized for the geothermal assessment of the Phle-
graean Fields zones (Rapolla et al., 1979)(9).
Temperature
Zone (area) Depth
Gradient
Cover 0 to 2.3 Ion 125°C/Ion
~ (34 Ion 2 ) Reservoir 2.3 to 2.8 Ion no grad.
Basement 2.8 to 5.0 Ion 50 0 C/Ion
Water
Rock Density,
Total Effective Rock Specific Specific
Porosity Porosity Density Heat Heat
Cover 0.30 0.06 2.2 gr/cm3 0.20 cal ~ V
Reservoir 0.05 0.02 2.65 0.22, /0 C
§ u
f
Basement 0.025 0.002 2.65 0.22 a e
s
- 138-
Table II
- 140-
INTEGRATED GEOTHERMAL RECONNAISSANCE
OF THE SOMMA-VESUVIUS SYSTEM
ABSTRACT
The main results of the first phase of the geothermal exploration of Somma-
Vesuvius active volcanic area (including geologic, volcanological, hydrogeologi-
cal, hydrogeochemical, gravimetric, aeromagnetic, geoelectrical and reflection
seismic surveys) are summarized and used for the elaboration of a preliminary ge£
thermal model. The heat source should be provided by the Ve"suvius magma chamber
that should occur at shallow depth (3-5 km) according to petrologic data. The main
reservoir is represented by Mesozoic calcareous formations whose top depth ranges
in the Vesuvius area, from 500 to more than 1500 m. The impervious cap rock is
found in Tertiary flysch or Plio-Quaternary sedimentary sequence overlying the r~
servoir rocks. The main problem of the geothermal exploration of the area, that is
the intensity and lateral extent of the thermal anomaly will be checked by drilling
in the next phase planned for 1980.
-141-
1. INTRODUCTION
The first interest of AGIP in the geothermal exploration of the Somma-Ves~
vius volcano dates back to 1952, when preliminary geological, volcanological and
geochemical investigation were carried out not only in Campania, but also in many
other italian areas, including Vulcano I sland and Sciacca hydrotermal basin in
Sicily, where preliminary exploratory drilling gave very interesting results.
The present geothermal project initiated in 1978 and AG I P is carrying out
exploration activity in the area as operator of the AGIP-ENEL joint venture rela-
ted to the "Ottaviano" 462 square kilometer permit. The aim of this project is to
test if and under which conditions high enthalpy fluids can be commercially produ-
ced from reservoirs related to recent or active central volcanoes of the Somma-Ve
suvius type.
This paper summarizes the results of the first exploration phase that is ai-
med at defining the geothermal model of Somma-Vesuvius by detailed geological,
geochemical and geophysical investigation. The target of the second phase will be
to test the model by drilling. The project is therefore focused on exploring an acti-
ve volcanic area for natural hydrothermal system that could be exploited with pre-
sent day geothermal technology.
The methodology used for the evaluation of the geothermal interest of the
Somma- Vesuvius system could be applied to other active or recent volcanoes not
only in I taly but also in other countries of the Community.
The program for the first phase has been planned in order to provide the fo.!.
lowing information, needed to define the geothermal model of the study area prior
to drilling:
- detailed tectonic and volcano-tectonic structure;
- detailed reconstruction of the magmatic features;
- detailed geometry of the top of the calcareous reservoir;
- identification of the cap-rock;
- reconstruction of the main hydrogeological features;
- identification of geochemical anomalies related to fluid leakage from deep reser-
voir.
-142-
2. GEOLOGICAL SETTING OF THE CAMPANIAN VOLCANIC AREA
The geological literature on Campania is very abundant this short synthe-
sis therefore refers only to the volcano and sorrounding sedimentary area.
The Campanian plain is covered by a thick sequence of quaternary sedime!!
tary and volcanic deposits. The latter ones are related to the activity of Phlegrean
Fields and Somma-Vesuvius potassic volcanoes developed on tre Tyrrhenian mar-
gin of the Campanian Apennines.
The structure of that orogenic belt consists of a pile of nappes emplaced
during Miocene, with overthrusting towards East. The various tectonic units
are formed by mesozoic carbonate rocks, limestones and dolomite, with a 4000
meters total thickness, and by flysch deposits consisting of shales, sandstones
and marly limestones of Oligocene. The carbonate complex is usually deeply frac-
tured and hence highly permeable; these rocks therefore represent the main target
(potential geothermal reservoir) of the geothermal exploration of the area.
Mio-Pliocene and Quaternary clastic sediments extend in all Campanian
plain and may reach relevant thicknesses in structural lows. The Tertiary and
Quaternary deposits include many impervious rocks that may be assumed as good
cover to the underlying reservoirs.
The region is affected by an intensive vertical tectonics of Pliocene-Qu~
-143 -
The activity of the Somma-Vesuvius composite central volcano was largely
contemporaneolls to that of the Phlegrean Fields. The oldest products of Proto-
Somma might correspond to those of the first Phlegrean period (rv 35.000 y). The
last Vesuvius eruption (1944) closed a cycle of quasi-persitent activity (1631-1944),
therefore the volcano has to be considered still active and possibly connected with
deep hot magmatic masses.
3.1 ~~~£gL'!!l.2-_V_0l.'2..a..!!£I~gy_
-144 -
the occurence of two main types of faults or eruptive fissures. One has a local
volcano-tectonic significance, and is characterized by short fractures with a mo-
stly radial trend, that probably affect only the "volcanic. cover.
The second type has a regional significance and faults display a much higher
vertical and horizontal extent, certainly prolongating across the basement of the
volcano. The 'feeding vent of Somma-Vesuvius is located at the intersection of two
fracturesof this kind, with a respectively apenninic (NW-SE) and antiappenninic
trend (NE-SW).
Somma-Vesuvius products have a mostly basic composition ranging from leu-
citities to tephritic leucitites and tephrites. The most evolved products are trachy-
tes found only as ejected blocks and old tephra, and phonolites or tephritic phono-
lites occurring in Plinian pumice. The basic lavas erupted in the 1631-1944 period
are richer in K20 and more silica undersaturated than most of the previous lavas.
The volcanic and magmatic history of Somma-Vesuvius indicates that the
last period of activity (1631-1944) was characterized by frequent uprising of deep-
originated basic magmas with short residence, time at upper crustal levels. The
quasi-persistent character of the activity in the last three centuries furtherly
suggests that the volcanic vent was almost continuo sly filled by hot magma. The vo-
lume of the magmatic column was however probably too low, being of the order of
0.1. km 3 (total volume of the products emitted by the last 1944 eruption), to have
produced a strong thermal anomaly in the country rocks. On the other hand, volca-
nological and petrological data suggest that the salic products erupted in the seve-
ral Plinian eruptive phases of Vesuvius could be related to fractionation within a
shallow magma chamber located within the calcareous basement. The last two of
these eruptions (79 A. D. e 3500 b. P .) were therefore studied carefully in order
to get informations on the size, depth and temperature of the magma chamber under-
neath the volcano. Results show that interaction with country rocks had limited im-
portance and all data indicate that pumice were produced by cry stat-liquid fraction~
tion within the undersaturated part of the petrogeny's residua system at about 1 kb
water pressure. The solid-liquid trend indicates that the two pomice originated by
lfractionation of slightly but significantly different parental liquids. Some lavas of
appropriate composition were selected as parental liquids to compute the entity of
- 145-
the fractionation. In both cases a fractionation of about 70 weight % solid was nee-
ded to produce liquids with the composition of the pumice.
The combination of all data indicates that the two Pli.nian eruption were fed
by a magma chamber (3-4 km deep) having a volume of approx. 2.0-2.5 cubic km.
The temperature of the magma that initially entered the chamber was about
1l00°C, whereas the temperature of the residual liquids erupted as Plinian pumice
was BoOo and 850°C respectively. There is no evidence that such a magma chamber
existed at Vesuvius after the 79 A. D. eruption.
Physical models aimed at describing the possible extent of the thermal anoma
ly induced by these shallow magma bodies in the surrounding calcareous rocks are
presently under study.
3.2~l~~&~l~13E~j~~CE~~~~1_
Hydrogeological and hydrogeochemical surveys (fig. 2) have shown the ex i-
stence, in the Vesuvius area, of a regional water circulation with a main NE-SW
flow. Water infiltrates through the calcereous reliefs outcropping to the NE of Ve-
suvius, flows down the volcano and rises up in the coastal zone, chemically and
thermally modified. This water flux is of the order of 6-7 m3/sec and affects
either a single or more interconnected aquifers. Local water recharge occurs in
the volcanic area, where perched aquifers are present, with limited lateral extent.
The regional flux is characterized by progressive chemical modification of
the original alkaline-calcium-magnesium bicarbonate water due to calcium-magn~
-146 -
possibly related to the uprise of deep hot waters, and where hydrogen traces were
detected in the past.
The ammonia anomalies are more dispersed and are possibly related to deep
NW and NE fractures. Their weak values however indicate low thermality at depth.
As a whole, there are only few thermal manifestations on the Vesuvius area, mostly
of thermal water with temperature ranging from 14°C to 36°C. Fumaroles with T of 200-
300·C occur only in the Vesuvius summit crater and their composition indicates an
origin from meteoric water degassed and heated by convective circulation within the
upper part of the Volcano.
Only few water samples are suitable for geochemical geothermometry. The
K/Na-Ca, and silica geothermometers gave however constrasting results.
3.3 ~~~E~~£~
Gravimetric, magnetometric and geoelectrical surveys in the Ottaviano per-
mit were carried out in 1979, together with some seismic profiles.
The treatment and intrapretation of the results is not completed yet.
In the gravimetric survey 850 new stations were measured and preexisting
data. from other 300 stations and land and 20 on sea were considered.
The Bouguer anomaly map is reported in fig.4 together with a preliminary
interpretations. First results indicate a gravimetric gradient probably related to
the deepening of the mesozoic limestone basement from the outcropping relief tow-
ards the sea and from the Sorrento peninsula towards N. The E-W deepening seems
to follow a rather regular monoc1inal structure, whereas from S to N it appears
affected by two normal faults. The few data on sea suggest a dip towards Nand W
of the limestone basement. It appears affected by a fault running along the coast
with an important landward throw.
The aeromagnetic survey (fig. 5) consisted of 2.220 km of lines flown at an
altitudine of 1. 450 mas.!. It shows a wide positive anomaly (up to 1800 n centered
on the Vesuvius Volcano. A first interpretation of this anomaly using the surface
samples suceptibility values indicated very large and thick magnetic bodies diffi-
cult to reconcile with geological evidence. I t was therefore decided to consider the
influence of remanent magnetism and of the Curie point variations on the magnetic
- 147-
properties of vesuvian rocks. This study is still under progress, but first results
show that both factors cannot be neglected for a correct interpretation of magnetic
data. The two SE and SW positive noses in the Vesuvius subcircular magnetic ano-
maly could correspond to thicknening of the volcanic pile or to zones deeply injec-
ted by lava dykes.
125 vertical electrical Schlumberger soundings with AB from 1000 to 6000
m were also carried out in this area (fig. 3). The difficulty of access has preven-
ted the realization of a more complete program. Evident resistivity anomalies have not
been revealed by the survey. The isobath map of the resisti ve basement (fig. 3) con-
firms the general trend toward Wand N of the probable carbonate basement as
shown by the gravimetry.
Four reflection seismic profiles, for a total of about 40 km, were carried
out using an explosive energy source and a six fold coverage. They were shot on
the plain located between Vesuvius and the calcareous relief to Est of the Volcano.
The directon of NA 311-79 line is approximatively N-S, of NA 310-79 line
is about W-E, of NA 308-79 and NA 309-79 are SW-NE.
The results are of fair quality: they show a seismic horizon corresponding
probably to the top of the mesozoic carbonatic basement ad overlying a series of
reflections probably corresponding to tertiary sediments, including probable vol-
canic products.
The top of the carbonatic basement is lowered by block faulting towards
S-SW; a further sink probably occurs in the area comprised between Vesuvius and
the Tyrrhenian coast. Its depth ranges from approx. 500 m near the calcareous
relief to 1500 or more in the coastal area.
4. GEOTHERMALPROSPECTIVES
The results of geolopical, volcanological, hydrogeochemical and geophysi-
cal exploration carried out in the Somma-Vesuvius area allow to draw a prelimina-
ry geothermal model to be tested by drilling.
In the model the heat source should be provided by the magma chamber of
Vesuvius, approximately located at 3-5 km depth. Its size could however be too
small to have produced a large thermal anomaly and no clear evidence has been so-
-148 -
\. \ . , , .... - -.--
" .\,1\."'- .... .,,'
-, Somma
-\ • V'SuVI.na
A'
Vesuvlo Somma IE
2~ s.",
4~~;'1
I
=I~-.-.-.-·- --_.
BOUGUER ANOMALY
--
VESUVIUS VOLCANO
-149 -
POltllg l. ano d Areo hOPHt511C, hne
•
'0 Wat ". I l ow (,end
HYDROGEOCHEMICAL SKETCH
Fig.2
RESISTIVE BASEMENT
DEPTH MAP
__ "0 __ D epth,
0;-;- Ve $',.... 0
<II.- "
/<1
'" 1-0
-<>", ,500
'".'" ., '91,.
/;-
., .,
" ..
oS'
,,~
..J
F.g.3
-150 -
BOUGUER ANOMALY
MAP
- + 70 -
Isoaoomaly contours
Probable
fault $
...
F.g 4
T OTAL MAGNETIC
INTENSIT Y MAP
, '"
Fig .5
- 151 -
far found for the occurrence of high enthalpy geothermal fluids at the perifery of
the volcanic area.
The hydrogeological conditions are instead very favourable for a geother-
mal field. Tha main reservoir is represented by mesozoic limestone formations (top
depth ranging from 500 to more than 1500 m) with secundary permeability by fractu-
ring related to the intensive recent tectonics.
The overlying Tertiary flysch and Plio-Quaternary clastic sediments
should act as impervious cover on the deep reservoir. I n these conditions, the
main target of further exploration appears to be the control of the intensity and
extent of the thermal anomaly. The occurrence of a thick surficial sequence of
permeable rocks descourages the use of temperature gradient boreholes.
I n the second phase of the exploration a 2000 m deep exploratory well is
planned. Its most favourable location appears to be the southern side of the Volca-
no at mid way to the coast, where the thermal indications are more encouraging.
REFERENCES
BARBERI F., INNOCENTI F., LUONGO G., NUNZIATA C., RAPOLLA A., 1976
Shallow magmatic reservoirs as heat source of geothermal systems: prelim i-
nary interpretation of data available for the neapolitan active volcanic areas.
Seminar on Geothermal Energy 1977 - EUR 5920 Vol. I.
CAMELl G.M., PUXEDDU M., RENDINA M., ROSSI A., SQUARCI P., TAFFI L.,
1975 - Geothermal research in western Campania (Southern I taIy): geological
and geophysical research. 2 nd U. N. Symposium on the development and use
of geothermal resources, San Francisco.
- 152 -
CIVlTA M., DE MEDICI G.B., DE RISO R., NICOTERA P., NOTA D'ELOGIO
E. 1973 - Memoria descritti va della carta idrogeologica della Campania Nord-O~
D' ARGENIO B., PESCATORE T., SCANDONE P., 1973 - Schema geologico del-
l' Appennino meridionale (Campania e Lucania); Acc. Naz. Lincei, Quaderno
nO 183, 49-72.
FINETTI I., MORELLI C., 1974 - Esplorazione sismica a riflessione dei Golfi
di Napoli e Pozzuoli; Boll. Geof. Teor. Appl. XVI, 175-222.
- 153 -
PROSPECTION GEOTHERMIQUE INTEGREE DANS LE MASSIF DU MONT DORE
par
Z Z Z zz
J. VARET , L. STIELTJES , A. GERARD, C. FOUILLAC
RESUME
Un important programme de prospection geothermique a ete mene
dans le Massif du Mont Dore en 1979 soue la conduite du Departement
Geothermie du B.R.G.M., aVec le soutien financier de la CCE, et la colla-
boration de nombreuses equipes (BEICIP, SNEA, CNRS, Universites, CEA).
Diverses methodes geologiques, geophysiques et geochimiques ont ete con-
centrees sur l'objectif, d'une dizaine de kilometres de diametre environ,
que constitue la caldera du Mont Dore.
On a pu degager ou confirmer les principau:x: elements de l 'his-
toire vUlcanologique du Massif du Mont Dore qui s'est elabore a partir
de 4 M.A. avant J.C. en deu:x: phases principales.
~ B.R.G.M. - Orleans
~ C.N.R.S.
- 154-
Cette synthese a ete realisee a partir des travaux de :
- lSS-
INTRODUCTION
- 156-
PROSPECTION GEOTHERMIQUE DANS LE MASSIF DU MONT DORE
I - GEOLOGI E
I. I - INTRODUCTION
Bien que des etudes approfondies aient deja ete menees a bien sur
Ie Mont Dore depuis plusieurs dizaines d'annees, leur caractere marque de
recherche scientifique laissait un certain nombre d'inconnues quant aux
donnees a caractere applique pour mener a bien une prospection de champ
geothermique debouchant sur des operations de forage profond. Ainsi, les
notions de localisation, d'evolution dans Ie temps de l'activite liee a
la source de chaleur profonde se devaient d'etre precisees. De meme, les
aspects de structure interne du massif volcanique, a savoir sa caldera,
la nature et l'epaisseur de son remplissage, la localisation des zones
d'injection preferentielle du magma representaient une aproche fondamen-
tale. D'autre part, la preparation d'une operation de forage profond ne-
cessitait de disposer de coupes geologiques previsionnelles redigees dans
l'esprit d'un montage technique d'une phase de forages. Enfin, une per-
ception synthetique globale actualisee de la geologie de ce volcan central
etait necessaire pour l'interpretation des resultats de la grande masse
des etudes geophysiques entreprises dans Ie cadre de ce contrat.
- 157-
Le B.R.G.M. est intervenu pour les etudes de structures geologi-
ques internes du massif (tectonique, volcanologie, problemes de constitu-
tion et de remplissage de la caldera), de geochronologie (en liaison avec
l'Universite de Clermont-Ferrand), et a ete charge par la CCE d'assurer
la coordination des travaux et realiser la synthese geologique et globale .
- 158-
L'altitude moyenne du massif est comprise entre 1 000 et 1 500 m,
les villes thermales de la Bourboule et du Mont Dore, situees au coeur de
la fosse centrale du massif ou coule la Dordogne sont a une altitude de
900 a 1 000 m. Le cours de cette riviere, qui prend sa source au Sancy,
change 3 fois de direction dans sa traversee de la fosse, selon des di-
rections majeures de fracturation du socle. Les cr~tes volcaniques domi-
nent la fosse centrale de 500 a 700 m sur les rebords nord et le rebord
est, et culminent au sud au Puy de Sancy (1 886 m) avant de se reabaisser
graduellement vers l'ouest OU la fosse s'egueule et permet a la Dordogne
de s'echapper (fig. 2).
S k.c=
I
1191'0 (le paruqe du e.AUX
-160-
dificatJo n du
Si gnal du Luguet
CEZALLIER
w-
~
en _____________________________ Prenli~ I 'e s intru-_________________________
z-
« sions .' et p
o...J de /" Mont Oore Sanc y
o
>
_ _ _ _ MONT DORE • • • • II 1- __ _ •
::;
-162-
/
<I .. 0 'X'
·2 .
i
~.. ~
~ ~
~
~
..· ~.
u
=
·j
!:l
~ c
0
i! N
E ~O
j
-, :\
~::;
~ ------
'°1 rrnOl1
/
'"'--.--- --
.. -
Clo
/ 'lis
--~--
··163 -
1.5. - LA CALDERA CENTRALE ET SON REMPLISSAGE
Son existence a ete suggeree par les etudes geologiques, qui ont
demontre l'absence du socle a l'affleurement sous la couverture volca-
nique de la partie centrale, puis par des etudes electriques CGG (1960).
Des mesures-de gravimetrie (1962) ont permis d'en ebaucher les limites
qui isolent une caldera d'environ 60 km2.
-164-
fig. 6 - Schema volcanologique du strato-volcan du i.jont ilore
- 165-
c rd. a.~
1 61 Cou16u ,,'centAu du Sancy
I. . ~c "!~:~i
py:'oclasc.it.t. pon.cc!"Ises Uilles a'\,lX itLt.:rusions Itt. c!O:e
~hyol~t.lques 01,.1 tr4ehyand~5it.lques.
/
~~·~.?;:1:" .J l~t.r\I:Sicr..s phonolit-iqu" 01.1 t'~h.t1t-1ques 0.9S ."'!AI
•/ 91
Stria t..raehYL"\l!hitique post.ealdera (2,1 & 1,50 ,)loA)
il
//
py::oe16sutes I
fOr;:laUons .,on.ctua ... rhyoUI:.1quU pr' at. syncalr:!era
(2.S .& 2,2 .to'I.A,
1/2.5-2.2""
Le reservoir magutlque
se limite aux failles s
poriphidqueo d. 14
caldera. Le upa. est
plus SIC et so tN-cdonne
ju.squf.l des phonolites,
+
aitel en place en fin de
C:). ::.1e. 11l long <10 La +
peripheri'i de 14 c41dlrl.
-168 -
En fait, les donnees plus recentes (MERVOYER, MAURY et VARET,
MAURY) ont montre que l'ensemble des produits volcaniques du Mont Dore
pouvait provenir d'une m~me source de magma basaltique alcalin unique
par cristallisation fractionnee sous differentes conditions de pression
partielle d'eau. Ainsi, un fractionnement sous forte pression d'eau (avec
separation d'amphibole en particulier) produit des magmas satures (rhyo-
litiques), alors qu'un fractionnement a sec produit des liquides res i-
duels sous-satures (phonolitiques).
-169-
1.8 - CONCLUSIONS GEOTHERMIQUES SUR LA BASE DES DONNEES GEOLOGIQUES
-170-
2. - GEOCHIMIE
F Cl- 504
par potientiometrie
Rb+ cs+ Al Ba
1 8.5 5.38 1.600 1. 257 1. 456 0.674 0.243 0.001 0.096 7.060
2 7.2 5.28 1. 267 1.213 0.921 0.970 0.325 0.001 0.082 5.468
3 10.5 6.16 1. 700 2.874 3.353 5.870 1. 309 0.104 0.203 19.218
4 10.3 5.02 1. 600 0.524 0.210 0.657 0.212 0.094 0.102 2.523
5 11.5 5.08 1. 550 0.519 0.202 0.678 0.207 0.073 0.065 2.623
6 10.0 5.30 1. 267 0.699 0.926 0.965 0.212 0.031 0.068 3.815
7 7.0 5.80 1.100 2.016 1. 802 0.600 0.212 0.115 0.102 8.461
8 15.0 5.86 1.467 2.390 1. 234 66.957 1. 790 0.916 58.082 14.788
9 30.4 6.32 2.067 2.725 2.081 64.696 2.271 0.958 45.049 27.570
10 39.0 6.52 1. 733 3.678 1. 304 113.913 3.836 2.249 85.614 33.687
11 8.5 5.86 1. 700 0.783 0.831 0.513 0.115 0.281 0.056 3.707
12 18.0 5.87 1.683 2.096 1. 296 37.217 2.077 0.999 25.642 17.026
13 12 .5 5.64 1.750 1. 262 0.967 35.739 1. 749 1. 083 16.220 22.056
14 40.5 6.18 3.000 2.071 1.473 14.870 1.138 0.375 6.121 15.252
15 44.2 6.18 3.033 2.056 1. 440 14.696 1.105 0.375 5.698 14.989
16 51.5 6.48 1. 917 0.873 0.362 76.174 2.476 1. 333 48.575 27.295
17 18.0 6.36 1.267 0.155 0.177 4.761 0.317 0.250 2.652 3.114
18 12.2 6.00 1.667 2.535 5.471 7.913 1.161 0.001 0.209 24.263
19 11.0 7.20 1. 367 1.038 2.065 7.961 1.141 0.052 0.409 16.452
20 23.5 5.62 1.967 3.553 1. 423 10.478 0.818 2.290 0.296 16.074
21 16.0 5.36 1. 817 2.490 1. 308 5.370 0.627 1.124 0.243 11. 58 3
22 22.5 5.74 2.033 4.137 1.650 12.391 0.928 2.811 0.296 18.356
23 5.5 6.34 0.425 0.135 0.082 0.213 0.051 0.042 0.062 4.474
24 0.0 0.00 1.000 0.793 0.518 0.865 0.146 2.394 0.846 0.000
* .::.
"
:t :t:t
"'* *"
~ :t:t :t
1 0.0 0.00 1.22 0.00 0.92 0.00 0.16 O. ~O 0.86 0.03 0.16
2 0.00 0.00 4.41 0.01 0.00 1. 30 0.44 0.00 0.10 0.05 0.44
3 0.04 0.02 1. 58 0.12 5.00 60.80 0.00 0.01 0.23 0.01 0.05
4 0.00 0.02 24.60 0.00 0.25 2.50 0.85 0.00 0.37 0.03 0.05
5 0.00 0.02 26.40 0.00 0.26 3.00 0.79 0.00 0.32 0.03 0.14
6 0.00 0.01 2.56 0.01 0.00 1.00 0.59 0.00 0.49 0.01 1. 36
7 0.00 0.00 0.57 0.02 0.29 1.00 1. 89 0.00 0.17 0.03 0.06
8 0.92 0.01 0.09 0.89 8.17 49.00 0.13 0.06 1. 59 0.07 20.90
9 0.83 0.04 0.01 0.79 7.10 68.00 1. 37 0.04 0.03 0.00 30.60
10 1.29 0.0 0 0.00 1.80 22.60 1640.00 2.21 0.09 0.01 0.0] 134.00
11 O.OC 0.00 5.04 0.00 0.25 0.00 0.86 0.00 0.27 0.01 0.00
12 0.37 0.04 1.14 0.36 3.94 8.00 1.92 0.03 0.11 0.02 28.10
13 0.41 0.03 0.23 0.40 3.42 17.00 1.10 0.01 0.04 0.00 34.80
14 0.18 0.03 0.20 0.17 0.00 62.40 2.36 0.01 0.05 0.01 5.10
15 0.18 0.03 0.24 0.17 3.00 64.20 2.51 0.01 0.08 0.01 9.00
16 0.92 0.27 0.19 0.84 15.75 1010.00 4.95 0.04 0.01 O.OG 74.30
17 0.07 0.05 0.08 0.06 0.92 9.40 0.17 0.00 0.08 0.02 6.60
18 O.OC 0.02 2.07 0.18 2.20 14.70 13.00 0.01 1. 58 0.01 0.30
19 0.00 0.03 2.08 0.13 2.31 3.00 0.00 0.00 0.67 0.00 0.ge
20 0.00 0.01 0.15 0.05 1. 49 40.00 0.80 0.04 0.17 0.02 2.80
21 0.00 0.01 2.40 0.02 1.10 23.80 0.92 0.02 0.35 0.04 0.84
22 0.00 0.01 0.10 0.05 1.71 40.00 0.56 0.05 0.07 0.01 2.18
23 0.00 0.01 113.00 0.00 0.00 40.00 0.00 0.00 0.09 0.01 0.00
24 0.00 0.03 0.00 0.01 0.26 16.00 0.00 0.00 0.00 0.00 2.87
-172 -
Tableau 3 : gaz dissous en mole/litre
- 173-
TABLEAU 5 TABLEAU 6
.. J_,. .... ~I
o .......... , II- u.~/_ • wIIwtI_ . . ~IArJ
Fig. 10
-174 -
2.2. INTERPRETATIONS GENERALES
-175-
2.3. - INTERPRETATIONS PAR GROUPES
- MDS1 Paiadiroux
- MDS2 Merdant;on
- MDS4 Fontpique
- MDS7 Le Vernet
- MOSS Goyon
- MDS15 Le Bouteix
- MDS26 Collange
- MDS27 Reboisson
-176 -
Les resultats sont donnes dans le tableau 6, on voit que l'on a
une assez bonne coicidence pour le Na/K et le Na/Li, et le quartz.
pendant cette periode, la chimie est restee constante. Ceci peut signi-
fier l'apparition des eaux tritiees et peut conduire a admettre un temps
de circulation tras bref pour Ie Mont Dore ('" 25 ans), sachant que la
teneur de l'atmosphere en tritium a brutalement augmente entre 1952 et
1956 a la suite d'explosions nucleaires en tres haute atmosphere.
-177 -
2.3.4. LA VALLEE DE CHAUDEFOUR
Les sources repertoriees MOSS - MOS6 - MOS30 - MDS28 - MOS29 -
ont ete denommees respectivement Chaudefour 1, 2, 3, 4, 5. Elles presen-
tent des particularites tres interessantes.
Ces deux proprietes nous font penser que l'on est ici en presence
d'eaux du type sulfate acide frequentes dans les champs geothermiques ac-
tifs. Il reste alors a expliquer les relations obtenues entre les compo-
sitions chimiques des diverses sources du groupe.
-178 -
2.3.5. SOURCE SULFUREUSE DU SANCY
CONCLUSIONS HYDROGEOTHERMIQUES
- flux de cO 2
- nappes chlorurees
3. - GEOPHYSIQUE
- 179-
3.1.1. ETUDES STRATEGIQUES
Il s'agissait, pour l'essentiel de prouver l'existence ou l'ab-
sence d'une eventuelle structure de type caldera dans la region du Mont
Dore puis dans le cas favorable, d'en cartographier les limites horizon-
tales, et de situer la structure dans son contexte geologique profond.
.. ... "",---
.---_ ..--
4 __ - ., _ _ ,
()~
V~ •• I~t'
. , ._"
. . ,our<. 'Iudi,,_
riG. 11
PLAN OE POSITION
FJC. 12
- 181 -
3.2. - DETERMINATION DES LIMITES DE LA CALDERA ET ETUDE DU MILIEU
ENCAISSANT
3.2.1. GRAVIMETRIE
- 182-
'-----'~~=---------------------
Fig . .13 -_..--.... _ . -
_~;..:-~.i.:..-
--------~ ......
"
MONT OORE
rIG. 14
-IH -
On remarquera au sud de cette limite une ser~e de discontinuites
traversales mQins marquees, bien visibles dans la topographie et tres
probablement liees aux limites des coulees. On notera au passage que par
exemple, si l'on attribue a la coulee formant la Montagne de Bozat et Ie
Rocher de l'Aigle une epaisseur d'environ 150 m cela suffit pour expli-
quer l'anomalie positive d'environ 5 milligals qui marque cette region.
Les donnees obtenues lors du leve INAG-BRGM de 1972 ont ete trai-
tees sur la region du Mont Dore par Ie B.R.G.M. en collaboration avec
l'I.P.G. Clermont-Ferrand (cf. annexes) et par Ie Centre Geologique et
Geophysique de Montpellier (cf. annexe).
~ (cf. annexe )
- 184-
i
Ii: I
\
~
\'l
'"
;0
Co
<>
6
w "5
a:u
0·"
o ~
>- .!!
5:!i
~i!
UJ ..
~ Co
~ ~
z~ .-~
~ .g-
:::r .0
u
Co
E
o
.c
u
\
i
u:
\
- 185-
observes. La figure 17montre les resultats obtenus et l'on peut consta-
ter qu'en effet, nous sommes ici en presence d'un champ d'origine super-
ficielle attribuable au modele presente et d'un champ d'origine probable-
ment plus profonde (obtenu par soustraction entre Ie champ du modele su-
perficiel et Ie champ observe) avec Ie maximum deja decele (fig.16 ) sous
Ie massif du Sancy, et dont la reduction au pOle fournirait probablement
une image un peu plus "contrastee" mais tres voisine de celIe de la fi-
gure 16.
-186 -
'. .. ... --
'" ...--.... / \ ... . -
"." ".-~
RII0Iet·Haut: .PY'I de I'.."h
Ro."" 00
" . :. . _ * , " Rulas.au l'AllIe I"'t Dora
_: o. Vendei..lc J Rul.aA4u
l (~
: ) c:lnVrru .. ~
-= ~~~ -.~
LE HO,1fT DOftE
00
..... ...... ANOl-lAUE KAGNETIQUE
Gu.",
RESIDUELLE A 1000 ~
Ul _ to
1 .....
- 188-
MONT-DORE
CHAMP MAGNETIQUE
D'ORIGINE
SUPERFICIE LLE
REOUIT AU POLE
o, 11I:m
,
00
'"
Limite. de 10
,;" caldera d'apres
/ 9rQvimetri~ eo'
Quires methodes
rIG. 21
3.3. - ETUDE DU FOND ET DES STRUCTURES INTERNES A LA CALDERA
-190 -
\ .
\, "I"H-
\
\
\
"
\
\
\\
\
\.
_ ... . . . ................... \
z
a ., '.
i=
u
""\ \
'.
<{ \ ~
0::
I.L. . \" .... o
l1J \ I;
..,
0::
l1J
=> iI " '.
C /
" ••~~
')
~ o
Vl
in
(\ :\,
I
.'\.
l1J
o o \ '.
...J
u::
a
\! !'1 _ _ _ _ _-
1
0:: o I -.;u~ ,CIOt ~
0..
=>
!
o
~ o
z
a
...J
.... ---
l1J
...J I
~ 0 } i
" i
l;j
J:
it
i,
\i " t I
~ l~ A'
~ !
l1J ~------.;..----
~
"
,/ ii
~ ,·~i-----:~Io~~i----~~>.
a : \
u " \
,/ i 1
: i ~
: i t
: i I.
I I It
: li
I
I
I iH
A_
'"
N
- 191-
Apres interpretation (en cooperation CRG-B.R.G.M.), les resul-
tats montrent la presence de deux couches conductrices. La base de la
plus superficielle suit approximativement la premiere interface sismique
et surtout plonge tres brutalement des que l'on atteint les bords de la
caldera.
-192 -
\ .~~ I
no 140 142 ...a
o ~
~
MONT-OORE
't~ I MAGNETOTELLURIQUE
co
5.EX
.97
. CONDUCTANCES
'72 (mhos)
\ .
l.
\
_ L1miles de 10
\ ~ caldera
/
I~.
. 20 ---
,•" 22 ()I~
,."
. 185 . \ ~
•
t
22 \ \\ ./{60~~400~/400
\
'\ ... \ - t.,
-.J\,.....-= ;,.
:.~501 •
7'i.
,., \ " '(210)"::......
. -. . . .: -----.,00___" "-37l FIG. 23
:: '' .... ,- ---.,
............. 200 __
f- ilO
- 1
J
J 'I
f i.
!
; ;' :. eli
~ ~. ;"1 i)
.. ~ ! ! i~ 1 ..... - .... - - - - --
o
:;l0l.
- 194-
Le CRG a utilise, de son cOte, la methode dite "magnetotellurique
harmonigue". Les resultats des sondages sont rassembles sur la figure 24
ou l'on a ajoute les courbes de niveau (quelque peu "lissees") de ce qui
paralt ~tre Ie toit d'une couche conductrice profonde.
POLARISATION
SPONTANEE
, - Urtule-s CI~ kJ
..", coLdero
,...........
,
--
...... ..... ,
----- .....
;::
- 195-
On s'interessera davantage aux anomalies negatives qui jalonnent
la vallee de la Dordogne et traversent le massif du Puy de l'Angle et du
Barbier (a l'est) pouvant indiquer des penetrations d'eau importantes
vers les profondeurs de la caldera.
- petite anomalie de PS
Cette seconde zone donne les meilleurs resultats avec les metho-
des les moins conventionnelles, mais apparait moins attractive si l'on
considere les resultats des methodes les plus classiques.
-196-
P'IG. 26
CALDERA DU MONT.DORE ;
ca~
.
COUPE INTERPRETATIVE SUD· OUEST - NORD.EST /
100mV
.... \=~
-- ~ /~\.....•...-../-........•.....,..••..\. .
.,.j
.. ...... .........•. .. -so
.......•........
..... 1
, Anomalt. 01- / --- zooo
,,
-10
, ao..t~r ( d;;l.4 ) /
/
/
\ /
\ //
- IS \
\ /
\ /
-%0
\
\
\
\ ., I
I
I
...
% 1000
---/-
.".
, 1' .....
, I \
' ....
-%S
SW
15<Xl
1000
!..JMII" 0. k) ~ld.rc
- 197-
4. - STRATEGIE D'EXPLORATION PAR FORAGE
4.1. - INTRODUCTION
- 199-
4.3.1.3. - Couverture constituee de materiaux volcano-sedimentai-
res (ponces, lahars, depots lacustres) souvent argilises, recouverte au
voisinage des bordures de la caldera de coulees recentes. Cette couvertu-
re est recoupee par une tectonique recente permettant les manifestations
thermales.
- 200-
Des mesures regulieres seront effectuees au cours du forage
elles sont detaillees par ailleurs dans Ie programmme de forage.
- 201 -
GEOTHERMAL SURVEY OF THE SOUTH-EAST MARGIN OF THE FRENCH MASSIF-CENTRAL
B. BOSCH, P. DEGRANGES, J. DEMANGE, M. LELEU, A. MARCE, C. SARCIA
Summary
The sprin~of Lamalou and Lodeve area are ancient waters, with depth
temperature close to 100°C ; the waters in depth are in one or several
non carbonate reservoir.
-202-
In order to study the geothermal possibilities of the south-east
margin of the french Massif-Central, we proceeded to a geological and
geochemical survey of this area.
2 - RESULTS :
The chemical results expressed in 10- 3 mole/liter are presented
!~~l~_! for major elements, and !~~l~_!!, for minor elements.
The isotopic results, expressed versus SMOW for aD and 00, versus
C.D.T. for oS, and versus P.D.B. for OC are presented !~~l~_!!!.
The dissolved gas analysis, expressed in mole/liter are presented
!~~l~_!Y ; free gas analysis, in percent, !~~!~_Y.
3 - INTERPRETATION AND DISCUSSIONS
3.1. Waters origin
The oD.oIS data, presented !!g~E~_t_roughly agree with the line for
meteoric waters, as shown by H. CRAIG (1961) and W. DANSGAARD (1964).
The data for the Vals-les-Bains area waters shows a supply from higher
altitude than for Lamalou and Lodeve waters.
- 2C3-
The whole results of the CO 2-HC0 3- system are presented on ~!g~!~_1_ ;
in the upper part of the diagramm are the waters with an initial magmatic
carbon, in the lower part the waters with a biogenic initial carbon.
3.3. !~~_~~~f~_~E~g~~
We present on !!g~!~_~, the SOq content versus the 6 3q S of the sulfate
this diagramm reveals three different sulfur origin :
A deep origin is the case for the major low SOq waters
A sedimentary origin is proved for the Lodeve area hot waters
The Asperjoc and Bagnols sulfur indicates a bacterial activity from
sedimentary sulfates.
3.4. !~~_!~~~~E~!~E~_E~~£~~~_eY_!~~_~~!~E~_~~_~~E!~
Various methods have been used for the depth temperature determination.
-~-
3.6. The Lodeve area
On a chemical view point, the 9, 10, II waters are very similar with
the Lamalou springs ; the 7 and 12 springs are distinct and geographically
more distant; the 8 water is a cold reference. For the 9, 10 and II springs
the Na/Li temperatures (C. FOUILLAC, G. MICHARD, 1979) are close to the
quartz geothermometer temperatures, and leads to estimation between 80
and loooe.
For the 18, 19 and 22 springs the silica and Na/K geothermometer leads
to similar depth temperature estimation, ranging from 80 to ISO°C.
The 3H, aD and 018 0 data show an important participation of recent waters
without any evident mixing phenomena and a supply area at higher altitude
than in the previous cases ; the important rate circulation could be due to
the reservoir fissuration.
1 lA.5 1i.31 0.41i7 ,.215 3.10A 7.7(,1 ) . ' 4 <; 1. <;62 0.677 1 n. 5 'IS
2 19.5 6.28 1. 067 1.<l7l 1.070 10 217 0.'121 0.101 0.733 1 <;. Ii 72
j 15.0 6.00 1. 083 2.1Ql 1.0Q4 3.100 0.744 0.062 0.40'1 9.2~5
4 22.0 7.~1I O. <'83 0.'l7R 0.'i02 o. no 0.01, 0.O'i2 0.152 2.76~
5 12.0 6.19 1. 383 'I.3,1i 0.6P) 0.304 O.OR~ O.0r.3 0.107 70.079
6 25.0 7.fi2 0.2fi7 1.0,8 0.8P'4 0.20'1 0.070 0.115 0.231 3.577
7 19.0 6. '10 0.500 10. 77~ l.<)6~ '.97Q 0.06 0 13.741 1. 326 LR??
8 18.9 6.<10 0.lfi7 2.485 LillI 1.,7.2 0.146 11.500 0.705 8.283
9 34.5 6.40 0.458 ~. 362 2.04 'l 15.926 1. 2S8 2.707 2.70R 23.790
10 31.0 6.45 1).458 8.50R 2 ?'i0 10.261 1. ~ 37 3.227 2.4R? 30.23~
11 4<;.0 fi.~l 0.6AJ 1;.7A6 2.081; i q. 4), 1.11Q 1.123 3.nG 25.334
12 22.0 7.2R 0.21lA 2.131 O.5'il 0.109 n.02() 0.141; 0.1;21 4.71'i
13 50.0 6.38 1. 000 5.763 2.271 10.783 ?. 4 q4 0.115 n.'iOR 25.·QO
14 'il.A 6.40 O.RQ) 'j.7RR 2.291 1'1.913 ?.HiO 0.125 1I.r;nQ 7."'.2lili
11) 44.5 fi.42 0.BB3 '.4A<I 2.1AO 10.7Q '.11'i 1).IR7 O.'iOR 74.641;
16 24.7 5. <10 0.~25 3.144 1. 37~ 1. Q7.<, 1.nq 0.313 0.417 11.0114
17 20.0 ~.55 0.4';7 1.IRR O. ~76 1. 4"1 n.A1~ 1. OR 3 0.451 1. 'i ~o
18 41.5 6.A8 1. 250 0.334 0.041 q • ~OQ n. 17 n 1.4QQ O. <Ill 5.224
10 13.5 6.211 0.311 4.017 2.9SQ 13. 478 1).~n6 n.146 1.R~2 42.527
20 13.5 6.68 1. 300 1. 337 1.641 74.7~3 1. J 71 n.<;Ii? 2.200 7Q.5'iR
21 11.1) 6.78 0.5A3 O.qO~ 0.814 a.OR? O.t''lq n.135 0.564 11.2~1
22 52.0 6.76 1. 300 0.344 o on R.on n.2?J 0.323 O.5~4 R.04<1
23 23.2 6.00 2. O~O r,.040 2.(,r,5 Q.1l2t; 0.Q41i (l.1~6 '1.2(1(, 21.4R~
2~ 12.0 5.52 0.7~7 1.043 0.M7 2. "7R J (l. HI 0.073 0.107 6.3 QlI
- 205-
TIIBLE II TRACE ELEI"ENTS IN 10- 3 /o()LE/LITER
Spring B F Li fr Fe ~"n
- 206-
TABLE IV DISSOLVED GASES, IN M:lLE/LITRR
- 207-
TABLE VI: THER"10~1ETR.IC DATA If\! °C
5 - BIBLIOGRAPHIE
S. ARNORSSON, 1975
Application of the silica geothermometer in low-temperature hydrother-
mal areas in Iceland.
Am. J. Sai., 275, 763-784
,
J.C. BAUBRON, B. BOSCH, P. DEGRANGES, J. HALFON, M. LELEU, A. MARCE,
C. SARCIA, 1977
Recherches geochimiques sur les eaux thermales d'Amelie-les-Bains et
de la zone de Plombieres, Bains-les-Bains
SeminaP on geothermal Energy, Brussels, 6-7-8 deaembre 1977, II, 385-
414.
H. CRAIG, 1953
The geochemistry of the stable carbon isotopes
Geoahim. Cosmoahim. Aata 3, 53-92
H. CRAIG, 1961
Isotopic variations in meteoric waters
Saienae, 133, 1702-1703
W. DANSGAARD, 1964
Stable isotopes in precipitations.
Tellus, 16, 4, 436-468
A.J. ELLIS, 1979
Chemical geothermometry in geothermal systems
Chemiaal, Geol., 25, 219-226
C. FOUILLAC, G. MICHARD, 1979
Un geothermometre empirique : le rapport Na/Li des eaux
C.R. Aaad. Sai., Paris, B, 34, 19 fevrier 1979
W.G. MOOK, J.C. BOMMERSON, W.H. STAVERMAN, 1974
Carbon isotope fractionation between dissolved bicarbonate and
gaseous carbon dioxide
Earth Plan. Sa. Let. 22, 169-176
A. SCHNEIDER, 1970
The sulfur isotope composition of basaltic rocks,
Contr. Mineral. and Petrol., 25, 95-124.
-209-
a
. !
ti
I I
io
o" 0
y
i' .
I !
88
o
+
o
11
•
~
!i
+
o
11
•
~
il
.
- - - - - - - - , - - - - - , - - ----~------~-----'''-
- 210-
• 504- mg/I
34 -
504 sedimentaire
15110 34504- oxvdo-rtlduit
07 par bacterias
1000
I ~
500 I I
4110
I
300 010
011
200
I 09
I 3+
I
.,8
100 .17
I
I
50 .2
0 8
40
1
30 ... 16 I t 22
I ~15
20
23 e+,2 .,9
.21 .,4
10
I +2 +61
13 •
ft"5
.26 .24
5~ +:+1
4
25.
3
o LodeV8
I
2 • Lamalou
+ Herault peripMrique
34~ .ft. . ..
I ,
• Ardoch. - Lazer.
~
-6 -4 -2 0 4 8 12 16 20 24 28
i.!.45 dans volcanitas 05%0
f"
- 211 -
·"
.'. ."
. ".,.
.,. ."
."
."
... ••
.,. -' 0 .,
Quartz
,. 2. ,. •• T6mtrgtnot
50
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
~/
- - AlbitelM.croeh M
_ _ MontmOhllonite NIJMontmatlllon,tJo t(
I"", II(.I HI
- 212-
GEOTHERMAL EXPLOITATION IN GREECE
by
M.D. FYTIKAS
SUMMARY
1. INTRODUCTION
- 213-
Greece has an aggregate area of 132 000 km 2, 18% of which consists of
isLands and a popuLation of aLmost 9 miLLion.
The country does not have a vast energy potentiaL. It has no Large coaL-
fieLds and the onLy oiLfieLd which has been discovered is smaLL.
Some 50% of the country's eLectricity requirements are met by power sta-
tion which operate chiefLy on Greek Lignite or by hydro-eLectric power.
FossiL fueLs are thus wideLy used to generate eLectricity and for aLL
other uses (urban heating which consumes 15% of imported oiL and trans-
port) and are obviousLy a major burden on the country's baLance of pay-
ments. These circumstances expLain the naturaL interest in the various
non-conventionaL energy sources, Like geothermaL energy, which couLd
provide benefits for the country's economy.
Figure 2 summarizes the heat fLow data so far avaiLabLe for Greece (Fyti-
kas and KoLios 1979) which seems to have five main areas with positive
flow anomaL ies C> 100 mW m-2).
- 214-
The most extensive anomaLies occur in conjunction with the active voL-
canic beLt and the areas affected by the vuLcanicity in the PLiocene-Qua-
ternary. The other major anomaLy area in continentaL Greece is in Makedo-
nia in a region which exhibits intensive destressing tectonic activity from
the PLiocene- Quarternary which has formed Large graben, e.g. the graben
of Strimon.
In 1971, IGME began prospecting on the basis of the MarineLLi report and
with the heLp of various foreign advisers. This confirmed the existence
of the geothermaL fieLd of MiLos. When the first expLoratory boring in
1971 proved positive, the nationaL eLectricity corporation (PPC) made a
financiaL contribution to part of the programme and had two deep borings
made. DriLLed in 1975 and 1976, the hoLes Led to the discovery of the vast
geothermaL fieLd under the isLand.
After the expLoration work on MiLos, the IGME began prospecting in various
other parts of Greece with the heLp of various foreign experts, some of
whom were sent by the UNDP.
At the same time, the PPC sought the advice of the geothermaL department
of the ItaLian ELectricity Board (ENEL) which worked out a new expLoration
programme based on three seLected areas of geothermaL interest apart from
MiLos which were practicaLLy the same as those aLready picked out in 1970.
- 215-
These areas are, in descending order of importance, Argennos and PoLichni-
tos (Lesvos), Sousaki-Loutraki (Corinth), Aedipsos (Eubea), PLatystomo
(vaLLey of Sperkhos) and Poros (ENEL, 1977). The whoLe expLoration pro-
gramme in these areas is expected to be compLeted by the end of 1980.
Some of the work wiLL be carried out by the IGME and the rest by service
companies and nationaL and foreign bodies.
The IGME in particuLar, is busy working out programmes for finding and
using Low-enthaLpy fLuids in various parts of Greece.
Many regions have been designated for this purpose. At first sight there-
fore, the geothermaL potentiaL avaiLabLe for agricuLturaL use and urban
heating wouLd seem particuLarLy high.
There foLLow brief descriptions of the areas which the basic information
aLready avaiLabLe shows to be of potentiaL interest. In aLL these areas
expLoration work is under way, but at various states of deveLopment.
- 216-
Vulcanicity has given rise to typically calc-alkaline rocks. It began in
the Upper Pliocene with large explosive eruptions which deposited a mainly
calctufa-ignimbritic series.
From a geochemical point of view, most of the thermal waters are salty
with a pH of below 7, a relatively high Band NH4 content and a low
Na/K ratio.
Some very large bLock of schist and limestone were flung out by the most
vioLent expLosions so that craters may be up to 600 m in diameter. Some
60 thermometric weLLs were driLLed in order to survey the geothermaL po-
tentiaL of the isLand and these showed temperature gradients of up to
1°C per metre.
It took two years - from 1971 to 1973 - to compLete aLL the geoLogicaL,
geochemicaL and geophysicaL prospecting work on MiLos. The programme cost
a totaL of some US Z 80 000. By the end of the project, it was possibLe
to identify the potentiaLLy most promising areas and to site the expLora-
tion and production weLLs.
The main particuLars of the two weLLs driLLed are given in Figs. 5 and 6
(Fytikas et AL. 1976)
At the bottom of the first weLL MZ1 (1101 m) temperatures of 310°C were
found. The weLL produced between 46 and 60 tonnes per hour of mixed fLuids
chiefLy steam with an enthaLpy of some 450 KcaL/kg.
The information on the two weLLs (particuLarLy the MZ1 weLL) can not yet
be taken as finaL as production times have so far been very Limited.
After a hiatus of about four years, the PPC has scheduLes another two
deep expLoration weLLs which are to be driLLed by 1980 with a view to
setting up a first smaLL 10 MW piLot pLant.
- 218-
4.2 Nisyros
The island of Nisyros is also part of the active volcanic belt. It has
an area of 40 km 2 and some 1000 inhabitants and is relatively hilly.
After the formation of the caLdera, a number of dacitic domes formed in-
side it (Di PaoLa 1974). There were numerous phreatic eruptions and in
some cases these have produced Large craters. The most important feature
of phreatic origin is a crater inside the caLdera of some 250 m in diame-
ter.
The most recent phreatic eruption occured in the last century. There is
intensive fumaroLic activity inside the caLdera. There are hot springs
with maximum temperatures of 55°C, even on the sLopes of the voLcano.
The geoLogicaL structures and the geothermaL data recorded to date indi-
cate that this area has considerable potentiaL.
- 219-
4.3 Thira
The island of Thira has a complex volcanic system which is still active.
The oldest volcanic products date from the end of the Pliocene (Ferrara
et al. 1979). The year 1500 BC saw the great eruption which resulted in
the formation of the calderic depression which gives the island its cur-
rent aspect.
Various eruptions inside the caldera, the last of which was in 1950, have
resulted in the formation of the small island of Nea Kameni.
The isLand has a number of energy probLems, chiefLy because of the subs-
tantiaL deveLopment of tourism, and a serious water problem.
4.4 Sousaki
-m-
tiaLLy covered by a Neogenic series which in some pLaces is very prominent.
The voLcanic formations date back to the PLiocene and consist of domes and
dacitic fLows which are generaLLy of Limited extent.
The main directions of tectonic activity run E-W and NW-SE. The Latter
is probabLy more recent.
4.5 Methana
-221-
4.6 Lesvos
The island of Lesvos is inside the volcanic belt on the eastern side of
the Aegean sea (Fig. 8).
The thermal area of Polichnitos is the most extensive on the island with
20 surface springs of a total capacity of between 13 and 25 litres per
second. The waters are particularly rich in chloride and their salinity
is between 11 and 12 grammes per litre.
Although there has been no recent volcanic activity in the area, it can
also be considered as potentially promising for geothermal energy because
of the major destressing tectonic activity which could have had a positive
effect on the formation of a geothermal anomaly faily near the surface.
- 222-
4.7 The vaLLey of Sperkhos
The vaLLey aLong which the Sperkhos runs consists of an east/west PLio-
Quaternary graben (Fig. 9). The southern part of the feature has fauLts
with marked dispLacements aLong which emerge severaL hot springs with
temperatures of between 35 and 42°C (TermopiLi). Emergent formations con-
sist of a Mesozoic carbonate series covered by an ophioLitic fLysch for-
mation. Inside the graben are fLuvio-deLtaic sediments dating from the
Quaternary which ties in with the very recent origin of the graben.
The springs of higher fLow (some 20 Litres per second) are in TermopiLi.
The waters contain some 6-8 grammes of saLt per Litre and are of sodium
chLoride and caLcium bicarbonate (Dominco and Papastamataki, 1975).
Nineteen weLLs have been driLLed in the area to depths of between 50 and
250 metres but have encountered onLy moderate temperatures of around 45°C.
In the Ypati areas there is an NH4 anomaLy. A smaLL test borehoLe driLLed
down to 325 metres showed a 280 metres aLLuviaL Layer above the carbona-
te series. The temperatures of the fluids at the bottom of the well are
around 52°C. Abundant CO 2 carries the hot waters up to 10 metres above
the wellhead.
4.8 Aedipsos
StiLL on Eubea, but opposite Aedipsos there is another hot spring (430C)
with a fLow rate of around 1 Litre per second.
The hydrothermaL system in this area seems to be connected with the huge
fauLts which extend through the region and aLLow the deep hot fLuids to
escape.
-223-
4.9 SaLonika
ImmediateLy to the east of the Large town of SaLonika there are two Large
graben of recent origin which run approximateLy east-west. The most nor-
therLy of the two has two Lacustrine basins and three hotsprings - with
maximum temperatures of around 50°C - which have deposited travertine.
There is seismic activity in the area and in fact the epicentre of the hu-
ge earthquake from which SaLonika suffered in 1978 was aLong a fauLt in
this graben.
The more southerLy graben of Sedes has two moderateLy hot springs
(27-36°C). The speciaL position of this graben, which begins aLmost imme-
diateLy south of SaLonicco, wouLd seem to warrant a more accurate evaLua-
tion of its geothermaL potentiaL.
4.10 Strimon
The thermaL springs are basicaLLy in two areas, one near Siderokastron
aLong the northern edge of the graben and the other at Nigrita on the
south-west side.
The first area has a number of springs with maximum temperatures of around
56°C which deposit Large quantities of travertine. These waters have a
reLativeLy Low saLinity (1.5 g per Litre).
- 224-
GeochemicaL thermometers indicate probabLe temperatures of between 90 0
and 1300 C.
The economic conditions of the area wouLd seem to favour the use of Low-
enthaLpy geothermal fluid for urban heating and agricultural use. All the
lowlands can be used and are in fact very fertile.
4.11 Thrace
There are various thermal areas in the region of Thrace. The main area
is the Xanthi pLain some 20 km south-east of the twon where there are a
number of springs with maximum temperatures of around 54 0 C. The waters
are relatively rich in gas and their salinity is around 5 g per litre.
It is highly likely that the springs are Linked to the main regional
faults, which run chiefly in an east-westerly direction and have caused
the lowering of the vast area which is now the Xanthi plain.
5. CONCLUSIONS
We hope that, with cur own resources and with the help of the European
Communities, which we are proud to be joining, geothermal energy can be
developed in Greece so as to make an appreciable contribution to the
country's progress.
- 226-
Fig.1 . Structural skatch map of Gr""k Area scala
o-=-60Km
Main vOlcani c cant "rs 01 acti ve arc
~
Main volcanic ca ntars of Upp.Miocene - Pliocene age
- 227-
- 228-
1 r-I Alluviua, fluviaU le - torren-
L--J tl III IUld coutal depo.! ....
• cr.e. lind fana, He.n' non
volcanic tuft••
q- -
: ~o.o;~,,·,;
3,'" 'I "Orun" and other hhue.
4S':~~~:~
:.:;::_::_....:,:-:.
'l\1rr • .aDd turCH ... , Y°\ln8er
and older thfJ1 5.. glowing
• &valanchlu (nuiu -.rd.tnhl),
18Dhlbrl tea, JNaiCIl r10'111 d..-
potUI &114 &lh:r.d tuff•.
DeIa•• and ILD.du1tic, ~1t1o
aDd tl\,Joliiic lava flow.,
N ocou1onallJ alhred, of
N older (0.5-1.5 lI.y.) volcU'liam
'.0
FIGURE 3
MILOS ISLAND - Comprehensive geological map
- 230-
TEMPfRATURE ANO PRESSuP.£ LOGS
STRATIGRAPHY
I:> _ _ _ _ r.".,.rAN:o-. 3"tNIW"tI4' " Or;to611" ",7
~ ~ ---- r~~,... ,1,tP'1P fM.~~4 dllt'mJ ~"I,., I~~ 10(117'5 to '51'171
...., Q.
- - - - - -Pr.ssat,. cu ..., .. ,., ,,77
~
1t'II • • Ot:~b~
'I: PRECEOING
v
INTERPRfTATION T(IttfPlItATIJIt( ·c
50 , 0 150 200 2:10 J<)O
; .. u,.oorMI df:pQSIU
\
,
100
If
Gr~f" /.IIlr
300
\
\
\
\
:..
.
\
'& 400 \
\
\
\
\
500 \
\
~,., f~.Jt '#nt's \
\
\
\
\
\
\
\
\
\
.
I
\
\
\
\
\
\
\
•
900 \
-"1tttc'o""t .. • chJon~ 'Cr,.lJ, I
_''I ~r~ s of c~·Sch)U \
\
I. ,d cMrmlo M,cNds \
\
\
\
.., tor. klnrU _t4 ,~,. MZ -/ \
, .,.d r pjl/ol~
/;'-,.£(£.;
• 4 • ~:;r o;... tl'''' ~JA _,a;z.
1100 k'l;; ,., Sch"" .,.,., qil.-tzllrs
- -- ... """",It c . kd~ ."tI
.~ Ido'~
'ItESSVI41 ",/, .. -
'0 ,00
500
.!Wt.
700
..
~
"Q
~ IKlO
900
,~
1000
~ MA-1
'Q 1100
tL
So 100
OEPMOKPAl'A' rEOTPH1EON
O~
I
20 )()
~~~~~~
~O 50 eo 10 10 90' C
,
100
10 •
yr
\\\\..
'" .
,0 L
.., . r1'~ \
.
~\
.:.,
\\\
10
\\ \
\ '\ '~'''''
\ \
00
90
\ c,
~
\ \
'00
\
110
rJ •
120
~
...
EB
--;.,
~.
-""
C"
'". . ""
'"-
w
'"
....
1- M .. tamorph lC' bl!ls~m~ nt :2 - PUldol/ Ces . ,.,a",ly 3-lgnimbYlI~s 4 - N"oge"ic vo/can,sm (16-IB",'1) 5 -N~0geni( conr,~e" ltl/
/d"Hlri~" .. d,menls 6-Fma/ volcanism 7-all~te,"ary 8- Hot springs
"
.. r--I "llun .. e.nd continental .. ~eDh
L--J (Rec.ent Phhtoc.ene - Oloeene)
N
[.~:<;::':):.~I1'raVer1in. depo81tl
SeQ Ie
oi-. -l
_ _ _ _'0
• "rhermal .prinSl'
FIGUliE 9
F'ig.l0. GEOLOGIC.AL SKETH MAP OF STRYMON AREA
Sca I~ 1: 500 000
Alluvium
k~':::<~::;;'.ii! 1 Loam sands terra rossa cong lom~r~tes tray~r tlne$ (Qua tevn~ry)
AC Jd '"trusiv~s
Ophio li t"s
Amph i bolites
Marb l ".
- 236-
BIBLIOGRAPHY
FERRARA G., FYTIKAS M., GIULIANI 0., MARINELLI G. : "Age of the Formation
of the Aegean Active VoLcanic Arc". Acta of II into Sc. Congr. 'Thera
and the Aegean worLd', Athens (1979)
- 237-
A GEOLOGICAL AND GEOTHERMAL APPROACH OF THE PAVIN AREA.
Summary
In the scope of the evaluation of the geothermal potential of the Chaflle des
Puys, studies are being effected in its southern part, in the Pavin benmoreitic
volcano area. The present results are mainly geological and consist in volcanolo-
gical and petrological data about the Pavin and the volcanoes with less differentia-
ted lavas associated to it. A conclusion has been drawn that the possibilities
of a geothermal anomaly in this area depend on two main factors : the fact
that a high level acid intrusion under the Pavin exists or not, and the depth
of the differentiation magma chamber. A model of a magma chamber is pro-
posed for the Chafne des Puys. Some parameters of this model, and particularly
its depth, have still to be precised. The first geological results of a drilling in
order to measure the geothermal flux are presented and discussed.
- 238-
1. INTRODUCTION
Some researches considering the evaluation of the geothermal potentiali-
ties of the Chame des Puys (Massif Central fran~ais) have been going on for some
years now. A first borehole in order to measure the geothermal flux was realized
in 1976 in the central part of the Chame, at the foot of the Puy de DOme, and re-
vealed a moderate heat flux value (1,7 UFT).
More recently, the investigations have been carried out in the farthest
southern part of the Chai~ in the neighbourhood of an acid lava volcano: the Pavin
crater-lake. The Pavin and three associated volcanoes with less differentiated
lavas form a volcanic group clearly a part from the rest of the Chame des Puys
and thus offering a slightly different volcanological, and perhaps geothermal
context. Not only the Pavin group is slightly younger (6760 to 3450 years B. P.)
than the rest of the Chame, but its very eccentric position also suggests that the
acid lavas originate in a magma chamber independant from the main differentiation
reservoir (fig. 1).
2. VOLCANOLOGY
The Pavin holocene group is located on the southern side of the Mont-Dore
plio-quaternary composite-volcano and rests on a volcanic basement (the crystal-
line basement is exposed all around for a few kilometers, at lower altitudes -
1150 to 1170 m). It includes four volcanoes : two strombolian cinder cones, one
basaltic maar and one acid explosion crater (fig. 2).
The strombolian Puy de Montcineyre and Puy de Montchal have spread
over ash and cinder blankets (respectively 30 km 2 and 13 km 2) and have producted
lava flows, some of them canalized in the actual valleys, covering great distances
(16 km for a lava flow from Puy de Montchal).
The Estivadoux maar, like many other explosion craters of the Chame
des Puys (1), results from a phreatomagmatic eruption of a basaltic magma. The
crater, largely filled by later products, is only today a slightly marked depreSSion
whose initial diameter must have been about 1 km. The maar ejecta are exposed
over several km2 and consist in finely bedded polylithological deposits including
both dense, vitrous juvenile lapilli and lithic essentially volcanic debris,
The Pavin explosion crater (2) is a subcircular depression with a diame-
- 239-
ter of 950 m, cut out in the substratum and filled by a deep (92 m) lake. The for-
mation of the crater, resulting from explosions accompanied with collapse, is due
to an eruption the complex dynamism of which has both plinian and phreatomagma-
tic characteristics. The ejecta have been blown apart by aerian clouds giving air-
fall deposits and by basal blasts producing base surge depOSits. The totality of the
Pavin deposits is loose and in most cases bedded and its thickness and particle
size diminishes rapidly with the increasing distance from the vent. The acidic ju-
venile magma consists mostly in highly veSicular pumice and in dense, vitrous and
slightly vesiculated lapilli and bombs, the latter resulting from the quenching of a
part of the magma in contact with superficial ground water, before and during the
vesiculation. The magmatic fragments are mixed, in a slightly equal proportion,
with lithic fragments of ancient lavas or basement rocks. Nowadays the Pavin de-
pOSits cover approximatively a 100 km2 ovordal area, the distribution of the ejecta
have been essentially southwards. From an isopach map (fig. 3) the volume of the
ejecta can be estimated at about 75.106 m3, which is a minimal value of the initial
volume. The corresponding unvesiculated magma volume is about 15.106 m3.
In most of the trachytic or benmoreitic volcanoes of the Chalne des Puys,
the initial explosive activity is followed by the extruSion of a gas-poor magma to
form an endogenous dome of viscous lava ("Puy de DOme" type) or a spine of al-
most solid lava ("Puy Chopine" type). The Pavin shows the particularity that no
extrusion took place within the crater. It seems thus likely that a gas-poor part of.
the magma was not able to reach the surface but now rests at a shallow depth.
Field investigations enable us to define a very reliable tephrochronology
of the Pavin volcanic group, with a Montcineyre - Estivadoux - Montchal and lastly
Pavin sequence. This chronology is inconsistent with the previous radiocarbon
dates (respectively 5750, 6660, 6760 et 3450 years B. P.) probably due to unrelia-
ble C 14 data measured on fossil soils.
3. PETROLOGY
The various lavas of the Pavin group correspond to three degrees of diffe-
rentiation which combine very well with the low-potaSSiC series of the Chame des
Puys, according to the petrographical, mineralogical and geochemical aspects.
- 240-
MAURY (3) proposed for this series a classification following the values of the
differentiation index (D. I.) defined by THORNTON and TUTTLE (4). According to
MAURY's nomenclature, the lavas of the Pavin group are a potassic basanite
(D. I. = 26 ; Montcineyre - Estivadoux), a X-hawaiite (D. I. = 40 ; Montchal) and a
sures which can explain the prominent role of amphibole. The water pressures,
estimated at 2, 5 to 3 kbars, would correspond (if water pressure is equal to the
total pressure) to a depth of the magma chamber of 7 to 10 kms.
4. DISCUSSION
The volcanological and petrological data (4) enable to define some of the
- 241-
characteristics of the magma reservoir where it is assumed the fractional crystal-
lization occurred: approximative depth, nature of the filling initial magma, possi-
bility of a shutting of the chamber with an evolution as a rather "closed" system in
the last stage. These characteristics, that we can reasonably extend to the Chame
des Puys as a whole with reference to other studies, added to other significants
facts such as the structural setting of the Chame closely connected with the tertia-
ry fault system, allow us to modify the model of magma chamber previously set
up by AUBERT and CAMUS (5). The model proposed here is based on the assump-
tion that the setting of the sub volcanic reservoir(s) of the Chame des Puys, like
the arrangement of the volcanoes at the surface, was mainly controled by the ter-
tiary distensive tectonics. The fig. 4 shows a possible mechanism of setting of a
magma chamber in the normal conjugated fault system of the horst forming the
basement of the Chame. Such a model implies a geometry of the high part of the
chamber in the shape of a rather acute corner. This geometry seems to explain
well some volcanological characters of the Chame, particularly the slight volumes
of the most differentiated lavas and the quickness (a few thousand years) of the
evolution.
From a geothermal point of view, the implications of this model are to be
considered. It seems indeed likely that "corner-shaped" magma chambers are able
to support overpressures. Thus the estimated water pressures could reveal over-
pressures and lead to overvalue the real depth of the reservoirs. More detailed
geophysical and petrological studies would allow to appreCiate this depth more pre-
cisely. The most recent geophysical data (6) (7) indicate that no magmatic body has
to be expected less than 5 kms deep.
Several basic conditions are present in the Pavin area to suggest its geo-
thermal possibilities. The results of a geological survey, briefly described here,
lead to a best appreciation of them. It finally appears that the possibilities of a
superficial geothermal anomaly depend on two main factors :
- the presence of an acid magmatic body at a shallow depth under the
Pavin or not.
- the depth of the differentiation reservoir.
N @uf fond
**
Son son
~B esse
C hou .~ -
'
,
!Y'*0~
Mo n lc ho l 0 I Q:-
~
o *
Men' ,"e' .......
cinder cone
L-"
( " basaltic maar
v'"
<:> acid explosion crater
~
• tracytic dome or spine
) 0
La Godtvelle 5 km
Fig. 1 - Sketch map of the ChaIne des Puys, showing the location of the
Pavin group
- 243-
I
I
i
I
o
L.... c DE
BOURDOUl_
,!
,
,
\,
\ ,
'.
\
\
\
t
" ' f'
\.\
..
".
oE I..A CHAUMOUJrU:'
o 11(1'1'1
1
' ==-1==",,/'
®
Fig. 4 - A possible mechanism of a setting of the magma chamber of the Chal'ne
des Puys
(C) 1 - Chilled margins
2 - cumulates
3 - active magma chamber
- 246-
m
Pavin depoSit
Montchal lava flow Holocene
]
sequence
Estivadoux maar formations
- 247-
500 m South of the Pavin. A simplified log is given on fig. 5; it shows that the
well crossed the whole holocene series and part of the ancient volcanic cover. The
principal aim of the drilling, which was to reach the crystalline basement in order
to carry out reliable thermometric measurements, was unfortunately not put to
an end. It results that the gradient measurements give a tricky interpretation;
their results and their discussion is not to be given here.
From a geological point of view, the borehole is offering lots of valuable
informations. It allows at first a useful stratigraphic control of the holocene
series. A complete log of the Pavin deposits is available, whereas the bigger part
of the formation is not exposed at the surface. The existence of a masked phrea-
tomagmatic crater at the place of Estivadoux depression is confirmed, for the
drilling cut lacustral sediments (interpreted as the result of the filling of the maar
by a lake) and maar deposits (interpreted as maar ejecta collapsed in the vent).
Among the Mont-Dore formations, a thick pyroclastic level unknown at the surfa-
ce has been revealed. It consists in block and ash, almost monolithological pro-
ducts and represents a pelean-type "nuees ardentes" deposit. "Nuees ardentes"
are frequently canalized on the lower slopes of volcanoes. The presence of their
deposits and the fact that the crystalline basement does not appear at 1100 m seem
to indicate that the borehole is Situated in the spot of a more or less significant
palaeovalley. Considering this eventuality, detailed electrical investigations would
be now necessary to appreciate in a better way the surface of the crystalline base-
ment and to determine eventual other sites for thermometric drillings tallying
with the two follOWing conditions: lateral closeness of the Pavin crater and ver-
tical proximity of the crystalline basement.
We are grateful to many people for their suggestions and help during the
field and laboratory studies, without being able to name them individually. The
Institut National d' Astronomie et de Geophysique supported the geological survey and,
partly with the C. E. E., financed the "Pavin" drilling.
- 248-
REFERENCES
(1) CAMUS G. (1975) - La Chafne des Puys (Massif Central fran9aiS). Etude
structurale et volcanologique. Ann. Be. UniversWi Clermont n° 56, 322p.
(2) CAMUS G., GOER de HERVE A. de, KIEFFER G., MERGOIL J., VINCENT
P. M. (1973) - Mise au point sur Ie dynamisme et la chronologie des
volcans holocenes de la region de Besse-en-Chandesse (Massif Central
fran9ais). C.R. Acad. Sc. Paris, D, 277, 629-632.
(3) MAURY R. C. (1976) - Contamination (par l'encaissant et les enclaves) et cris-
tallisation fractionnee de series volcaniques alcalines, continentales
(Massif Central fran9ais) et oceaniques (Pacifique central) : l'origine des
laves acides. These Doc. Etat, Paris-Sud, 455 p.
(4) BOURDIER J. L. (1980) - Contribution a l'etude volcanologique de deux secteurs
d'inter~t geothermique dans Ie Mont-Dore : Ie groupe holocene du Pavin
et Ie massif du Sancy. These 3eme cycle, Clermont-Ferrand.
(5) AUBERT M., CAMUS G. (1974) - Structure profonde de la Chafne des Puys.
Bull. Volc., 38, 445-457.
(6) DUPIS A. (1978) - Electrical investigation in the Chafne des Puys (France).
Phys. Earth. Planet. Int., 17, 240-264.
(7) MOSNIER J. (1976) - Recherche d'une anomalie geothermique sous la Chaine
des Puys par sondages electriques et electromagnetiques. Bull. Soc.G~ol.
- 249-
SESSION II - Local Exploration - Drilling
-251-
DEEP INVESTIGATION OF THE GEOTHERMAL ANOMALY OF URACH
Abstract
Until recently the exploration of the geothermal anomaly of Urach was
limited to water-bearing sedimentary horizons, mostly in the "Upper Mu-
schelkalk" of Middle Triassic. By the deep well Urach 3 exploration was,
for the first time extended to the basement. Originally the well was
planned to a depth of 2100 m or 2500 m resp. but the final depth is
3334 m below surface with a 1700 m thick penetration of basement. In
this research-well 7" casing was set and cemented at a depth of 3320 m
after all the planned geological-petrophysical, hydro- and geophysical
data were cOllected. After a short phase of interpretation of the
technical and geological parameters the feasability of the basement rocks
for the hot-dry-rock concept was tested. Although only one single well is
available at Urach we succeeded in fracturing the basement rocks in four
sections and even to produce small amounts of geothermal energy by
circulating water within the tubing-fracture-annular-system.
According to our working schedule the research-operations can be sub-
divided into 3 parts:
- 253-
In order to obtain valuable electrical logs and to get a "guidance"
for coring the well was in certain sections pre-drilled with smaller
diameters, 12 1/4" and 8 1/2" resp. Average drilling progess in sedi-
ments was 2 - 3.5 mlh and 1.5 - 2.5 mlh in the basement. In the sedi-
ments soft to medium-hard tooth bits and in the basement medium-hard
to very hard insert bits were used. 1/10 of total depth was cored;
progress during coring was 1 -2 mlh in the sediments (with Diamond
core bit typeC20 of Christensen) and 0.3 - 0.7 mlh in the basement
with type C20, C23 and C24 core bits. The best results were obtained
with type C24. Drilling was twice as fast as coring in the sediments
and three times as fast in the basement.
According to the casing scheme the well was drilled in 4 steps. For
more technical details including diameters see fig.1.
With the exception of the 9 5/8" Liner all casings were run to the
surface. Cementations were made in such a way that a later production
of water in potential horizons is possible. The cementation of the
7" casing was made in two steps with API mixed cement; by this we
hope that the rocks were not fractured by cementing operations and
that the cementation is perfect behind the casing.
Summary
There are no severe problems when drilling through basement for the
hot dry rock concept. This is probably true for large parts of
South-Germany. Average drilling progress ist expected as 1.5 - 2.5 m/h.
Best tools when drilling in the basement were medium-hard insert bits
(type 3JS - 5JS, J44 and M88) for drilling and very hard Diamond bits
(type C24) for coring. In spite of rock tensions the uncased hole
(up to 1500 m!l offers no casing problems. A final casing is necessary
because of decomposed zones.
-254-
1.2 Geological results
The top of the basement was reached at 1604 m. It can be subdivided in-
to 3 units:
Ortho gneis (coarse grainedMetablastites resp.) on top
Para gneis, in the middle part, underlain by "Glimmer-Syenites"
(plutonite-like rocks with Granodiorite to Tonalite minerals).
All three units are generally marked by open and by sealed fissures.
Their dipping is between 450 and 700 ; Calcite and Pyrite are frequently.
A severe loss of mud (5.6 m3/h) occured at 1775 m in an open fissure
in the Orthogneis. Smaller losses occured in the Paragneis and near
the bottom of the hole (averagely 0.1 - 0.4 m3 /h; maximum was 1.2 m3 /h).
Oriented cores showed a slight dipping of the sediments toward SEj the
basement is steeply dipping towards N.
- 255-
circulation (via volcanic piercements and - in the basement - addi-
tional fissures) presumably influences all aquifers. The fissures of
the basement were partly signalized by severe mud losses.
- 256-
2. Fracturing and Circulating
Breaking well head pressure was 200-400 bar. 1200 l/min of water
could be injected at well head pressures of 600-650 bar.
When the pressure was no longer applied the artificial fissures closed
again; therefore a high-pressure-resistant propping material (Bauxit-
Sinther) with gel was pumped into the fractured rock. These tests were
successful and the resistance for the entering water was reduced
during later circulation-tests.
- 257-
!J' 4,0 Ip '? ~,'\'> .. '\" "l'l ,~ 2<p " ",HO .. ' 1/f ,~~ 21\0
J10 lage (Ooys]
'.
'- ,E .",..
soc Abb l ' BOhrfortsc'hrdtskurve. Geologlsches Profll
und Verrohrungsschema der
Ll.~~"~ .E Erdwarme - Forschungsbohrung URACH 3
\.E
I,
1490 --------- .. -
'500 KOlbon(St.fon)
..,, '- ~--~
N
00
'" " • ~E Orth.)IJnt l 5
\ 0.6-1,J_ ~~-.- ( M.tablostltel II~SIl'l SI 51'-
\~~~Jw . . <. ~ 16.,0"",12 !W
19S0 :,: ~
"
2000
A Arbf.ll$unt",b,,"chung Iln'~rruptlon)
I':;"
':.'.:::.
F Fongolbtll (flstung Job I '.
I
G GfSlon9. ·Tt5~ 10nil Stem T"st) \
:':/j Pa,agnIIS.
mftatelc\lscI'I
!rot MltSsung IMtosurtml. Logging)
bult'lfluflt
P Pympvl'tSuch (Pumpi ng T.,s t) l._nt,.., .. lul. 2~1'O.om
2500
., Pack" 2Sn"I""
R RotH e'lnbou I Coslnq Seulngl 'f......""s ~
~
N
S Spvlungsv"lIu5 t 'Mud L05SU I
Zeme-ntatlon I CemfnllnQI -.~-
I
'I
I
1000
- - - - - Bohren, Kernen (Drllllng,Corlng)
Erweltern (E), Nachbohren{N) (Reamlng.Redrilllng)
.. "
~~ nlQ,ClI'I'I ?"
.. 'I~'~~ ___ ~ ~"p.J.. _____ _ MI\.l 311L ).)3 "".o.IJZ .,
":2 ......... Stilistands2elten (Stand Still. Wailing T,me) M'
Bohrung URACH 3: Reih en folge und Bohrlochbilder der Pumpversuche und GestCingeteste
We ll URACH 3 Sequence on d Well Schemes of Pumping ond Drill Stem Tests
'0
PuIBp.ntinlcwt tlS1 .nI. 186.1 706, II 7a1!i.l ... 9s a .2' ft'I lComJll,.II •.tlolnO""
...... m ..... _or laut 2l!t Pv ,"8..- ~
. -.. ..
r D099·' I I
-- - - - - --
-1.40'"
,I - - - - --
-- - -- - - - -- -- - - -- , ---- - -- -- -- --
:I'I I
lC.up.'
_ ~L _ ' - - - ~ , .. __ - -- - - , - --
- - - - - -- -
MU.f;~.Ulolk :11 I
.. _----------- -- --- P -- - _.
a-:. ;::;~;~;~ ~ - -
---------- - - -- --- -- --- - ~ '. I -- I -
leoo p. fo- I=-
's
~
P.,,,,
DECKGE BIRGE
OVER 8\JRO EK "
--- --- ICQtbOf'l
-- -- .- . - - --P, f-:- - p -
5
-
~ --
l
,<0, - s-
-- - -- - _.
N f-'- ..- - -
V. .. -
". r= r
'" KRISTALUN
5, 1-'110 ,-"It.' i 1- 171 •• '
~ "
20DO0
m ~
m ~ 0
c
, c I ,~ c
,
~ P."orotlon .. str.c.Irc. " ,
l~ ". I ~ ~ ! j
~ 8entonllStopten
I II .
P PumpversuchlPumplng Tosti
~-- ~~- j l
, ~;. r' F'
S : GestongetestlOrlll Shm Testl
...
Abb IFlg 2 ,
L I r~
§
~
··
,··
'2~ L-.~-,=__ ~ u
333' LEndt~ute - - i.....
~ '- ~-
~ ~ U~
1I............
-- --- " tt"
a. "'~.
- - 1.26m +NN
QUI'! r to r P : Pumpversuch
Hydros\atische Druckhoh. Pump ing Te s '
Hydrasta1lc Pr.S5ure l-ifoad
Dogger 26 6 S : Ges tong e tes t
00 .226 Dull Ste m Teo!>.
ca 20 4m
L ias
Keuper
I
~ 665 · Y> 12 1/5
I Muschelko l k ,YyYYY-
~~~~~I------"'"
;:-:f~f £-
Perm
( Rotliegendes)
Ko rbon (S telon)
Or thogneis
(Met o bl ost ite)
a1
20
1
1.0
1
60
1
80
1
100
mvol %
Por ogne i s ,
metotekt isch No' K' Mg " Co " I Fe
beeinfl un t
1:!¥fP!1.! .. : ." i MM?7AjjjIIIIINt~I!ll~~
Cl'
8 1e tcnzont ~ :.
':
"Gl i mmer -
Syenit " Abb.lFig . 3
(Gronodior i t
bis Tonol i t)
Thermalwasser - Stockwerke in URACH
Thermal Water Bearing Layers In URACH
C1E T FtlCH79J 8 0
- 260-
"[
/~oo,~i(
Quorlo r 50 100 150
f~ ~
Dogger
Ii -
L-ia- S------f28l
f-- Temperatur - Log
31-t-
1
- - ~r;~~f'
f-------4398 · f~ I URACH
Keuper
~OO 1\ I 1 1
1_ -
-=
lbbJig
1 --
f--- - - ----f1665 B
. '
"",h.,""," A ) I
Orlhogneis I I II \ I
iMetoblost i tel - - t- -.l - i •
- 1~
.. - l---+_I.--+--+-
\
2~ ~-r-+-+-------+~
\ -~~~
- ,
~
[ . - . \
- ,
~ ,-- -
I ,
Paragne i s.
1
metotektisch
bee i nfl urI!
2500
1
r
I
I
i
- f-.-
1 I \
1
3000 -+-f----+-f--+--_-+-_+~--~_+_
"Glimmer-
Syen i I "
IGranodiorit (
1--' ___ --I:--'-+\~r- 1_
bis Tonalll)
1
L.-
_ _ ~,
Endteufe 3334m
J __ "- - - I -I------+-----+1\r --i
DIE TIIUCH 7. 1M CPI. HArtE-l'ZO'" 111
- 261-
!11
Hohenangaben
t I~,?:-
mar3stabstreu
. '. ?~ 3259 m
. ~ fl - 3264 m
Perforation 1
( ~l tll['~
,
.?
. '
. /?
c
~
?
,
' .
.
- 3271 m
Perforation 2
. rI. ." . - 3276 m
.? ?
3293 m
Perforat ion 3
3298 m
- 3302 m RTTS - Packer
I (
, ...
,.. \ { (
<' < (
- 3334 m ET
~
-~
!i
iii
-~
II
~
1l !
~
~ !
-~ ~
!l ~
Ii!
'"
....
!!!
..,-
:x:
i u
E" <I
0 o-:l~
u
-I '"
:::l
Ii! :it ::!
'"
0
- 263-
Abb'/Fig.7
Circulation Model
Urach 1979
u~~~~--~3293m
(frac 4)
~-\:-+-:+------l,-- 3320 m
(open
hole)
- 264-
GEOTHERMIK-FORSCHU NGS-PROJEKTE und ZIRKULATIONS-SYSTEME zum HOT- DRY-ROCK-KONZEPT
ProJekt d.., los Alamos $c.lenldlc: loborotory fnlt 197L, Forschunc;p.bohrung Uroch tu' t 19171
!OOO", !OOOm
luff
~Om
0099"
P.rm
-- -28lm
-380m l ias
- -lll!!!
Keuper
Pa'aOlolkum
(unOl9h,der t I
-665",
-7l0m
Mus(:h.IIcQlk
·842m
- -- - -
BuMsandsUln _ _ ____ ~
P,rm
,R o' ''.g.nd.s)
KUslalt l n
G,antt u 0:
-160'm
K'istalhn
OrlhoQn.ts
Poro9 n'tS
GtonodlOu, U Q.
hydroullsCh
jnduz .• rt.,
FRAC IRiss).
S)'s t.""
Modell Uroch
:Zwei - Bohrloch-Sys tern : Eln-Bohrloch -
-lOSOm System
Abb/F'1I- 9 - ll34",
- 265 -
3. Bibliography
-266-
LATERA EXPLORATORY WELL (NORTHERN LATIUM) ITALY
Abstract
Contract No. 571-7B-1-EG-I, drawn up and signed by ENEL and
the EEC, provided for the drilling of a geothermal well (Late-
ra 1) in northern Latium to explore a new area of geothermal
interest individuated by ground surveys.
The objective of this well was to ascertain whether there
are geothermal fluids at depth and, simultaneously, to provide
data for verifying the results of the ground surveys.
During drilling the well crossed Quaternary effusive volca-
nic products, Paleocenic-Triassic (?) sedimentary terrains and,
finally, intrusive syenitic-type rocks.
No significant permeability exists in these terrains due to
the intense mineralization that has sealed the fractures origi-
nally present in the sedimentary and intrusive rocks.
However, it has been shown that the area is indeed of geo-
thermal interest, considering the high temperatures (about
345 0 C at 2BOO m) recorded in the borehole.
-m-
1. INTRODUCTION
As stipulated in Contract No. 571-78-1-EG-I. ENEL. in joint
venture with AGIP, have drilled an exploratory geothermal well,
Latera 1, west of Bolsena Lake in North Latium (Fig. 1), on
the inside of the volcano-tectonic depression commonly known
as "Latera Caldera". The site of this well was defined on the
basis of the results of geological, geochemical and geophysic-
al surveys conducted in preceding years throughout North La-
tium.
These surveys revealed, in particular, the presence of stru~
2. OBJECTIVES
Latera 1 well was drilled to ascertain whether geothermal
fluids exist in Latera Caldera area and to verify the conclu-
sions drawn from the ground surveys.
Drilling was therefore programmed to reach a depth of 2500
m. so as to cross the Quaternary volcanic terrains, the Creta-
ceous-Paleocenic flysch facies terrains and the carbonate ter-
rains of the Mesozoic complex.
Having established that the terrains crossed to 2500 m were
impermeable it was decided to continue drilling to 2800 m.
3. DRILLING WORK
Drilling began on 1st December 1978 with an Oilwell SSO rig
capable of reaching 3000 m.
An RB 0 12"1/4 was used to 402 m. with mud circulation to a
depth of 178 m, where circulation loss occurred.
-~-
The first casing (0 18"5/8) was then lowered, with the shoe at
324 m.
On 16th December 1978, just before the above operation was
completed, the front part of the mast platform collapsed.
Drilling began again on 22nd August 1979, reaching 1284 m with
an RB ~ 12"1/4. The second casing (~ 13"3/8) was then lowered
and cemented, with the shoe at 1222 m. Continuing drilling to
1996 m with RB 0 12"1/4 the third casing (0 9"5/8) was lowered
and cemented, the shoe being at 1510 m and the head at 1176 m.
From 1996 m to bottomhole drilling continued with an RB 0
8"1/2.
-269-
flysch facies terrains of the Cretaceous-Paleocene, made up of
clay chaotically enclosing siliceous limestones, calcarenites,
mar~s and quartzitic sandstones.
The formations of the Tuscan Series were encountered be-
tween 1380 and 1720 m. These include siliceous and cherty li-
mestones with radiolaritic facies (MaIm) (1380-1430 m), marly
limestones ascribable to the Posidonia Marls Formation (Dogger)
(1430-1450 m), breccia with limestone clasts in clayey matrix,
attributable to the Cherty Limestone Formation (Lias) (1450-
1520 m) and very re-crystallized saccharoidal limestones re-
ferrable to the Massive Limestone Formation (Lower Lias) (1520-
1720 m). Three small syenitic-type dykes were also noted in the
last interval.
A syenitic sill encountered at 1720 m continues to 1940 m.
It shows various grades of crystallization, with fluidal struc-
tures on the edges and xenoliths of host rock. Absolute data-
tion with the K-Ar method gave a value of at least 0.5 m.y.
Thermometamorphosed sedimentary rocks appeared between 1940
and 2750 m, with frequent syenitic igneous dykes.
Metamorphism increases with depth. The mineralogical associa-
tion suggests that the original rock was prevalently made up
of massive limestone of the Tuscan Series. However, Triassic
limestones and perhaps anhydrites may represent the lower part
of the metamorphic series.
From 2750 m to bottom-hole there is a syenitic body of holo-
crystalline structure and equigranular texture.
The mineralizations and thermometamorphism are of particular
importance when defining the permeability characteristics of
the various formations.
The effects of thermometamorphism can already be seen in the
1520-1720 m interval, near the dykes, and especially on nearing
the sills between 1720 and 1940 m.
-270-
Newly formed minerals such as phlogopite. vesuvian and garnet
are associated with the re-crystallized limestone.
The rocks between 1940 and 2750 m are completely metamorpho-
sed. with mineralogical association of biotite. muscovite. gaE
net. vesuvianite and clinopyroxene.
Pyrite mineralizations can be found throughout the 25-1720 m
interval; silicification. chloritization and zeolitization pro-
cesses appear between 410 and 1280 m; calcium carbonates and
sulphates are found in the 410-1280 and 1380-1940 m intervals.
while calcite occurs in the 1380-1720 m interval and between
1940 and 2750 m. where it is associated with anhydrite. Clayey
minerals are found between 1520 and 1720 m.
5. THERMOMETRIC LOGS
During drilling eight thermometric measurements were made
near well-bottom in order to reconstruct the temperature pro-
file (Table I and Fig. 2. Log A). Temperature was measured paE
tly with a Kuster-type device and partly with a highly sensi-
tive thermo-probe designed and constructed by ENEL (EEC Con-
tract n. 166-76-EG-I).
The data were extrapolated to obtain the equilibrium tempe-
rature by a computer program.
-271-
6. IN-HOLE LOGS, MEASUREMENT AND ANALYSIS
-272 -
The first injection test was conducted with a bottomhole
8. GENERAL REMARKS
- 273-
......... N
...
c!J ~
o 2 3
, , km
~ \?-:::
MARTA
....
(m) C (OC)
100 200 300
0. , , • •
Alluvium
N
-
~... - .
-L. ~v
!, Pyroclastic rocks
\
>- \
r;
~
'"
~
:=::
-
I ~~ \
III -~ II::
It!
\
~b"!' -C-;~O
- L.custrine deposito
0( \
\
f6~'P9.~, ~
500 :__ ' Q._o" Z \
,;o.:.g'Q:_ \
~;O ,_ ()_
II::
o96°o'~'6 '
..,
~ ~b\f,b-'- III
\
OQ ' O':'O~ \
~
...
'. o'C:i' :'-o'
~ '90 Q.d :"·'O,·O \
" ~()'~bO~( Polygonic volcanic \
~ j;:)~ Q'c> . .,.'0 breccia 0(
\
~ D:09.d.O•
~O.:;so,~ :::I
iii 1000 ~DgC)&,\,
iii ~qo~. \
o-QO~\/Q a \
'tl 0 '
~?bo ,:) \
,
_ ~ '-C7o
" o/ljo·~o @ \®
~'i:).dt\;":
___eF:;(~"-.. ..3:
", .
\
..
-'
~ • (f>vii Allochthonou. terrains
Radiolarit•• :I
\
'90 Posidonia mar I ...J
e \
1500 I~ Chorty lim.ston•• :I \
I
A. }/L \
~ ).... Massive IimestoM' ~.
...J
\
~ y'Z ...i
,
iii
iii v v
'"
" v
"
"v vv S,enitic Sill \
" v '" v (~0_5 M. y)
v
V v ... \
W-.....L-
\
-L.
2000
~
-:.,..
\
r--'- -'-- \
t--C-~ \
p:-=
-1- -'-
\
p:= -'- Thermomot.morphic
S' ~- -~ s.q ....nc.
... '\
\
ai ·
iii
2500 ~~ .\
.-L ......l.-
~ _J \
.-L ---L-
-L J.
\
L- L-
Srenit.
\
- 275-
and partly to the relatively cold aquifers circulating in the
- 276-
OPERATION GEOTHERMIE A STRASBOURG-CRONENBOURG
-277-
A l'origine (1974) l'exploitation de la Geothermie dans l'agglomera-
tion strasbourgeoise a ete envisagee dans plusieurs secteurs
(Illkirch - Graffenstaden, Elsau, etc ••• ) avant que les perspectives
economiques de ses debouches dans le domaine du chauffage urbain con-
duisent a retenir, pour une premiere experimentation, le site de
CRONENBOURG.
1 - Cam~agne geophysique (vibrosismigue)
Toute la zone a forte densite de population pouvant ~tre interessee
par la Geothermie correspond a des formations sedimentaires en posi-
tion synclinale et, de ce fait, avait ete evitee par les campagnes
geophysiques a objectif petrolier. Mais celles effectuees alentour
y faisaient presager le passage d'un reseau de failles normales
fortement inclinees et de rejets importants. Les emplacements dis-
ponibles pour des forages, compte tenu des problemes d'environne-
ment, etant en nombre tres limite et l'exploitation des doublets de
puits geothermiques devant @tre conduite au niveau du reservoir
dans un bloc parfaitement homogene dans ses trois dimensions, il est
apparu d'emblee un probleme de comptabilite des operations avec les
accidents tectoniques supposes. D'ou la necessite de definir parfai-
tement ces derniers, jusqu'au Socle, c'est-a-dire a une profondeur
importante (plus de 3000 m). L'analogie entre ce probleme et ceux
de l'exploration petroliere, et la realisation d'une nouvelle
campagne a buts petroliers a proximite de Strasbourg ont conduit
ELF-AQUITAINE a mettre en oeuvre, en 1975 et pour la premiere fois
en Geothermie, une equipe vibrosismique dans toute la banlieue de
Strasbourg. La methode vibrosismique, completement affranchie de
l'emploi d'explosifs a permis de travailler dans les voies urbaines
tant que les conduites anciennes du reseau de distribution de gaz le
permettaient. Dans certains secteurs la progression a ete effectuee
en slalom (Ie passage des enregistrements en central informatique
restituant un profil rectiligne) afin d'eviter d'eventuels deg!ts.
Les quatre profils pour Geothermie ont presque tous prolonge des
profils "petroliers", beneficiant ainsi directement des tirs de
petite refraction, pour mesures de vitesse, de propagation qui ne
pouvaient ~tre effectues en zone urbaine et d'un calage sur sonda-
ges pro fonds. En neuf jours de travail 32 kms de profils ont ete
realises en couverture d'ordre 24. Au moment de son execution, Ie
- 278-
co~t de la campagne vibrosismique a Strasbourg representait environ
2,7 %du co~t d'un doublet de puits geothermiques.
2 - Sondage de reconnaissance
Le sondage qui doit @tre fore, verticalement, tres prochainement a
CRONENBOURG a pour but de reconnaitre le potentiel geothermique de
la nappe aquifere des Gres Bigarres a un emplacement qui tient
compte a la fots, de la proximite d'un centre important de consomma-
tion d'energie pour le chauffage des habitations, des possibilites
d'installation d'un appareil de forage de type lourd et du comparti-
mentage des reservoirs geologiques en profondeur Mis en evidence
par la campagne geophysique rappelee ci-dessus.
II est destine - si le potentiel geothermique Mis en evidence
apparait exploitable economiquement - au soutirage de l'eau de gi-
sement des Gres Bigarres (Trias) entre (previsions) 2550 m, toit de
la formation, et 2900 m, profondeur finale dans le Socle.
Dans un premier temps, il est prevu d'evaluer Ie potentiel geother-
mique des Gres Bigarres, le sondage etant laisse en trou ouvert sur
toute la hauteur de la formationdetritique. Par la suite, il est
possible que l'on selectionne (bouchons de ciment et/ou perforations
en colonne) dans ce puits les meilleurs horizons producteurs -
appartenant soit aux Gres Bigarres ou aux formations Muschelkalk,
Lettenkohle ou Grande Oolithe - en vue d'obtenir un rendement supe-
rieur. Auquel cas Ie puits de reinjection sera fore et conditionne
pour interesser les m@mes niveaux.
L'objectif est de disposer en t@te de puits d'un potentiel geother-
mique equivalent a 150 m3/h d'eau (mineralisee) a 140 °C.
3 - Implantation du sondage
Suite a la cartographie precise des accidents locaux et de trois
horizons sismiques fournie par des interpretations des pro fils sis-
miques pour Geothermie et de profils petroliers recents on peut
souligner que, dans Ie secteur de Cronenbourg, Ie trait geologique
majeur est constitue par une faille normale de direction SO-HE in-
clinee a 45 0 avec plongement Ouest et de plus de 400 m de rejet
qui decoupe deux compartiments:
- un compartiment Ouest, Ie plus bas structuralement, affecte par
un axe synclinal SSO-NNE, Ie long de la faille, et qui plonge
- 279-
regulierement vers l'Est. La profondeur du Socle y est voisine de
-3400 m (par rapport au niveau de la mer),
- un compartiment central d'allure convexe, c'est-a.-dire marque par
un axe anticlinal dans sa partie occidentale, qui s'enfonce regu-
lierement vers Ie Nord-Est et dans lequel Ie reservoir objectif est
engage.
Le sondage est implante sur Ie profil sismique Geothermie N° 4.
Celui-ci, tres court parce qu'interrompu a. proximite de l'Autoroute
A 35, ne fournit un bon calage stratigraphique que dans Ie compar-
timent Ouest "au-dessus" de la faille. Pour ce qui concerne Ie
compartiment Est "sous la faille", dans lequel les Gres Bigarres
objectifs sont engages, on formule actuellement deux interpretations
structurales : avec ou sans reduction de la tranche "toit du Sannoi-
sien inferieur -toit de la Grande Oolithe".
- 280-
~ 20" 150 m pour isoler la nappe phreatique pendant l'approfon-
dissement du puits
~ 13"3/8 1200 m pour recevoir les equipements de production assis-
tee (lift ou pompe immergee)
~ 9"5/8, en colonne perdue de 1100 a 1850 m, pour couvrir les forma-
tions au-dessus de la serie Secondaire dans laquelle deux
reservoirs (Grande Oolithe et Muschelkalk superieur) seront
etudUs.
~ 7", en colonne perdue de 1750 a 2550 m, pour couvrir les forma-
tions au-dessus du reservoir objectif et assurer sa mise
en production en trou ouvert sur toute sa hauteur si
necessaire.
La composition des boues de forage tiendra compte de la non pollu-
tion de la nappe phreatique, de la presence d'anhydrite et,
eventuellement de sel dans Ie Sannoisien, des pertes possibles dans
la Grande Oolithe et Ie Muschelkalk superieur, et du non colmatage
(cf. fraction argileuse) du reservoir objectif.
Des essais de courte duree sont prevus sur chacun des trois reser-
voirs d'inter&t geothermique :
- essai de la Grande Oolithe sur toute sa hauteur, avec garniture
ancres dans Ie tubage ~ 9"5/8
essai du Muschelkalk superieur sur 25 m environ, avec garniture
ancree dans la partie superieure de la formation
- essai global des Gres Bigarres, avec garniture ancree dans Ie
tubage ~ 7".
Un programme complementaire sera etabli en fonction des resultats
de l'essai des Gres Bigarres et de la tenue du decouvert, avant de
proceder a des essais de longue duree interessant les niveaux rete-
nus pour la mise en exploitation.
Des carottes seront prelevees dans Ie reservoir objectif Gres Bi-
garres, si possible dans les niveaux les plus permeables et leurs
epontes.
Les diagraphies differees seront ajustees de fa~on a realiser une
parfaite investigation des qualites reservoir de l'objectif et de la
serie Secondaire sus-jacente. II serait souhaitable quelles puissent
-281-
COUPE TECHNIQUE Coordonn~es appr o Appareil
X : 997.090
de Forage
SONDAGE : CRONEN BOURG V :1 114.275
Geofherm ie GCR 1 Zs: U3 m
CH 1400· e 23
D.S.T
i
on ___
--_ -_ " )
., f ~ ~. : --.--- U
Co CD '! .=-~.-~.
j
~ ~
_!
..
.,... u
i
~
g! ----
N • --
~.!
., ,'1,.000 ~:E~~u;:;Jr.;,ll.:I'.
_ _ ~ .. ':':,~\p r v::t •• '".,
1200 __ A~3 i~
_----=-_
- - z
z
~
~·
ISo
.! ..!
---A ~ 1 .; ~
c ..
.! .:
-;
:
~
1850..r
i
i
~
t-
•
!:10 0-t
'" ~ "'5
.. ~ i !
1 ~ '!
iI !: ••..~ '. It...!
~
~ .~ ~
'" 0
c ..g
f'
~ v
• 1i • i ~
}......
~ i t-
o
~
..J
a r
,L
·.
..! ;. ,-,0
'" .J _
t .H
11<
I 2
-
f! ! !
I--l--l--L ~
...o ...'"
.J
c l -
·.. .
.J .. .J
I o •• ~ "-
I
I ~l ~ ~ %
...'" ...
~ :I
.J
"&
i ~ I-L--'- -'-'-
.
o
·
co ~
...
>
.:: Vi o
o
Observations! Inultfltud., lour 10 p,c .. ne~ III I'';po fu euf 4 . . . lilt de Cr"ndtl Ootfthe- eon rodl j l.M du ,onoilt d' une foill ..
we' .. laSO m.
- 282-
s'accompagner de toutes les mesures necessaires de pendagemetrie
afin de localiser le plus precisement possible l'accident majeur
et les complications techniques qui pourraient l'accompagner.
On doit souligner en effet que si necessaire - et c'est une alter-
native a une serie de competitions et de stimulations - une plus
forte production des reservoirs pourrait @tre recherchee, a l'instar
de ce qui est etudie de l'autre c&te du Rhin par nos collegues
allemands, en procedant a une reprise partielle du sondage en
deviation dirigee de fayon a se rapprocher de la zone affectee
directement par la faille.
- 283-
DEMONSTRATION PROJECT
MILANO
by C. SOMMARUGA, R. GHELARDONI
AGIP S.p.A.
ABSTRACT
- 284-
1. INTRODUCTION
Well Metanopoli 1, drilled in 1979, the results of which are described herein,
was located on the SW flank of a trough in the "Ghiaie di Sergnano" Formation and
is part of a programme pertaining to a first experimental low enthalpy geothermal
space heating project which will be carried out in I taly through a couple of deep
wells by AGlP and SNAM, companies of tl'\e ENl Group (EEC Contract GE 02/79).
- 285-
The project covers winter heating by hot geothermal water and summer cooling
of dwellings and offices (ENI Group's offices, Metanopoli) for a total space of
467,000 m3 , at present heated through a conventional centralized thermal system,
and equivalent to 2350 dwellings of a standard volume of 200 m3
The first phase of the project includes the drilling of a couple of wells, one
producer and the other for reinjection. Bott. wells, more than 2000 m deep, have
the same location but the second well will be directional with a lateral deviation at
bottom hole of about 1000 m. The expected cooling of the formation is 1°C in about
50 years.
The target of well Metanopoli 1 was to ascertain the reservoir lithologic
characteristics and fluid temperature, salt content and potential flow rate.
The second well will be drilled after the hydrothermal and hydrodynamic para-
meters obtained from the first well have been processed so as to allow preparation
of a hydrogeologic model of the aquifer.
At this stage, the risks pertaining to the geothermal system conditions at a 2000
m depth are connected with:
- aquifer porosity and permeability;
- water temperature;
- aquifer thickness;
- water salt content.
The second phase of the project contemplates the utilization of 100 m3/hour of
58°C water for direct or indirect heating, even through the use of a heat pump.
The net saving of primary energy achievable in a year is:estimated at approxima-
tely 1500 t.e.p. taking into account the heat pump and facilities consumption.
In particular, this project contemplates the heating of a first residential quar,ter
through heating panels, of offices and mess hall through hot air, and of a second
residential quarter through a natural gas powered heat pump which will be used for
office cooling in summer.
It is estimated that the project. ,,,ill meet 87% of the thermal energy requirements
and 47% of the power peak required.
- 286-
3. WELL METANOPOLl 1
The well was drilled using a Cardwell 03 rig from August 10 to September 20,
1979; height of rotary table 100m a. s .1., total depth 2505 m.
The effective porosity analyses of the tested beds were carried out using the
following electric logs:
ISF - SL 1967 - 2310 m
FDC - CNL 1967 - 2311 m
ISF - SL 2299 - 2505 m
-287 -
1967 - 2039 m interval: average porosity between 4.5% and 25%
2102 - 2144 m interval: average porosity 17.7%
2181 - 2293 m interval: average porosity between 4% and 12.96%
2304 - 2424 m interval: average porosity between 5.2% and 11.6%.
The chemical analyses of the reservoir water indicated a total salt content of
88.5 gil with prevaling NaCl and scarce Ca and Mg, sulphates were practically ab-
sent. The water contained dissolved gas in the porportion of 0.5 Nm 3 per m3 of
water.
The temperature recorded by Amerada technique during static condition of the
fluids indicated 60°C at reservoir top and 75°C at bottom. However, during fluid
dynamic condition the temperature rose to more than 81°C at reservoir bottom.
Production parameters obtained from production tests in the upper section of
the reservoir were as follows:
The results of the first well confirmed the anticipated outlines down to the
reservoir top (1967 m), but the situation downwards was marked by the following
differences:
a) Smaller reservoir, divided into thin sand and shale beds belonging not only to the
"Ghiaie di Sergnano" Fm but also to the underlying shale-marl beds, all Upper
Miocene. The lithology represents a lateral variation of a conglomerate complex
belonging to the "Ghiaie di Sergnano" Fm, penetrated through 150 m by a well
previously drilled on a similar situation a few kilometers South of Metanopoli.
b) Lower permeability affecting adversely the water flow rate. Flowing tests made
by air lift indicated a maximum flow rate of 60 m3 Ih, therefore, lower than
expected.
- 288-
c) Higher salt content, up to 88.5 gil, which could cause scaling problems in the
water lines.
Based on the phsycal data obtained by the air lift technique and on formation
data from core analyse'S and electric logs, a preliminary mathematical model of
the hydrogeological system was prepared. The results confirmed theoretically
the possibility of using submersible pumps with 400-500 m head and 40-50 m3/h
output. Such submersible pumps should be at least 150 kW and suitable to operate
in 80 D e temperature. Equal power (150 kW) is required for reinjection.
The solution of the problmes which emerged in drilling well Metanopoli 1, par-
ticularly those concerning production of gas-out geothermal waters, is of the
utmost importance for the implementation of similar projects in the Po Valley.
New projects are being prepared to be carried out on the outskirts of Milan,
where gas wells drilled by AGIP have already revealed a higher thickness of the
"Ghiaie di Sergnano" formation.
-289 -
EIANDPOLI P lEel
•u
- 500
.
- - - ~
..
- - - - - f- 1000
- - - 't
g
II'
-- I-:::~
~i
-----
- --
-=-=- f.-i
--
___
- -
4:
!-
;~
..g i \I
-~ li
..
\I '
- - - ~
~
i
-
-.
- -
;.-:"'.......... ;.~
. ~.;-A'7
~
X
.'... ,\
,
i
\
~
2 " 2 I I
60' 7f1
- 290-
DEMONSTRA nON PROJECT:
EVALUATION OF THE MOFETE HIGH ENTHALPY RESERVOIR
(PHLEGREAN FIELDS)
Abstract
Mofete 1 and 2 wells' drilling is the first phase of a geothermal evaluation study
of Mofete field, the final target of which is the installation of a modular plant for
electric energy production of at least 10 MW e .
The whole project includes also drilling of deviated wells and reinjection wells
from a few location plats in highly urbanized country. Drilling -crossed volcanic
grounds consist mainly of pyroclastic and lavic volcanites i1wl of sedimentary
layers.
A few mineralized intervals were found at a depth ranging from 500 m to about
2000 m with a fluid temperature between 200°C and 350°C, which are being tested
a serie of several analyses.
The first gathered data show that drilled wells ascertained a pressured hot
water, with high salinity (30 gr/l) and a possible vaporization close to the boreho-
les.
The achieved results show the possibility of ,developping a program for deeper
drilling, down to 3000 m. to recover fluids at higher temperature.
- 291 -
1. INTRODUCTION
Between 1939-1953 SAFEN carried out a drilling exploration in the Phlegrean
Fields and, expecially in Mofete area, drilled about 15 wells a few-tens to
1200 meters deep. These wells showed the presence of a water dominated system
with high temperature (225 0 C) and high salinity (about 30" 000 p. p. m.).
Due to the lacking of adeguate technologies at that time, the potentiality and
physical-chemical parameters of the geothermal fluids were not completely defined.
The CVL 7 well produced water and steam (max 40 t/h) but after four months the
outflow stopped due to the progressive self-sealing.
Notwithstanding these first results, in 1953 SAFEN ceased any exploration
activity.
Later on, in the years 1977-1978, the activity was resumed by the joint ven-
ture AGIP-ENEL (operator: AGIP) which carried out the geological, geochemical
and geophysical investigations necessary for evaluating the geothermal field of
Mofete with the aim of producing electric power.
Because of the highly urbanized area, particuiar care was paid to the protec...:
tion of the environment with controls and surveys carried out before, during and
after the drilling. Natural microseismics controls,were, infact, carried out, or are
still being carried out, by means of a geophone grid, altimetric, acoustic surveys,
as well as gas diffusion survey in the atmosphere during production tests.
The target of Mofete 1 and 2 wells was to explore a lithological section of
Mofete area, at least, 1500 m thick,and to analyse its porosity and permeability
characteristics; as well as ascertain the existence of geothermal fiJilds deeper
than those discovered by SAFEN at 350-600 m and to control its physical and
chemical properties.
The results achieved with the first two wells, drilled in 1978-79, have alrea-
dy enabled the Mofete 3 well to be drilled, as scheduled in the second phase of
the project.
This phase (1979-81) includes a few development and reinjection wells. The
latter ones could be necessary in order to avoid the possible pollution and re-
charge the reservoir.
-292 -
In this phase the installation of an experimental 3 MWe power plant is planned.
During the last part of the project (1983-84), a first modular plant of 10-15
MW_ could be installed.
e
The success of the project might encourag.ethe exploration of geothermal
systems in analogous geological situations of Southern I taly and other active volca-
nic area dependent on the European Community.
The Phlegrean Fields, to which the Mofete hill belongs, are a quaternary
alkaline-potassic volcanic area that covers a wide sector of the Campanian Plain.
The last one is a large graben in which the mesozoic carbonatic rocks are pro-
gressively sinking from the Appennines (NE) towards the sea (S W).
The volcanic products of the Phlegrean Fields are related to three principal
periods of activity which go from 70, 000 years ago to historical times.
The Mofete hill is part of a volcanic system of the second phlegrean period.
This volcano was partially destroyed by successive volcanic and tectonic activities
and is partially covered by the third-phlegrean period products. The Mofete area
is a fault-block higher than the other surrounding sectors.
The whole phlegrean region is affected by numerous normal faults with trend
NW and NE and throws of some hundred meters.
The geological and geophysical survey - particularly the gravity, airborne
magnetic and geoelectric surveys that have been specially carried out - show that
the Phlegrean Fields, with their volcanic structures and local calderas, belong
to the margin of the vast sinking caldera corresponding to the Pozzuoli gulf.
This caldera is characterized by a large negative gravimetric anomaly with
positive local anomalies on the margins ~Annular and radial fractures favoure the
rising of magmas, hydrothermal fluids and, consequently, the isotherms.
The Mofete area, on the W margin of the great caldera, is characterized in
particular by a low magnetic susceptibility which has to be related to the predomi-
nance of toufaceaus grounds and conspicous hydrothermal-alteration phenomena.
The h.ydrogeochemical study recognised three kinds of waters in the Phlegrean
- 293-
Fields: bicarbonate-alkaline-calcium-magnesium waters (representative of the
H/CH 4;
- Area surrounding Averno lake, with shows of dominant nitrogene and low ratio
H/CH 4 ,
The differences could be due to the fact that in the area surrounding Averno
lake an effective cap-rock should limit the rising of CO 2 towards the surfa-
ce,
In the area there are several water sampling points with anomalous-temperat~
re values; in fact,the medium gradient of the zone is five times the normal one,
The thermic semiquantitative balance seems to confirm the existence of a wide
high-geothermal gradient area, particularly on the west margin of Pozzuoli gulf,
where the Mofete wells are located,
From what stated above it is possible to say that the phlegrean geothermal
system includes a series of hot acquifers fed by other hotter and deeper aquifers may.
be corresponding to a carbonatic reservoir,
- 294-
3.1 Lithostratigraphy
- subaerial pumiceous tuff of the second and third phlegrean period. Thicknes.S:
180-3S0 m;
- chaotic tufite formed by tufaceous conglomerates more or less coherent with
lithification increasing toward the lower part. A'rgillii:atibn phenomena are ob-
served. Thickness: _6SO-7S0 m;
-lathitic and trachitic lavas with tufite and tuff alternation. Thickness: 2S0-300 m;
- subaerial and marine volcano-sedimentary sequence. formed by chaotic tufite.
marly-silty-sandy thin beds. tuff and some lathitic lavas. Thickness: 100-200 m;
- marine chaotic tuff and alternate lathitic lavas. Thickness: about 3S0 m.
From a general point of view the argillization and chlorifizationphenomena.
connected with the hydrothermal alteration of volcanic rocks. noticed in the first
600-900 m. have given a cap-rock character to this interval. The rocks of the
deepest interval are somehow hardened by the deposition of silica, epidote, adula-
ria and carbonates. Consequently these rocks have acquired a fracturing chara-
cteristic and, therefore. the possibility of becoming reservoirs.
The hydrothermal paragenesis made possible the spotting of the following zo-
nes:
- argillitic zone, down to 300 m, with prevalent montmorillonite. Temperature:
130-1S0·C;
- phyllitic zone, from 300 to SOO/600 m. with prevalent chlorite and sericite. Max
temperature: 180·C;
- phyllitic-propilitic zone, from SOO/600 m to 8S0/1000 m. with chlorite. adularia,
illite and epidote. Temperature: 200-2S0·C;
- propilitic-potassic zone, from 8S0/1000 m to at least 2000 m, with adularia, al-
bite, epidote, quartz and pyrrotine. Temperature: over 300·C.
- 295-
der to give a prompt definition of the lithological and mineralogical aspects of the
crossed rocks, petrographic analysis of cuttings and cores on the site are carried
out.
- <!.~o.£~~mj£'!La~~i.2!.aE£~ mud samples, mud filtrates, reservoir fluids (gas and
water) are examined with the scope of obtaining information on the parameters of
porosity, permeability and fracturing, temperature and physical state of the geo-
thermal fluids in reservoir rocks.
Such an assistence is carried out by means of a mobile laboratory, which is
owned by AGIP and is an innovation in the geothermal-drilling sector thanks to the
completion of activites it can perform. I t is handled, under the scientific manag-
ment of Prof. Tonani, by AQUATER, a rompany which is part of the ENI Group.
The described methods are very useful in order to detach tractured intervals
which, in rocks with almost no primary pe.rmeability like those crossed by Mofete
wells, are the only place where geothermal fluids can gather.
The use of light weight mud in the geothermal exploration in the Mofete field
was suggested to avoid that the reservoir is damaged during the drilling. On the
other hand, light mud use makes fractures'detection more difficult and consequen-
tly the recognition of reservoirs becomes one of the most important problems that
have to be solved.
AG I P technicians are on the way to get the right answer using different re-
sponses: geochemical log, injectivity tests, geophysical logs and particularly aco-
ustic attenuation interpretation.
The enclosed paper shows the reliability of responses obtained from the stu-
dy of compressional and shear-wave attenuation, geochemical log data and injecti-
vity.test results (fig. 4).
3.3 Fluids
In this first phase, with the drilling of Mofete 1 and 2 wells it was possible to
assess the presence of high enthalpy fluids in deep intervals investigated by former
exploration.
-296 -
During the initial production tests a remarkable series of water, steam and
4. CONCLUSIONS
Since the results of Mofete 1 and 2 proved to be successful and in order to
explore a major part of the geothermal system, a third deviated well was carried
out (MF 3d - TO m 1909 - vertical projection: m 1744 - offset: m 634 towards SW).
The wells' production tests, made with particular care in order not to ~amage
the environment, are still being carried out and, therefore, we do not have enough
information on the geothermal model, on its chemical and physical {:haracteristics
and the potentiality of the wells and fields.
The Mofete geothermal system is a hydrostatic-pressure hot-water single-
phase-system, with .Scarce natural shows, low-medium permeability and fractu-
ring zones. The temperature increases with the increasing of the depth.
As for the dynamic characteristics of production, the system seems to evolve
in two-phase with in- situ vaporization around the wells.
If the reinjection is suitably measured, it will be possible that the system
evolves even to produce dry steam.
In fact, the first tests carried out on wells which were not stabilized yet, sho-
wed excess of steam in the water (90% in volume and 50% in weight) with fall of
pressure at the well bottom up to a few tens of atmospheres .The fluid temperature
varies from 350°C in static conditions to around 200°C in dynamic conditions as
the vaporization heat is given by the water itself.
The Mofete field has, therefore, a considerable scientific interest as it
gives the rare opportunity of studying the initial conditions and the evolution pro-
cess from a geothermal water system.:into a steam system.
From the first evaluations it seems possible to obtain potentials of MW /
2~5-3
e
well. However, since the number of the wells that can be drilled in that particular
territory is limited, it is possible to plan the installation of a plant of just a few-
tens MW total capacity. But it is likely that the field stretches towards North,
e
- 297-
£uther the area presently investigated. At this end a detailed geological and geo-
physical survey is being carried out with the aim of locating an outpost.
In the test campaign in course, measurements are made by means of lips, ac-
cording to James method in order to evaluate the production of the first three wells.
In future separators will be implemented too.
Later on, another test campaign will be carried out with the aim of evaluating
fluid-dynamic characteristics and the geometric extent of the reservoir (interfere!!.
ce,limit and pulse tests).
Finally,prolonged injection tests will be made in one of the wells while the
other two will be alternatively the producer well and the "control" well.
It is likely that during those tests electric power will be produced by means
MF2
o
MF3 0
OMF1
f iG I
- 298-
1- -
MOFETE 1 WELL
(E.p · LAGO 01 PATRIA · )
SnATIGaAftHY CASINGS 51ATlC TlM'lIATUR. ('C I HYPOTHITICAl COM,OSlTl I'tIOFlLl
0, ' IGO' 200' 300· 400· If' lGO' 2GO' 300' 400· °c
m 0, I I I
I
(;
(;
(;
<D APR 1 79
°
(;
(;
(;
l
® MAY 4 ,.
®
'*~\\:
(; (; MAY 15 79
(;
@ SEPT 13 '.
'~ "
(;.(;
;(; .
1(; =(;
-~ =(;; ®4'~
"
+ '00
(; =c. '~ .
'I )
~I \-----0
:1 z: '"'"
M
\~ (.
\
1000 ~ ~ ~ h
II
+ 1000
[, (; 1 ~m
\\
II'
'h
'\\::
II
1500
16001~ .~
(;(;
U~
'. 0'
r oo 1600
TO 1606 m
16 61 pomICMi.l5 tun D l"llhc r.i~a
16761 (.11.011(.1",,1,10 B 'loed'rT'If>"'I6r't~III1.l~'llltetl
I-.K.-I tf.tCnytlC I......
---- - - ------- Fig 2
,u
M
! Ol
iL
i•
I~
i ,
I
,.
of
t
~
~~t~~~
...I"
,.,
:!
:i
•
:; G e .....
...I
11111:
S Ifflii
~
:... I!! e®®ee®
i
,
i!5
..E
C\I It
~I
~
0.,
...
2~
I.
I I
o
<J
II
<l
II
<]11
jt
<l • <l
<l ..
<l " <l
II
<311
II
<l
<l
II
<]1.
II
<311
<J II
- 300-
MOFETE 2 WELL (SKETCH)
1250
.., 1272
1275
\
due \
\
I _ FI9 __4_ _ ~
~
:zoo'
- 301-
have not been met yet, and can cause new problems and the development of new
technologies.
I n this respect ,AG I P is planning a series of laboratory experiments with
samples of rocks and fluids placed in autoclaves under static and dynamic conditions
Bibliography
APR I LE F., OR TOLAN IF., 1979 - Sulla struttura profonda della Piana Campa-
na - I stit. Geolog. Geof. Univer. Napoli.
BARBERI F., INNOCENTI F., LUONGO G., NUNZIATA C., RAPOLLA A.,
1977. Shallow magmatic reservoirs as heat source of geothermal systems;
preliminary interpretation of data available for the neapolitan active volca-
nic areas~ CEE Seminar on Geoth. Energy - Decem. 1977.
CIVITA M., DE MEDICI G.B., DE RISO R., NICOTERA P., NOTA D'ELO-
GIO E., 1973.
Memoria descrittiva della carta idrogeologica della Campania Nord-Occiden
tale-II Conv. Intern. Acqua Sott. Palermo.
FINETTI 1., MORELLI C., 1974 - Esplorazione sismica a riflessione dei Golfi
di Napoli e Pozzuoli-Boll. Geof. Tecn. Appl. XVI 175-222.
MUSMANN G., HUNSCHE U., DUS KE J., KER TZ W., 1977 - Magnetotelluric
and geoelectric measurements for geothermal esploration in the Phlegrean
Fields (preliminary results) -CEE, Seminar on Geoth. Energy, Dec. 1977.
rie nei Campi Flegrei per vapore, acque termali e forze endogene in genera-
Ie - Ann. Geof. n. 4.
Abstract
Sasso 22 well was programmed in the Larderello field (Italy)with
the objective of verifying whether geothermal fluids could be recovered
from the deep horizons underlying those producing at present; other
objectives were to obtain better estimates of deep energy reserves and
gather information on their exploitation potential. Furthermore, data
from this deep well should improve the phenomenological model of the
field. Further information has been obtained on the stratigraphy of
the deep layers of the crystalline basement; below 2500 m the well pene-
trated gneiss and amphibolite, with a gradually increasing metamorphism.
In-hole logs revealed the presence of many fractured horizons within
the basement down to well-bottom and increasing temperature values to
around 380°C at 3970 m. Study of the parageneses indicates the exist-
ence of an initial hydrothermal phase at temperatures well above those
measured at present and also of fluid circulations in unexploited sec-
tions of the basement as deep as 3-4 km.
- 303-
1. INI'RODUCI'ION
Drilling of Sasso 22 well was partly financed by the BEe and repre-
sents part of the research programme directed at exploring the deep geo-
logical structures in Larderello geothermal area. The main objectives of
this research are: to ascertain whether geothermal fluids can be recovered
from horizons underlying those producing at present; to define the charac-
teristics of the deep reflecting horizon individuated by the seismic re-
flection survey: this horizon exists throughout the area at depths ranging
between 3500 and 7CXX> m; to study whether the field could be recharged ar-
tificially by re-injecting water into different levels of the reservoir;
to improve the field model and acquire the necessary data for a better
evaluation of its potential.
The well reached a total depth of 4092 m, which is the deepest so far
in any of the exploited geothermal fields throughout the world.
2. DRILLING
Nearly all of the well was drilled without return circulation because
of the large number of unseal able fractured horizons.
The formations encountered during drilling were very hard and unhomo-
geneous.
The lack of return circulation forced us to use water as drilling
fluid,which meant less cuttings could be removed. We were consequently ob-
liged to use a less stiffened drilling string to avoid blocking the bit
more than necessary.
As a result it was practically impossible to achieve a straight well.
Despite the above-mentioned difficulties drilling operations proceeded
successfully to a depth of about 3CXX> m, after which the problems increas-
ed continuously to 4092 m.
At this point same unusual drilling problems arose with the breakage
of the drill-pipes, probably caused by the high temperature conditions in
the bore and its corrosive environment.
Fishing operations had to be undertaken before continuing drilling.
These proved to be an extremely complicated feat because of the high tem-
- 304-
perature conditions, with bottom-hole values ranging around 400°C. Thus
explosive -type fishing operations could not be used, nor hydraulic equip-
ment (e.g.,back-off, jars, bumper sub, turbine).
Directional drilling had some success, despite the fact that, because
of the prohibitive high temperatures, the appropriate cement plugs could
not be placed exactly as desired at well-bottom to center the directional
equipment. The difficulties increased at such a rate that drilling had
eventually to be abandoned.
However, the attempts later made to overcome these difficulties by
using commercial equipment adapted to the particular conditions in this
well have represented a useful experience for future deep drilling opera-
tions.
3. IN-HOLE LOGS
Systematic in-hole measurements were programmed for evaluating the
thermodynamic parameters of the fluids and the lithological and physical
characteristics of the rocks. However, tkis programme could not be respec-
ted in full because of the technical problems.
During drilling some temperature and pressure measurements were made,
along with production and injection tests and a series of in-hole geophys-
ical logs (Fig.l).
The geophysical logs and absorption tests were conducted when the well
reached 3000 m, in the interval between 1500 m and bottom-hole.Comparative
study of the sonic,electric and Y-ray logs individuated lithological va-
riations and fractured zones. The latter were also revealed by the absorp-
tion profile drawn from the flow-meter log.
Fracturation is spread throughout the tested interval and especially
between 1500 and 1900 m and 2400 and 3000m. Fractured horizons also exist
below 3000 m. Measurements of temperature at bottom-hole were also made
during drilling stops, obtaining values of 255°C at 1320 m, 270°C at 19l7m
and around 380°C at 3970 m.
-305-
4. STRATIGRAPHY AND PETROLOGY
The sequence of formations in the Larderello geothermal area is as
follows (1), from top to bottom: a) a sequence of mainly clayey and shaley
units that act as the cap-rock for the geothermal system. These consist of
i) Neoautochthon units of the Mio-Pliocene and ii) flysch-type allochthon-
ous units of Jurassic to Eocene age; b) a sequence of formations that rep-
resent the potential geothermal reservoir, made up of i) the arenaceous
and carbonate Tuscan Nappe, ranging in age between the Upper Trias and Oli-
gocene and ii) the underlying Triassic and Paleozoic terrigenous forma-
tions, that represent a discontinuous layer of tectonic wedges; c) a
slightly permeable metamorphic basement of Paleozoic age.So far few wells
have penetrated this formation and only to depths of 2000-2500 m. Its main
lithotypes are phyllite,quartzite,metagreywacke (Ordovician-Silurian) and,
at greater depths, micaschist and amphibolite.
Apart from defining the stratigraphy of the upper part of the base-
ment, Sasso 22 well also reached metamorphic units that were unknown until
now in this region.
The stratigraphic column is described briefly below (Fig.l). From top
to bottom the following units have been encountered:
1) 0 - 190 m: an Eocenic Ligurian Nappe, which is a flysch sequence of
clayey,shaley and sandy sediments.
2) 190-270 m: Oligocenic arkosic sandstone ('macigno') of the Tuscan Nappe.
3) 270-600 m: Mesozoic carbonate formations (Upper Trias-Lias) of the Tus-
can Nappe (270-380: stratified cherty limestone; 380-530: massive lime-
stone; 530-600 stratified marly-dolostone).
4) 600-1325 m: Upper Triassic carbonate and anhydrite ('Burano Formation')
representing the base of the Tuscan Nappe.
5) 1325- 1440 m: Triassic slightly metamorphic quartz pebble conglomerate
and coarse quartz-arenite which constitute the basal levels of the Tuscan
Verrucano.
6) 1440-2315: quartzite,phyllite,metagreywacke,metabasite of the Lower Pa-
leozoic 'Filladi inferiori' Group.
- 306-
7) 2315-2375 m: garnet-bearing plagioclase micaschist of the Micaschist
Group (l)_(Lower Paleozoic-Pre-Cambrian?).
8) 2375-4092 m: fine-grained gneisses and thin interbedded amphibolite le-
vels, (Lower Paleozoic-Pre-Cambrian?).
The latter is the most interesting unit, as it has never been found
before in the crystalline basement at Larderello-Travale. The presence of
gneisses and amphibolites confirms the analogy between the Tuscan basement
and that of North Sardinia (2), Calabria and north-east Sicily (3), the
Alps (4) and Provence (S), already pointed out by some of the present au-
thors (6).
Another result of the petrological studies on the Sasso 22 core samp-
les is the individuation of two main Hercynian metamorphic events: a first
phase of medium pressure and medium grade and a second phase of HT-LP (Ab-
ukuma type) (7) .
The micaschists and gneisses from the core samples taken between 2369
m below ground-level and 2636 m contain the following minerals belonging
to the second phase: andalusite, cordierite,newly-formed biotite, besides
those of the first phase: quartz,muscovite,chlorite,kinked biotite,andesi-
nic plagioclase, almandine, with tourmaline ,apatite ,zircon and ilmenite as
accessories. Andalusite and cordierite referred to a late Hercynian HT-LP
metamorphism of Abukuma type indicate the attainment of the medium grade
(T ~ SOO°C (8)); Hercynian metamorphism during the first phase also reach-
ed the medium grade, as indicated by the anortite content constantly ~30%
- 307-
3686 m and 3800 m. The corresponding P""T conditions seem to have been
reached during a late Hercynian thennal event; indeed, metamorphic rocks
of similar Hf-LP to those of Sasso 22 well are widespread and very thick
arotmd the Herycnian granites of Sardinia and Calabria.
All these petrological data provide new knowledge on the deep struc-
ture of the Larderello-Travale area with regard to the nature of the deep
seismic reflecting horizon. Of all the previous hypotheses (10),(11), the
most convincing is to assume the existence of a granitic-gneissic Hercynian
basement.
The existence of an Alpine thennal event is shown by the appearance
of calc-silicate bands (core samples of 1600 m, 1984 m, 2263 m and 2636 m),
sometimes cutting the main schistosity and made up of minerals (diopside-
hedenbergite, andradite,wollastonite) very similar to those of the Alpine
skarn ore deposits of southern Tuscany. The presence of such minerals even
at relatively shallow levels (core sample of 1600 m) could be related to
convective circulation of hot fluids, perhaps of part magmatic origin,
along major faults during a first phase of hydrothennal metamorphism(12).PT
conditions could have been in the range 370°-550°C and some hundreds to
1 Khar. During a cooler stage of hydrothennal metamorphism more abtmdant
hydrothennal minerals were produced: adularia, epidote,chlorite, sphene,
pyrrhotite, chalcopyrite, actinolite, prehnite and datolite, that are very
widespread in veins and vugs down to about 3300 m. A temperature range of
2SOo-3S0oC can be inferred for the latter minerals. These values are very
near those measured in the Sasso 22 well and the consistence between gas
partial pressures measured in the field (13) and those inferred from vein
minerals indicate that hydrothennal metamorphism was active tmtil very re-
cently.
S. CONCLUSIONS
Preliminary interpretation of the Sasso 22 well data has led to some
general observations.
The basement in the area crossed by the well is extensively and inten-
sely fractured, especially in correspondence to a gneiss fonnation between
- 308-
SASSO 22 WELL (420m • .• . 1)
o
~ knysec
.600
1100
1000
'800
1000
13' 3/ 8 '2000
'100
2000 2200
2300
2400
2500
1
)
noD
2800
i 2900
SonIc log
4000
:;;;:'11
References
(1) G.Bagnoli, G.Gianelli,M.Puxeddu,A.Rau,P.Squarci and M.Tongiorgi (1979)
A tentative stratigraphic reconstruction of the Tuscan Paleozoic
basement. Mem.Soc.geol.it., in press.
(2) P.Di Simplicio,G.Ferrara,C.Ghezzo,G.Guasparri,R.Pellizer,C.A.Ricci,
F.Rita and G.Sabatini (1975) II metamorfismo ed il magmatismo pal-
eozoico nella Sardegna. Rend.Soc.it.Min.Petr. 30, 979-1068.
(3)L.Amodio-Morelli,G.Bonardi,V.Colonna,D.Dietrich,G.Giunta,F. Ippolito,
V.Liguori,F.Lorenzoni,A.Paglionico,V.Perrone,G.Piccarreta,M.Russo,
P.Scandone,E.Zanettin-Lorenzoni,A.Zuppetta (1976) L'arco calabro-
peloritano nell'orogene appenninico-maghrebide. Mem.Soc.geol.it.
17, 1-60.
(4) M.Frey,J.C.Hunziker,W.Frank,J.Bocquet, G.V.Dal Piaz,E.Jager and E.
Niggli (1974) Alpine metamorphism of the Alps: a review. Schweiz-
erische Mineralogische und Petrographische Mitteilungen S4, 247-
290. --
(S) M.Seyler and M.Boucarut (1978) Donnees nouvelles sur la lithostrati-
graphie du massif des Haures Ie long d'une transversale Real Mar-
tin-Grimaud. Bull.BRGM 2 erne serie, section l,no.I,3-18.
(6) G.Gianelli, M.Puxeddu and P.Squarci (1978) Structural setting of the
- 310-
Larderello-Travale geothermal region. Proc.69th Congress Italian
Geological Society, Perugia, in press.
(7) G.Gianelli, M.Puxeddu and P.Squarci (1979) Studio dell'assetto strut-
turale della regione geotermica di Larderello-Travale. Proc.lst
Informative Seminar of the Geothermal Energy Sub-Proj ect, Final-
ized Energy Program, 224-235.
(8) H.G.F.Winkler (1976) Petrogenesis of Metamorphic Rocks.4th Ed. ,Spring-
er-Verlag,Berlin, pp.334.
(9) H.J.Greenwood (1976) Metamorphism at moderate temperatures and pres-
sures. In: (D.K.Bailey and R.McDonald,eds.)The EvulutiOll of the
fEystalline Rocks. Academic Press, London, 187-259.
(lO)M.Puxeddu,P.Squarci,A.Rau,M.Tongiorgi and P.D.Burgassi (1977)Strati-
graphic and tectonic study of Larderello-Travale basement rocks
and its geothermal implications. Geothermics ~, 1/2, 87-93.
(ll)F.Batini,P.D.Burgassi,G.M.Cameli,R.Nicolich and P.Squarci (1978) Con-
tribution to the study of the deep lithospheric profiles: 'deep'
reflecting horizons in Larderello-Travale geothermal field. Proc.
69th Congress Italian Geological Society, Perugia, in press . - -
(12)G.Cavarretta,G.Gianelli and M.Puxeddu (1980) H)'drothermal metamorphism
in the Larderello-Travale geothermal field. Geothermics ~-3/4, in
press.
(13)F.D'Amore,G.Gianelli and M.Puxeddu (1980) An attempt at reconstructing
a hydrothermal alteration process of the basement rocks of the
Larderello geothermal system. Geothermics, in press.
- 311-
SESSION III - Geochemistry
- 313-
HYDROGEOLOGY AND GEOCHEHISI'RY OF THE THERJ.lAL SPRINGS OF SOUI'H-l'iEST lliSCANY
SlUIIIllary
Various circulation systems have been identified in thermal areas of
south-west Tuscany: the main one develops in Mesozoic carbonate and/or
evaporitic formations, while other minor circulations flow through permea-
ble rock masses within the cover formations of the main reservoir.
As most of the springs emerge near widespread absorption areas the
temperatures computed with the various geothermometers represent minimum
values that cannot be extrapolated to deeper zones of the reservoir or
zones very far from the emergence point.
The geothermometers were also applied to the low salinity springs
within the classical geothermal area (Larderello); in this case, despite
the fact that the circulation proved to be very shallow in the cover for-
mations, the geothermometers indicated the presence of a marked thermal
anomaly.
- 315 -
1. INTRODUCTION
This study was aimed at defining hydrogeo1ogica11y and geochemica11y
the main hydrothermal circuits of centra1.,.southern Tuscany, between the
Monticiano-Roccastrada-Iano ridge in the east, the Arno river in the north
and the Tyrrhenian Sea in the west (Fig. 1) •
Once these data have been processed and integrated with the preceed-
ing studies conducted for the European Conmnmity (6) the hydrogeochemical
survey of central-southern Tuscany will be completed. This survey will
utilize unpublished data and others drawn from the literature (1),(5).
- 316-
224,226, 214 and 96. The characteristics of this group of springs are:
a) they emerge from the main Mesozoic carbonate reservoir;
b) sulphate-calcium-magnesian type;
c) salinity between 1300 and 3300 ppm;
d) saturated in calcite,dolomite, fluorite and anhydrite (Table I);
e) rather high Sr concentration values that are in proportion to the sul-
phate concentration;
f) salinity expressed as ionic strength (see Fig.3)is proportional to the
disequilibrium index of anhydrite;
g) in the Ca-Mg activity diagram they lie along a calcite-dolomite straight
line for temperatures of the order of 60°C.An average CO z pressure of
the order of 0.1 atm is typical of over-saturation in calcite and dolo-
mite;
h) the Na-K diagram shows that they are not in equilibrium with albite or
microcline, so that the Na/K geothermometer cannot be applied;
i) most of the samples have a less negative ~180 value than that correspon-
ding to the emergence elevation. These wat~rs can therefore be assumed
to belong to a regional low-temperature circulation that undergoes no
obvious isotopic exchange with the reservoir rocks.
The average elevation of the recharge areas of the springs emerging
in the Grosseto area (nos.136,138,139 G and D) is 150-ZOO m. The latter
figures suggest that much of the recharge takes place through the carbonate
outcrops north-northeast of Grosseto.
The springs further north that belong to the hydrothermal circuits of
Venturina (Z13) and Massa ~brittima(ZZ6) are fed at elevations of more
than Z50 m. The absorption areas for these two systems are mainly the wide-
spread carbonate outcrops immediately north of the springs.
These three groups of springs,associated with local absorption areas
of widespread carbonate-type outcrops and with flow-rates of around one
m3/min, are typical examples of the type of hydrothermal circuit found in
many parts of central-southern Tuscany (e.g.,Saturnia no.140,Le Galleraie
n.Z14, Bagni di Casciana no.96).This type of circuit is shown in sketch
- 317-
form in Fig.4 (A); after infiltrating the extensive outcrops it flows in-
to the deepest carbonate-anhydritic complex.
Table II shows the values computed for the reservoir temperatures,ob-
tained either from the silica geothermometer in equilibrium with chalce-
donyor quartz (when the latter phase is present in the reservoir rocks) ,or
from the gas-geothermometer (3), whenever there are local manifestations
for sampling.
These calculated temperatures can, generally speaking, be said to
provide indications on the temperature regime in zones of the reservoir
very near the springs in question which, in our case,lie near widespread
recharge areas. Further away from these areas the reservoir temperatures
beneath the cover may be much higher (a typical example is that of spring
no.Z14, with calculated temperature of around 70°C, sited in the immediate
vicinity of Travale geothermal field).
- 318-
Na-montmorillonite and kaolinite and along the albite-microcline
straight line at 60°C;
j) chlorine content is about one quarter of the sodium content.
ite.
The presence of chlorides may be tied, in this case, to the circula-
tion of the waters in Upper Miocenic evaporitic deposits.
The temperatures calculated by the silica geothermometer (chalcedony)
are of the order of 60°C for no.97, which is less affected by mixing with
the Na-bicarbonate waters of the clays, and 30°C for no.103 which is the
richest in Na-bicarbonate. The Na/K geothermometer gives temperature val-
ues of the same order (Table II).
Figure 4 illustrates the hydrothermal circuit (in sketch form) hypo-
thesised for family (B).
Spring no.101, near Miemo, belongs to the water family (E) ;although
lying in the same zone of the Piper diagram as the CD) family, it has some
distinct characteristics:
a) it emerges at the contact between the shaley formations and a thick
- 321 -
mass of ophiolitic rocks, both belonging to the cover complex of the
carbonate reservoir that does not outcrop in this area;
b) low salinity (about 800 ppm);
c) saturated in calcite,magnesite and dolomite at an emergence temperature
of about 30°C. Unsaturated in anhydrite and fluorite;
d) calcium and magnesian bicarbonate water, with CO 2 pressure of the order
-2
of 10 atm;
e) very low Sr content;
f) it lies along the extension of the albite-microcline straight line at
about 100°C;
g) similar to nos.97 and 103 (B family) ,it has an anomalously high Mg con-
centration, equal to that of Ca.
Drilling data from near the spring suggest that the ophiolitic rocks
housing this circulation are around 1 km thick and that the ophiolites are
separated from the main carbonate reservoir by a marly limestone formation.
The high Mg contents can therefore be easily interpreted as the re-
sult of an interaction between the water and serpentine. The serpentine-
water alteration processes in the presence of relatively low CO 2 pressures
are not simple, but one of the metamorphic reactions suggested in the li-
terature (7) accounts for serpentine alteration to talc,magnesite and
quartz at low temperatures.
A rather high pH, as in our case (7.3) permits precipitation of dolo-
mite. The composition of the solution is capable of being saturated in
quartz and magnesite, but not talc, at the emergence temperature: this is
only possible for temperatures above 60°C.
Calcite precipitation buffers the Ca content, which is kept low com-
pared to the Mg concentration released by the ophiolites.
The temperature obtained with the quartz geothermometer is 80 0 Ci this
would become 109°C were we to use the Na/K geothermometer. Note that the
gas-geothermometer (3) applied to the nearby manifestation of Orciatico
gave a value of the order of 115°C for the reservoir. In-hole measurements
in the carbonate-anhydritic reservoir underlying the ophiolitic-flysch
- 322-
formations gave values of the order of 150°C at about 1500 m depth.
Three springs (A,B and C) have more positive 6180 values than those
of the emergence elevation. Spring A, lying in the Monticiano-Roccastrada
structure on the eastern margin of the study area, belongs to the group of
manifestations (springs and gas) characterized geochemically by a thermal
anomaly. This factor was revealed both by a study of Siena Province (6)
and by a new gas-geothermometer (3). In fact, waters of an anomalous ~180
value (-2.8 ; -5.8) emerge in the Farma and Merse rivers on the margin of
Siena graben, immediately east of spring A.
These values have been interpreted as resulting from an isotopic ex-
change reaction or from evaporation phenomena.
On the contrary, the ~180 value of spring B (-2.8), which emerges
very near the coast-line at Talamone, together with the chemical composi-
tion (2), suggest that almost one third of the water is sea-water.
4. CONCLUSIONS
The calculated temperature values give indications on the local res~
ervoir temperature that are in agreement with the data calculated from the
deep wells and geothermal test-boreholes (2). However, as the springs gen-
erally emerge near widespread absorption zones for cold meteoric waters,
the temperature value cannot be extrapolated to more distant zones beneath
the cover. Temperature anomalies in the deep reservoir cannot therefore
be excluded, even in nearby zones.
The temperature values given by the geothermometer for the low salinity
springs in the classical geothermal area (Larderello) reveal the presence
of a marked thermal anomaly, despite the fact that the circulation is very
shallow in the cover formations.
Table II and Fig.4 summarize the main results of the study,including
~
-323-
Table I - Some geochemical parameters calculated from the chemical analyses for thermal springs:
emergence temperature, CO 2 pressure, ionic strength (I.S.), silica activity, disequilibrium
indices for some minerals.
3D
4~
••
'--~~---J'
....
Figure 1: Index and geological sketch map of central and southern Tuscany.
1.Alluvial and marine deposits (late Quaternary). 2.Granite (a)
and volcanics (b) :0.4-S my. 3.Neoautochthon: marine deposits,
mainly clay (Pliocene-early Quaternary) and fluvial and lagoonal
deposits(late Miocene). 4.Ligurian series(late Jurassic to middle
Miocene). S.Upper terrigenous complex of Tuscan series,including
sandstone (~~cigno) and shale (Scaglia) (early Cretaceous to ear-
ly Miocene). 6.Carbonate complex of Tuscan series(late Triassic
to Jurassic). 7.Basal terrigenous (mainly quartzitic)complex of
Tuscan series(middle to late Triassic) and Paleozoic basement
(mainly phyllite).
- 326-
SOf·cl
40 30 - 20 10
40 10
.103
/215 ·97
30 20
.217
• I ~.+
~
<0
20 30 ~
219.
220
10
221~
ri :S
40
138 211
139".1 14O
101.
•223
·218
130-
. 103
110
~ 70
230.
~
U;
...
~50 1~~
2~:::',~
,
30 • 13'
• 213
218 211
10 !t220 • 101
223
- 327-
E A B
o
L . . '_
lkm
. . . . . . . . ._ - ' - -_ " ' - - - - - - - '
,
2 3 4 5 6 7 8 9
- 328-
References
(1) Bencini A., Duchi V. ,Martini M. (1977) Geochemistry of thennal springs
of Tuscany (Italy). Chem.Geol. 19,229-252.
(2) Cataldi R., Lazzarotto A., Muffler L.J.P., Squarci P.,Stefani G.(1977)
Assessment of geothennal potential of central and southern Tuscany.
Geothermics 1-2/3/4, 91-131.
(3) D'Amore F., Panichi C.(1980) Evaluation of deep temperatures of hydro-
thennal systems by a new gas-geothennometer. Geochim.et Cosmochim.
Acta, in press.
(4) Francalanci G.P. (1959) Contributo per la conoscenza delle manifesta-
zioni idrotennali della Toscana. Atti Soc.Toscana Sci.nat. 65,2,372-
432.
(5) Panichi C., Celati R., Noto P.,Squarci P., Taffi L., Tongiorgi E.(1974)
Oxygen and hydrogen isotope studies of the Larderello (Italy) geother-
mal system. In 'Isotope Techniques in Groundwater Hydrology 1974',
IAEA, Vienna.
(6) Panichi C.,D'Amore F.,Fancelli R.,Noto P.,Nuti S.(1977) Geochemical
survey of the Siena Province. Interpretation. Proc.Seminar on Geother-
mal Energy, Brussels ~,48l-503.
(7) Winkler H.G.F. (1974) Petrogenesis of Metamorphic Rocks. Springer-
Verlag.
- 329-
ORIGIN AND CIRCULATION OF THERMAL WATERS IN THE UPPER RHINE GRABEN:
by Hans Friedrichsen
Universitat TUbingen, Abt. Geochemie, 7400 TUbingen, F. R. G.
SUl1lllary
The reservoir temperatures of 22 hot-spring waters from 9 localities
in the Upper Rhine ~alley and the northern Black Forest have been deter-
mined. The 180 (S04 -) thermometer and the Si02 thermometer yield concor-
dant temperatures 1n those thermal waters which have not been diluted by
near-surface ground water components. Deviations of the Si0 2 thermometer
- on the other hand - can be applied for an estimate of near-surface water
components in hot spring waters. The alkali content of the waters can not
by used for geothermometry.
The highest reservoir temperatures have been calculated for the hot
springs of Baden-Baden (1560 C). All other hot springs have substantially
lower reservoir temperatures {115 to 650 C). The 180 and D-values of
the thermal water component are comparable with the mean of the 180 and
D values of modern meteoric waters from higher elevations of the Black
Forest; the cold components have stable isotope ratios of local precipita-
tion.
- 330-
1. INTRODHCTION
Thermal waters from the Upper Rhine Valley and the Northern Black
Forest (Fig. I) shown large variations in their chemical composition, and
minor variations in their stable isotopic ratios. It was the purpose of
this study to obtain information on:
1.) the flow paths of the hot-spring waters and the possible recharge
areas
2.) the thermal history and the reservoir temperatures, and
3.} the contents of near-surface water components in the different
wells.
~1
Since the work of Craig (1963) it has been known that all hot spring wa-
ters are ultimately of local meteoric origin. Juvenile water components
with their characteristic isotopic composition have never been detected
in hot spring waters and can not be a major component.
As meteoric waters percolate through the upper crust, many reactions
occur between the rocks and the water which change the chemical compositon
-331-
of the soluble components in the water as well as the stable isotope com-
position of the oxygen, and the gas content (Fig. 2).
rock
water
clay minerals
'80. "He, Rn zeolites etc
IncreaSing temp
deeper crust. mantle
fig. 2
- 332-
t (oC) = -~~---
777 - 273.15
log (Na/K) + 0.7 (Fournier and Truesdell
1973/1976)
1647 _ 273.15
log (Na/K)+ B log (VICa/Na) + 2.24
(Fournier and Truesdell
1973/1976)
These relations are valid if the feldspars are the only phases which
react with the water.
2.) from the Si0 2 content of water:
t (oC) = A - 273.15
B - log Si0 2
B = 5.768, A = 1533.5 for quartz, adiabatic cooling,
B = 5.205, A = 1315 for quartz, conductive cooling.
B = 4.655. A.= 1015.1 for chalcedony. conductive cooling.
3.) from the 180/160 ratios of the sulfate:
10 3 In<< = 2.88 . 10- 6 • T- 2+ 4.1. T: (oK)
(180/16 0) sulfate
where tt = ....--"'"----'------
(180/16 0) water
Several other thermometers have been used (gas contents. '13C/12C and
DIH ratios in gases) for the determination of reservoir temperatures. but
are not applicable in this area (low content of relevant gases)
The 0- and H-isotopic compositions of meteoric water at a given lo-
cality are determined by several parameters. mainly the geographic situa-
tion (distance from the ocean. mean annual temperature and topographic
height) (for details see: Craig, 1963). In the area studied. rain water
from the Rhine Valley is enriched in 180 and 2H in comparison to rain
water (or snow) from the Black Forest hills. Seasonal variations (180 and
2H depletion in winter. 180 and 2H enrichment in summer) also have to be
considered. Using these natural isotopic tracers. it should be possible to
identifly recharge areas.
2. EXPERIMENTAL PROCEDURES
Na+, K+, ca 2+. Mg2+ and Si0 2 contents of the filtered waters have been
- 333-
determined by standard AA5 methods, The standard deviation is + 1 %for
~
Na, K, Ca and Mg, and ~ 2 %for 5i0 2, 504 and Cl were analysed by ti-
- -
tration with a standard deviation of~ 1 %,
I f'CI
"o/SO, 1'-
Ihfrmom.
100
- ••
- •
•
50
•
fig. 4
I f'CI
150
Si021QI
thermom.
100
50
fig. 5
- 334-
t('CI
150
SiO] (01
/hermom.
100
50
fig . 3
- 335-
due to the complex and unknown chemistry of the rock types which reacted
with the water.
Highly concordant temperatures have been calculated using the 180
(S04) and Si0 2 thermometers. Only a few exceptions have been found and a
reasonable interpretation can be given for these deviations. Within ex-
perimental error the Si0 2 thermometer never yielded higher temperatures
than the sulfate thermometer.
The temperatures which were derived from the sulfate data did not
change with time. The Si0 2-derived temperatures were rather constant in
some cases (! 1-20 C, Baden) but showed larger variations in others (! 6-7 0
C, Rotenfels). There is a principle difference in the application of these
two thermometers: if the hot-spring water is diluted by (Si0 2-free) near-
surface ground water of a similar isotopic composition, this has no effect
on the temperatures derived from the sulfate thermometer. However it does
change the Si0 2-content of the mixture, and a lower apparent temperature
will then be derived from the calculation.
In Fig. 3 the deviation of the temperatures is plotted as a function
of the Si0 2-free. near-surface ground water component.
These d~iations can provide important information for the calculation
of the different components (groundwater and hot-spring water) for a two
component mixed-water model.
If we assume that the sulfate thermometer gives us the reservoir
temperature of the hot-spring water and that the hot component carries the
equilibrium Si0 2_content of the hot reservoir, than we can calculate the
amount of the near-surface water components. The different thermal waters
of Rotenfels (Ro). Liebenzell (Li) and Bergzabern (Be) contain consider-
able amounts of the cold component. In Figs. 6-9 the discharge temperatures,
the isotopic ratios and the salt content are plotted against the cold wa-
ter content of the thermal waters of different springs from Rotenfels.
Discharge temperatures and salt contents are lower for waters with (cal-
culated) high near-surface ground water contributions. The J180_ and J D-
ratios are high for the cold water component and are in good agreement
(Fig. 14) with the stable isotope data for the precipitation in the Rhine
Valley. The hot component is higher inJ 180 andJD, and charactaristic of
precipitation from the Black Forest hills. The data fgr the mixed waters
-336-
from different wells in Bad Liebenzell (with up to 20 %calculated cold
component) show a similar pattern (Figs. 10-13). The cold component is the
carrier of the Mg2+ in these waters.
ROTENFELS
cl 0 1%.1
\ \
- 68
:II
- 67
-66
19
-65
1%) noor sur /ace 1%ln 5 we
50 OJ 50 OD
WO'£II' componf!nl
\
Soli conten~
-99 J 1801%.1
{g/I/
6
- 95
:I
~, ,
"- ,
-90 a
f'%J n S w e 50 /00
{%Ins ",e
50 100
fig. 6-9
BAD LlEBENZELL
~.
- 68
~
-(iT
- 66
10
''' ]nswc {'Xi ltr. we
10 10 10 20
-9 5
0
10 20
I"JI't5WC 10 JO "" n.s we
fig. 10 -13
- 337-
The dO-values and the ci 180-values of the various hot-spring waters
fit the meteoric water line (mwl) derived by Craig (1963) (Fig. 15), and
are in good agreement with the mean of modern meteoric waters. The hot wa-
ter component for all hot springs (except Mingolsheim) must have recharge
areas with higher elevations, while the cold component is heavier in J180
and J 0 and probably represents a mixture of precipitation from the higher
regions and the valleys.
-15 -10
-50
Fhne volley
mean QfQn~'
preellito/lon -
Black Forest h"s F ~Rhine valley
orrplitude of
seasonal YOriotion _
_.
V_arpltUde seasonal
fprecip/otlOflJ \ vario/lon(river water J
Block Fores!
-100
fig. 14
-110
_! 180
-10 -8
-60
fig . 15 -70
- 338-
4.CONCLUS IONS:
I.} Reservoir temperatures for hot-spring waters in the Upper Rhine
Valley and the Northern Black Forest have been calculated from Jl80-values
of sulfates and from Si0 2 contents.
The highest reservoir temperatures have been calculated in the Baden
area (155 0 C). The reservoir temperatures in Wildbad and Liebenzell are
between 110 and 1200 C. Lower temperatures (between 65 and 90° C) were
calculated for all other regions investigated in this study.
2.} The Si0 2 contents of the hot-spring waters yield temperatures
concordant with those derived from the sulfate thermometer for waters with
minor contributions of cold components.
3.) Cold water components in mixed waters have been calculated using
deviations of the Si0 2 thermometer from the sulfate thermometer.
4.} The Na+, K+ and Ca 2+ contents of the thermal waters reflect the
complex chemistry of the rocks which interacted with the fluid.
5.) The JD- and JIBO-ratios of the thermal waters are essentially the
same as modern meteoric waters.
REFERENCE S:
CRAIG, H. (1963) : The isotopic geochemistry of water and carbon in geo-
thermal areas; in Tongiorgi, E., ed., Nuclear geology on geothermal
areas. Spoleto, 1963: Pisa. Consiglio Nazionale delle Ricerche La-
boratorio di Geologia Nucleare, 17 - 54.
FOURNIER. R.O., TRUESDELL. A.H. (1973) : An empirical Na-K-Ca geothermo-
meter for natural waters; Geochim. et Cosmochim. Acta 37. 1255 - 1275.
TRUESDELL. A.H. (1976) : Summary of section 3; Geochemical techniques in
exploration; Proc. 2nd U.N. Sympos. on the Development and Use of
Geothermal Resources, San Francisco 1975, 1, u.s. Govt. Printing
Office.
- 339-
REFINEMENTS IN THE USE OF CHEMICAL GEOTHERMOMETERS
(FRANCE)
Summary
- 340-
I. INTRODUCTION
Some springs in the group of cold ones, were considered as polluted and
not taken into account because showing a quite high concentration of K, CI
and above all, N03' The (N03) concentration decreases rapidly when tempe-
rature increases, which suggests (not surprisingly) that the pollution is
from a cold reservoir. These few springs were avoided in the following cal-
culations (Savonneuses and 2 principally). Two other reservoirs, a cold
one and a hot one can be clearly distinguished.
The temperatures of emergence range from 9°C to 73°C. Correlations
between concentrations (Si02), (504) and (Na) are excellent and are better
than those observed by plotting same concentrations versus the emergence
temperatures. This suggests that some lost of heat by conduction creates
cooling and that temperature of emergence has been corrected in order to
obtain the "right" temperature, given by the relationship (l2):
TOC corrected = 47.62 (S04) (mmoles / kg H20) + 29.47
and range now from 31°C to 84°C. Furthermore, the cooling corrections al-
low to show that all concentrations are very well correlated with tempera-
ture • One can suggest that chemical compositions result of a simple mixing
in different proportions of two waters, one of a cold, the other one of a
hot reservoir. The position and the slope of this straight line are suppo-
sed calibrated on the rightest points and on the general trend of the
- 341-
NU, Li N. K Hg C. ST t meq CI NO, SO, Ale. l: mcq 5iOa pI! 10gfC02. T·cm. T~or.
Al -:"601 0,094 0,048 0,086 0,198 0,710 0,036 0,0':'1 0,0"4 0,025 0,514 0.707 0,202 6,73 -1,362 9,5 30,7
Al -76b 0,096 0,048 0,081 0,205 0,716 0,060 O.l)j~ 0,075 0,016 0,496 0,701 0,202 6,94 -2,588 9,4 30,2
Al -77 0,083 0,050 0,081 0,190 0,675 0,058 C,040 0,076 0,015 0,460 0,674 0,192 6,68 -2,34011,7 30,2
Table I
Al -78 <0,002 U,OOI 0,100 0,047 0,087 0,214 <0,001 0,754 O,OhO 0.0.\0 0.083 o .(lIS 0,550 0,7,)<) 0,218 6,1.6 -2,Ot'l7 9,9 30,2
5a2-76 0,569 0,128 0,221 0,485 2,108 0,090 0,944 0.208 0,116 0,528 2.002 0,237 6,81 -2,336 ~1,7 35,0 Chemical analyses of'
5a2-77 0,640 0,131 0,240 0,560 2,371 0,074 1,150 0,160 D, 152 0,460 2,208 0,210 7,34 -2.974 16,5 36,7
5a:!-78 <0,002 0,004 0,465 0,101 0,167 0.184 <0,0011,676 0,095 0,650 O,Ib4 0,111 0,540 l,b72. 0,234 6,65 -2,181 21.8 34,8
5al-76 0,540 0,131 0,257 0,490 2,105 0,095 1,099 0,218 0, 133 0,488 2,166 0,237 7,32 -2,857 24,6
water of the principal
3~.8
5al-77 0,745 0,133 0,247 0,610 2,592 0,095 1,100 0,140 0.192 0,518 2,331 0,240 6,76 -2,315 19,7 38,6
S81-78 <0,002 0,004 0,425 0,107 0,202 0,374 <O,OOII,MO 0,100 0,740 0,140 0,114 0,495 1.704 0,242 7,20 -2,74624,9 34,9 springs of Plombieres.
583-76 0,320 0,067 0,122 0,219 1,069 0,137 0 1 219 0,043 0,097 0,468 1,061 0,280 6.70 -2,233 26.0 34, I
S.3-71 0,284 0,065 0,116 0,218 1,017 0,144 0,215 0,037 0,066 0,440 1,019 6,61 -2,165 26,S 32,6
S.3-78 <0,002 0.005 0,307 0,063 0,123 0,213 <0,001 1,052 D,142 0.217 0,064 0.OB5 0,440 1,033 0,276 6,40 -1,98225,7 33,S
Sa4-76a 0,800 0,070 0,105 0,238 1,556 0,211 0,103 0,029 0,205 0,800 1,557 0,470 7,00 -2,272 30,0 39,2
Sa4-76b 0,832 0,071 0,105 0,241 1.595 0,247 0,099 0,031 0,214 0,808 1,613 0,497 7,30 -2,564 30,6 39,7
Al=Alliot;
5.4-77 0,780 0,074 0,110 0,254 1,582 0,252 0,115 0,044 0.IB4 0,765 1,609 0,470 7,08 -2,369 30,4 38,2
5a4-78 <0,002'0,013 0,832 0,070 0,105 0,235 <0,001 1,599 0,263 0,099 0,054 0,198 0,800 1,612 0,505 7,00 -2,290 30,2 38,9 Hy=Hypocauste;
Sa5-76 0,416 0,074 O,tOI 0,212 1.116 0,168 0,080 0,050 0,152 0,524 1,126 0,312 6,94 -2,380 31,4 36,7
5a5-77 0,390 0,072 0,098 0,198 1,054 0,157 0,085 0,047 0,112 0,510 1,053 0,300 6,57 -2,023 31,2 38,4
5aS- 78 <0,002 0,005 0,398 0,066 0,099 0,203 <0,001 1,077 0,163 0,088 O,06t 0,133 0,520 1,098 0,320 6,38 -1.48131,5 35,8
Ca.=Capucin;
5a6-76 0,328 0,079 0,094 0,187 0,969 0,129 0, )49 0,085 0,115 0,412 1,000 0,260 7.21 -2,76431,4 34,9
5a6-77 0,320 0,081 0,091 0,180 0.943 0,118 0,140 0,069 0,085 0.420 0,911 0,275 7,10 -2.644 30.3 33,S Da.=Dames;
5.6-78 <0,002 0,003 0,326 0,078 0,094 0,200 <0,001 1,009 0,126 0,160 0,076 0,108 0,364 0,942 0,258 6,78 -2,404 30,6 34,6
5a7-76 1,420 0,085 0,084 0,252 2.117 0,315 0,108 0,048 0,364 0,976 2,175 0,690 7,46 -2,56341,2 lt6,8
5.7-77 1,330 0,095 0,079 0,232 2,047 0,325 0,1 ~5 0,017 0,310 0,750 2,057 0,660 7,28 -2,504 40,0
Sc=Sainte Catherine;
44.2
...,... 5a7-78 <0,002 0,021 1,430 0,082 0,075 0,229 0,001 2,145 0,316 0,126 0,058 0,332 D,950 2,114 0,675 7,08 -2.19742,0 45,3
N 5.8-76 2,040 O,08S 0,073 0,263 2,797 0,389 0,126 0,039 0,450 1,360 2,814 0,890 8,26 -3,21247.5 50,9 So=Sondage 9;
5.8-77 1,900 0.089 0,069 0,256 2,639 0,390 0,120 0,034 0,425 1.130 2.714 0,835 8,14 -3,16846.8 49,7
5a8-78 <0,0020,034 1,900 0,089 0,066 0,252 0.'001 2.663 0,374 0,148 0,047 0,435 1,280 2,719 0,890 7,94 -2.90547,0 50,2 Ro=Romaine;
5a9-76 1,980 0,084 0,074 0,265 2,742 0,358 0,126 0,041 0,466 1,348 2,805 0,900 8,16 -3,10941,7 31,7
5.9-77 1,880 0.090 0,072 0,258 2,630 0,378 0,125 0,030 0,420 1,130 2,693 0,835 8,06 -3,088 46,0 49,5
5a9-78 <0,0020.031 1,770 0,085 0,067 0,253 <0,002 2,532 0,368 0,123 0,053 0,445 1,280 2,715 0,875 7,94 -2.913 <+b,O 50,7
Sa.=SavonneuseS;
5al0-76 2,030 0,091 0,073 0,255 2, ))7 0,326 0,131 0,043 0,518 1,336 2,872 0,900 8,25 -3,195 ':'9,5 54, I
5aI0- 77 1,940 0,091 0,070 O,2b6 2,703 0,410 D,140 0,033 0,420 1,150 2,782 0,850 8,14 -3,14149,5 49,S Cr=Crucif'ix;
Salcr78 <0,002 0,031 1,860 0,087 0,066 0,248 <0,001 2,610 0,370 0,126 0,047 0,440 1,300 2.723 0,885 7,96 -2.894 50.0 50.4
By -76 2,74 0,101 0,048 0,236 3,409 0,526 0,172 0,031 0.645 1,520 3,539 1,150 7.83 -2,72145,6 bO,2
By -)) 2,64 0, 108 0,044 0,222 3,280 0,525 0.170 0,014 0,660 1,460 3,549 1,140 8,18 -3,12044,5 60,9
MO=Mougeot;
I!y -78 <0,002 0,045 2,83 0.102 0.041 0,218 0,001 3,499 0,526 0,165 0,033 0,645 1,510 3,525 1.160 7,86 -2.75746,5 00,2
Cr -18 <0,002 0,061 3,18 0,135 0,012 0.184 0,002 3.774 0,632 0,200 0,028 0,785 1,530 3.961 1,460 8,32 -3,238 !.8.S 66,5 5b=5bis;
Ca -76 2,27 0,098 0,056 0,290 3,060 0,432 0,203 0,059 0,540 1,336 3,110 0,970 7.91 -2,830 49,6 55,2
c. -77 2.14 0,107 0,057 0,330 3,021 0,457 0,240 0,052 0.550 1,295 3,249 0,990 7,85 -2,79647,8 55,7
2,24 0,104 0,050 0,280 0,002 3,051 0,460 0.193 0,055 0,530 1,320 3,189 1,030 7,51 -2,429 49,5 54,7
Va.=Vauquelin;
Ca -78 <0,002 0,041
Ko -77 3,68 0, ISS 0,007 0,190 4,229 0,790 0,265 0,021 0,890 1,720 4,577 1,600 8,50 -3,391 49,7 71,9
Ko -78 <0, 002 0,074 3,70 0,152 0,005 0,155 0,002 4,252 0,737 0,270 0,045 0,090 1,700 4,553 1,640 8,13 -2.911) 57,0 72,3 Rr=Robinet Romain.
Da -76a 2,94 0,107 0,025 0,208 3,513 0,537 0.180 0,029 0,680 1,524 3,630 1,230 8,17 -3,037 52,0 61,9
Da -76b 2,93 0,103 0,042 0,217 3,551 0,484 0,175 0,031 0,670 1,560 3,590 1,240 8,05 -3.037 52,0 61,4
Da -77 2,96 0.110 0,038 0,216 3,578 0,578 0,170 0,018 0,650 1.430 3,626 1,160 8.40 -3,32552,1 60,4
Da -78 <0,002 0,046 2,93 0,104 0,036 0,202 0,001 3,560 0,532 0,176 0,029 0,675 1,520 3,608 1,210 7,98 -2,R27 S3,O 61,6
58 -78 <0,002 0,059 3,66 0,133 0.011 0.187 0,001 4,252 0,737 0,210 0,027 0.820 1.640 4,258 1,500 8,28 -3,063 65,0 68,S
5c -77 3,28 0,130 0,015 0,198 3,836 0,682 0,210 0,021 0,830 1,605 4,213 1,400 8,28 -3,G75 62,2 69,0
5e -78 <0,002 0,056 3,50 0,116 0,0)5 0,184 0,001 4.077 0,684 0,204 0,033 0,805 1,560 4,094 1,420 8,10 -2,856 62,0 67,8
Va -77 3,92 0,148 O,C02 0,149 4,370 0.774 0,215 <0,001 0,940 1,820 4,689 1,700 8,50 -3,273 67.0 74,2
Va -78 0,009 0,074 4,12 0,143 0,001 0,139 0,002 4,630 0,R42 0,220 0,001 0,990 1,780 4,824 1,770 8,28 -2,982 69,3 76,6
50 -76a 4,55 0.153 <0.001 0,134 4,979 0,947 0,232 0,001 1.005 1,920 5,110 1,810 8,50 -3,227 72,,3 77 ,3
50 -76b 4,54 0,146 <O,et)1 0,136 4,958 0,805 0,256 <0.001 1.070 1,948 3,150 1,800 8,20 -2,83fJ 72.B 80,4
50 -77 4,52 0, i SO 0,002 0,142 4,958 0,215 O,22j <O,C81 0,945 1,840 4,810 1,760 8,44 -3.16573.U 74,5
50 -78 0,009 0,078 4, II (}.142 f),OOI 0,135 O,C02 4,61 :5 0,789 O.~27 <.],(101 1,01 () 1.800 4,834 1,790 8,12 -2.762 71.0 77 ,6
o N
0.40 00,40
N
I I
-
CJ' CJ'
.x -'"
"- "-
VI
j 0,30 ~
§
0,30
:i: .-••
¥0,20 • £'0,20
•
~
0,10
0,10
~o 'fiR> R>
8'~
~
o~
O~~~~~~~~~~~~
0~~~=25~~~5~0~~~75~~~100 25 50 75 100
TEMPERATURE O( TEMPERATURE (ORRIGEE t
pH 8,5 GO
0
0 0
00
00 0 g
o 0.10
cP e 00 0
0
I
N
0 0
8,0 0 C7'
.:.:
Q 0
0 "-
-5
VI
8 0,08 o
~ 00
15 0
0
l:
00
0 -.J 0,06 o
o
0 0 0 o
7,0 0 0
0
0 o
80
0,04
0
0 o
6,5 o
oeD
0,02 o
o
6,0
° 25
TEMPERATURE (ORRIGEE o(
50 75 100
° 25 50
TEMPERATURE CORRIGEE
75 100
o(
- 343-
points. A supposition of a non-preferential cooling of the cold springs was
retained; so that, the slope of the cloud of points after mixing (fig. I)
is the same as the observed one.
4. CHEMICAL GEOTHERMOMETERS
The solution saturation (7) (9) (10) with respect to calcite and fluo-
rite (in waters for which temperatures are higher than about 50°C) involves
a continuous redissolution of precipitated minerals, the different mixings
yielding a temperature above 50°C. In such cases, the use, in hot waters,
of thermometers based on calcite and fluorite solubilities, will only indi-
cate the temperature of mixing. In cold waters, the dilution model can only
indicate a mixing of a "hot" water reservoir at 50°C. Furthermore, the use
of calcium in Na-K-Ca thermometer is, for these reasons, inappropriate, for
finding the temperature of the hot deep reservoir.
The 8£02 thermometer
Because of the mixing processes, the direct application of the Si02
thermometer (I) (3) (4) fails. The straight line obtained plotting
(Si02)tot as function of temperature can be extrapolated until crossing the
quartz solubility curve of (13) at the supposed temperature of the hot re-
servoir which is equal to 288°C. This temperature is first, certainly
overestimated because the pH of the waters increases with temperature.
Then, the silica measured after mixing between hot and cold waters is co-
ming from a part of H3SiO~ transformed into H~SiO~o when temperature and
pH are lowered. The problem is in the dispersion of pH between T = 47°C
- 344-
'".-
o
x 50
~a
~ II">
o
";; 1.0
: 1,0 o
30
20 20
10 10
e
o
D
o 25 50 75 100 o 25 50 75 100
TEMPERATURE (ORRIGEE O( TEMPERATURE CORRIGEE t
10
TEMPERATURE t TEMPERATURE °c
- 345-
to T = 80°C so that a precise extrapolation at higher temperature is not
available. In the hot reservoir, an equilibrium with quartz requiers a
pH = 9.2 at 100°C and 8.75 at 200°C. But, for temperatures lower than
180°C, the pH conditions yield an alkalinity greater than this obtained by
linear extrapolations of concentrations:
Alk. = (Li) + (Na) + (K) - 2(S04) - (F) - (Cl)
In conclusion, it seems that our solutions can never be controlled by
quartz in the Plombieres system.
The (Na}/(K) variation as function of temperature shows a trend which
is in the reverse sense compared to the one experted so that the (Na)/(K)
geothermometer cannot be applied directly. Therefore, it can be used in-
directly. If one supposes that in the hot reservoir, the ratio of
[Na+]/[K+] (activities) is controlled by the equilibria with albite and
microcline, this ratio changes as function of temperature of mixing and is
described by an hyperbola (molalities):
(Na) tot 0.090884 T - 2.717
(K)tot 0.001981 T - 0.009
There are two different curves (Fig. 4) one for activities, the other
for molalities which crosses the microcline-albite equilibrium curve at
temperatures between 130 and 140°C. The curve calculated with activities
is slightly but dependent of the pH because of the formation of the com-
plex ions such as KSO~, NaSO~, and above all NaCO~, which are pH dependent.
It seems, at this stage, that the NaK geothermometer is the one which gi-
ves the most reasonable temperature of the hot reservoir. This determina-
tion may be accurate, taking into account the fact that the equilibrium
curve is steep in that range of temperatures.
5. CONCLUSIONS
6. BIBLIOGRAPHY
-348-
GEOCHEMICAL SURVEY OF SHALLOW GROUND WATER WELLS IN DENMARK.
O. S. JACOBSEN
Abstract
The stability of SiO contents in water samples have been
tested as a function of time and salinity. There does not seem
to be a correlation between salinity and decrease in SiO? con-
tent with time. The regional ion-exchange of Ca, Na and K have
no significant effect on deep ground water. Thus, the Si0 2 may
co-precipitate due to oxidation of iron(II).
The highest concentration of Br occurs in saline ground
waters, alth~Mgh the Cl/Br ratio increases by increasing sali-
nity. The 6 0 investigation on selected saline wells confirms
that probably no wells have marine infiltration. The measured
tempsrature range of the investigated wells was between 7.5 and
14.5 C. No indications of migrating hot waters from lower geo-
logical strata were found, whereas migration of saline solu-
tions was determined.
The use of different types of chemical geothermometers can-
not be tested satisfactory due to the very small range in the
temperature interval. Thus, the silica-geothermometers appears
less reliable within the Danish sedimentary bassin, whereas
the Na-K-Ca-Mg geothermometers may not be refused as usable,
even in calcareous reservoirs.
-m-
1. Introduction
- 350-
2. Methods
Water samples were taken from the bottom of the open wells
before and after the main pump was connected, using a tube
sampler in the lowermost) meter of the wells.
-351-
2.3 The stability of 5i0 2_
Table I
Reservoir geology no of samples
Quarternary 30
Upper Tertiary 11
Palaeoceen 119
Danian 27
Upper Senonian 34
- 352-
The depth of the reservoirs was in the range 75-Z00 meter
below surface and 10-185 meter below sea level.
- 353-
solution of Si0 2 from the sediments is caused by amorphous
silica.
-354 -
Samples from the deepest part of the well showed no evi-
dence at all of surface contamination and as already mentio-
ned neither of marine infiltration.
According to the heavy metal contents remarkable high
concentrations were determined in reservoirs situated in
Upper Senonian near the border of the old sedimentation bas-
sin or at areas, where the deposition has taken place in
shallow water, Table II.
4. Test of geothermometers.
4.1 The silica-geothermometers
The application of the silica-geothermometers is based
on a known and uniforme mineralogical composition of the
aquifers. Further, consideration of equilibration kinetics,
hydrological retention time, and well enthalpy must be done.
Accordingly to the investigated reservoirs in this study it
is known that the enthalpy is insignificant due to low tempe-
rature and that the reservoir contains different species of
silica. Therefore, it might be expected that different silica-
thermometers must be used in different reservoirs governed
by the relative maturity of silica conversion to more stable
species.
Using chalcedony as equilibrium solid the predicted tem-
peratures were in the range 320 to 375 oK. As the aquifers
presumable still contain amorphous silica a modified geother-
mometer is set up.
-355 -
INVESTIGATED WELLS IN DENMARK
- 356-
SALINITY
• \I .. ..
-.
-.
-. E
1--0
- I
- ..
c
-"
Fig. 2. Saline wells, salinity vs. 6 18 0 • Vertical hatch area:
lakes and riwer water, horizontal hatch area: sea water and
brackish water, not hatch area: recent ground water in Den-
mark. Suggested processes: E: evaporation, C: cold period ef-
fect and D: dissolution of fossil salts.
./
350
.. ..-
.
300
..
.. .
0
.
:/: . 0
...~/ 0
. . ..
~ ... < . .
,~ ~
.
11
"
250 "
250 300 TEMP oK 350
•
/
V
•
/
/
/
V·
•
~
...
•
~
v~' •
. . .
••
;f ~
'•"
- 358-
geothermal waters exceeding lSOoC in non-sedimentary aquifers.
-359-
GEOCHEMISTRY OF THE~L WATER IN THE MJNT-DORE AREA
SUDDIlary
The thermal area of Dordogne valley springs constitute the most interes-
ting zone for geothermal purposes.
On these waters, confident temperature determination can be made (~ 150°).
Occurence of deep a water level at higher temperature (180°) is also
possible.
- 360-
Water geochemistry studies have been used in the Mont-Dore geothermal
project to get all possible informations about geothermal fluids of this
area. This includes deep temperatures, homogeneity or heterogeneity of sup-
posed deep reservoirs, fluids and solutes origin and circulation paths.
Twenty four springs were sampled in October 1978. Locations of springs
are given in f!g~~_!. Each spring was analysed for major and minor ions,
and isotopic compositions. Chemical and isotopic analyses of the gaseous
phase were also carried out. Tables listing the complete analytical results
can be found in the paper of GERARD and al (1980), presented at the same
symposium. In this present work are only given the useful data for interpre-
tation ; a detailed study of gas analyses has been already published (LELEU,
DEGRANGES, 1979).
At first look, waters of the area exhibit great differencies in their
chemical composition. As for their isotopic behaviour, none shows a 018 0
geothermal shift (f!~~~_?2 and the domain of variation is too small to dis-
tinguish different aquifers. On the contrary, the isotopic composition of
sulfur in dissolved sulfate displays different fields. Using a 034S-f[S04=]
plot to avoid possible dilutions effects (f!g~~~_~), shows that the Mont-
Dore city waters differ from the Dordogne valley waters.
Considering these chemical and isotopic variations we have split the
samples into four groups :
I Cold, CO 2 rich, springs
II Dordogne valley waters
III Mont Dore city springs
IV Chaudefour valley springs and sulfurated springs of Puy de Sancy.
-362 -
TABLE I CHEMICAL AND ISOTOPIC ANALYSES
pH Na K Ca Mg Li CI
- - S04 = Na/ Ca 60 6 14 0 6'"S T
GROUP Name RH. t HCJl 3 Si0 2 Clllco3 - C11 so.
---------- -- -- -- -- -- -- ---- ._------ ------ ------ ------ ----- ------ ------ ------ ------ ------ ------ ------ ------- ------ ------ ----- ------ ----- - -----
-------
Pa'iadi raux MD 1 8,5 5,38 0,6] 0,24 1,25 1,47 0,003 0,096 7,06 11.0. 1,60 0,013 0,53 - 59,S - 9,3 + 4,5 8
0,93 0,006 0,081 5,47 N.D. 1,27 0,015 0,79 58,8 - 9,3 + 8,9 23
--
Merdan~on 110 2 7,5 5,28 0,96 0,32 1,21
fontpique MD 4 10,5 6,16 5,86 1,28 2,87 3,39 0,118 0,203 1,92 0,10 1,70 0,011 2 2,04 - 61,6 - 9,8 +14,8 4
Le Vernet MD 7 10 5,30 0,96 0,21 0,70 0,93 0,003 0,068 3,82 0,07 1,27 0,018 1 1,37 - 59,9 - 9,1 - "I
I Goyon MD 8 7 5,80 0,60 0,21 2,0 1,82 0,007 0,101 8,46 0,03 1,10 0,012 3,3 0,3 - 59,S - 9,5 + 7,2 19
le Boutelx MD 15 8,5 5,86 0,51 0,11 0,78 0,84 0,003 0,56 3,70 0,28 1. 70 0,151 2,0 0,65 - 61,0 - 9,2 + 3,2 6
Col lange MD 26 12,2 7,20 7,91 1,13 2,53 4,23 0,18 0,21 124,3 N.D. 1,67 0,009 - 3,1 - 59,2 - 9,3 - 4
Reboisson MD 27 11 5,62 7,96 1,12 1,03 1,73 0,13 0,41 16,5 0,05 1,37 0,025 8,2 7,7 - 62,3 - 9,8 +15,0 13
Moneyron MO 12 IS,S 5,86 66,9 1,75 2,39 1,25 0,89 58,0 ~4,8 0,92 1,47 3,92 63 28 - 61,1 - 8,9 +10,0 2
felix MD 13 30,4 6,32 64,7 2,22 2,72 2,11 0,79 45,0 7,6 0,96 2,06 1,63 46,8 23,7 - 63,2 - 9,4 + 7,7 3
Croizat MD 14 39 6,52 113,9 3,75 3,67 1,32 1,80 85,S g~,7 2,25 1,73 2,53 38 31 - 61,8 - 8,9 +11,7 6
Clemence MD 17 18,5 5,87 37,2 2,03 2,09 1,31 0,36 25,6 2,1 I,D 1,70 1,15 25,6 17,8 - 60,4 - 9,1 + 8,0 3
II ';1
La Verniere MD 19 12,5 5,64 35,7 1,71 1,26 6,98 0,40 16,2 ~5,3 1,08 1,75 1,05 16,0 28,3 - 61,7 - 9,0 + 9,3
Choussy MD 24 53,2 6,48 76,2 2,42 0,87 0,37 0,84 48,S 7,3 1,33 1,91 1,77 36 87 - 61.2 - 9,0 +12,7 -"'I
Fenestre MD 25 16 6,36 4,8 0,31 0,15 0,18 0,063 2,64 3,11 0,25 1,25 0,85 10,5 32 - 58,3 - 8,6 + 4,7 33
.... MO 21 40 6,18 14,9 I,ll 2,07 1,49 0,17 6,11 15,3 0,37 3,0 0,46 16,5 7,2 - 62,3 - 9,5 +16,3 8
[ St Jean
....'" III Chanteurs MD 23 44,2 6,18 14,7 1,08 2,05 1,46 0,17 5,70 15,0 0,37 3,08 0,38 15,4 7,2 - 62,4 - 9,2 +15,6 13
I
Chaudefour 1 M05 10,2 5,02 0,65 0,21 0,52 0,21 0,0015 0,1 2,52 0,10 1,7 0,04 1 1,25 - 61,0 - 9,5 + 6,2 50
Chaudefour 2 MD 6 11,5 5,08 0,68 0,20 0,52 0,20 0,003 0,1 2,62 0,10 1,6 0,04 1 1,30 - 61,6 - 9,2 + 7,8 48
Chaudefour 3 MD 30 22,7 5,74 12,4 0,91 4,13 1,67 0,547 0,291 18,4 2,81 2,03 0,016 0,1 3,0 - 61,1 - 9,8 +14·,4 6
Chaudefour 4 MD 28 23,S 5,62 10,4 0,80 3,55 1,44 0,490 0,291 16,1 2,29 1,97 0,018 0,13 2,9 - 61,1 - 9,7 +14,8 3
IV 0,21 2,17 - 61,1 - 9,7 +19,6 6
Chaudefour 5 MD 29 16 5,36 5,4 0,61 2,49 1,32 0,230 0,24 21,6 1,12 1,82 0,02
CO uze Chaudefour MD 31 5,5 6,34 0,21 0,05 0,13 0,08 0,002 0,06 - - 0,42 - - 1,6 - 59,8 - 9,4 + 2,4 53
Su 1fureuse Saney A I 4 - 0,86 0,14 0,79 0,52 0,010 0,84 N.D. 2,39 - - 0,35 1,09 - 61,2 - 9,4 +13,2 21
---~-.
I
I
.
I
I 4
0
I
I
, I
0
26
,,
"'''~)
'"I ... '"" 1
iJ''''
0
20 1§
27
a
0
0,
o ",roup!
• II
c ill
• rv
- 365-
~D 0
, •
0 ,I
0 0 ,
-60 ,'0
,
•
•• • .' , I
0
-62.0
,I
,
• ~
0 p 0
• ~IO
-10 -9
~
+20 TIl
® ~ . ---- -
• •
N
- - - .- - .-
-
•••
n
o
'Z. So~
~o 20 30 40 50
t ·c
- 366-
Thermometricaly, the Mont-Dore springs have long been considered as
witnesses of hot (200°C) waters. This is due to their high silica content
and their low Na/K ratios, both leading to high temperatures (!!~!~_!!!).
However some features of springs composition tend to moderate this opinion.
1. The silica content, very near to amorphous silica saturations could
signify equilibria with sinter at about fifty degrees.
2. Calcium and Magnesium are too high for equilibria computations at
thermometricaly calculated temperatures. This could either mean an
overestimate of temperature by cation geothermometers (silica beeing
ruled by sinter) or a secondary dissolution of Ca, Mg beeing
minerals. In this later case, the constancy of Na/K ratios is ques-
tionable.
3. The Na/Li goethermometer gives lower temperatures (150°) than the
others.
In summary, we cannot assert that deep temperatures in Mont-Dore area
are as high as 200°, consequently, we do not consider that this zone is more
interesting for geothermal further work than those of the Dordogne valley.
5 - BIBLIOGRAPHY
DESGRANGES P., LELEU M. (1980) - Etude des gaz dissous dans les eaux ther-
mominerales : le cas des eaux de la region du Mont-Dore.
C.R. Acad. Sci. Paris~ in ~ess.
- 368-
EVALUATION OF RESERVOIR TB1PERATURES IN TUSCANY BY APPLICATION OF A GAS-
GEarHEmDMETER
Summary
Analysis of the relationship between the main gases from many natural
thermal manifestations and from different geothsrmal boreholes has led to
the definition of a new geothermometer based on the analytical composition
of CO Z-HZS-HZ-CH4•
An attempt was made to apply this gas-geothermometer in the thermal
areas of central-southern Tuscany; the results have proved to be of consi-
derable interest for a regional-scale geothermal prospecting and were used
to compile a reliable temperature map for deep reservoirs.
- 369-
1. GAS-GEOTHERM:MITER
The chemical composition of gas mixtures emerging in thermal areas
can be used in evaluating the deep geothermal temperatures as a useful
tool in geothermal exploration. Chemical analyses of gas compositions for
34 geothermal systems were considered and an empirical relationship devel-
oped between relative concentrations of HZS-HZ-CH4~COZ contents and the
reservoir temperature (l).The evaluated temperatures can be expressed by:
o Z4775 CH4 HZ HZS
t C = ~+~ + 36.05 - Z73 where ~= Z log COz - 6 log COz - 3 log COz
(concentrations in %by volume) and ~ =-7 log PCO Z'
The geothermometer was defined by means of the following main steps:
- A suitable chemical equilibrium model had to be devised for the gas mix-
ture, basing it on reasonable hypotheses on the origin of the single spec-
ies.
- 370-
- A suitable empirical equation had to be evaluated to individuate oxygen
partial pressure in the system as a function of temperature.
has been assumed to give the 'actual' values of Po in natural systems and
Z
has been considered in our final computations.
-371-
The main source of doubt in using this geothermometer lies in the cor-
rect estimate -of the coefficient ~= -7 log POOZ' The reservoir Peoz can
only be evaluated with precision if we have a complete knowledge of the
minerals and parent fluid of the system at depth.
However, the chemical composition of the gas mixture can be used to
obtain a reasonable order of magnitude for ~. This is based on the hy-
pothesis that POO z is a locally fixed independent parameter. Carbon diox-
ide acts mainly as a dilutant and carrier of the other gas species. Then,
in the same thermal conditions, reservoirs with lower Poo produce a gas
Z
mixture with higher HZ,HZS, CH4 and NZ concentrations and the evaluated T
would consequently be higher than the actual value. Whereas HZ partial
pressure is dependent on T only, the HZS and CH4 partial pressures are al-
so chemically dependent on PCO so that the dilution effect is considerab-
ly different for the three speZies.
In dealing with a large number of gas analyses from different geo~
~
= + 7 if OOz < 75%
0 if OOz ~ 75%
- 7 if l)OOZ ~ 75%
Z)temperature computed by =0 is more than 150°C
3)CH4 > ZHZ
4)HZS > ZHZ
This evaluation of ~ should not be considered a rigid one. An improve-
ment of its computation is given by D'Amore and Panichi (Z).
- 372-
samples from 54 localities and from some typical geothermal wells in the
Larderello,Travale and Mt.Amiata fields, distributed over an area of about
8,000 km 2
It should be noted that at low temperatures (~70°C) the calculations
are often about 25% inaccurate. This uncertainty is due to the fact that
some gas species were not easy to individuate because of their very low
concentrations.
The calculated values were used to compile a temperature map of the
following ranges: below 75°C, between 75° and 100°C, between 100° and 150°C
and above 150°C (see Fig.2).
The gas composition was assumed to represent the equilibrium temper-
ature reached locally by the species in the top of the main reservoir con-
nected to the manifestations. This reservoir should generally correspond
to the permeable formations of a carbonate-anhydritic or phyllitic-quartz-
itic nature generally lying below a clayey-marly cover.
Due to the very low ratio between the number of gas manifestations and
areal extent studied the contour lines may obviously be approximative only.
Gas and temperature data from the deep wells drilled on the margins
of the geothermal fields and the structural data on the zones with the
limited number of manifestations were used: the same order of magnitude
of temperature was assumed for adjacent parts of a reservoir having similar
structural characteristics.
Note that the isothermal limit (150°C) in the map refers to the ther-
mal systems outside the Larderello and Mt.Amiata geothermal fields,which
show temperatures of 250°C.
Apart from confirming the presence of temperature anomalies tied to
the known geothermal fields, this study also individuated zones of interest
for a future development of geothermal research, i.e., the wide graben south
of Siena and that east of the Monticiano-Roccastrada horst,where the cal-
culated temperature in the deep reservoir was of the order of 140°-170°C.
The application of the gas-geothermometer to Tuscany's geothermal
area confirmed the validity of this method in that the temperature data
-373-
TEMPERATURE ·c
150
'"
0
"-
••
00
~
- 40
•
log P0 2 ~ 8. 20-23643/T I Kltl
-35
••
-30 •
1.8 • .0 2.2 2 .4
(liT K) 10'
Fig. I. Data of log Po . vs (\ TK~ and last square straight line. These data were calculated from gas
. composition and are reported in Table I.
- 374-
~T'C.'00·
~
~ 10G"c T'C .. 'SO·
~71'4T"C4'tOO'
~T"CC75'
I I ..... gas from well
-I
m
z.
o
References
(1) F.D'Amore and C.Panichi (1980) Evaluation of deep temperatures of hy-
drothermal systems by a new gas-geothermometer. Geochim. et Cosmochim.
Acta, in press.
(2) F.D'Amore and C.Panichi (1980) Geothermal reservoir temperatures of the
hydrothermal systems of central Italy calculated by the H2-H2S-CH4
geothermometer. 3rd Int.Symposium Water-Rock Interaction, Edmonton,
Canada, in press.
(3) T.N.Kozintseva (1964) Solubility of hydrogen sulfide in water at ele-
vated temperatures. Geokhimiya~, 758-765.
Table I. Oata of log Po, calculated from gas composition for some geotnermal fields of different temperature and Pm,
values .
LOCALI'IY t(reservoir) P(OJ2) log POz (HZS) log P02(CH4) log P0 2 (H2)
·c aon
Equations used to compute log Po. from the H,S CO,. CRt/CO"~ H,,'CO, ratios at a gIven temperature and P,o ..
log Po,(H,S) = 12.5 - 26888 T - 97 log (PH,s:Pm ,) + 6·7 log Pm, .
log Po,(CH.) = 14.8 - 24312 T - 2.4 log T - 2;3 log (PCH " P co ,) - I 3 log Pea,
log Po,IH,) = 16.2 - 29592 T - 2 log (PH, Pm,) - 2 log Pco,
(I) Average 4 wells (NEHRING and FAl·STO. '1978\. .
i2I TRt;ESD~LL and MANON. 1978.
(3) Computed from data of O'AMORE and TRlJESDELL (1979).
(4) Average 6 wells (ARNORSSON. 1970).
(51 ARNORSSOl' et al .. 1974.
(6) Well VC 10.
(7) Average 12 wells IBROWSE and ELLIs. 1970).
(8) TRUESDELL and FRYE 11977).
19) ElliS (19701.
(10) Avera2e wells (ELUs. 1970).
(11) GLOVER (1970).
(12) Well Fabiani.
(13) Average 6 wells iD'AMORE and TRUESDELL. 1979).
il4) LIPMAN et al. (1978).
1151 Computed from data of O'AMORE and TRl·F.SDELL 11979).
(161 Well 230.
tl7) Well 8.
1181 Well R4.
119) Well 2bis.
1201 Well 7.
- 376-
TAllLETI
Locality Date n' Gas CODqlOsition (lOOles \) Teale
(lJ2 H2S H2 014 N2
,
Bagno Vignoni-S. ~irieo d'Orcia) (SI) 77 G47 97.18 2.82 4 76
Baoni Chiecinella -Pa1aia (PI) 65 G48 97.95 0.01(-) 0.823 1.23 ~ 50
Pergine-Arezzo (..,11 12) (AR) 65 G49 98.14 0.24 0.58 54
Cura Sigliano-Arezzo (AR) 65 G50 96.21 0.01(-) 0.18 3.61 ~ 55
Levane-Arezzo (AR) G51 99.38 0.01(-) 0.034 0.58 4 61
Itmtespertoli (FI) 65 G52 91.82 1.04 7.14 .$ 49
*
I.e Fonti Poggibonsi (SI) 65 GS3 95.02 0.56 4.42 51
Baccanella·-Palaia (PI) 65 G54 93.8 0.01 0.01 1.88 3.95 ~8O
- 377-
FLUID-ROCK INTERACTION IN GBOTHERMAL ENERGY RECOVERY
B. Althaus
Summary
Reaction of aqueous fluids with minerals of geothermal energy
reservoir rocks produces both secondary solid phases and more
or less concentrated solutions. Kind, amount, and rate of
reaction are governed by the degree of deviation from
equilibrium concentrations. Ionic ratios (metal-to-hydrogen)
are the most important parameters for these effects. Heat
production from chemical energy is possible in relevant
amounts, but involves production of dissolved material.
Conditions of optimal operation have to be evaluated individu-
ally for each rock system. Ionic concentration and PH can be
used to determine the reactivity of sdlutions.
- 378-
1. INTRODUCTION
In both natural and man-made geothermal systems water is the
most important medium for heat transport. In a dynamic
system water is neither at thermal nor at chemical nor at
mineralogical equilibrium with the solid phases constituing
the heat reservoir. Disequilibrium generally causes reactions
to occur; in an aqueous system, several types of reaction
are possible: The most simple type is plain stoichiometric
dissolution, including changes in kind, amount, and
composition of solids and liquid.
-379-
2. KIND a, REACTION
The common type of reaction between minerals and fluids is
triggered by hydrogen ion exchange. A schematic reaction
equation can be written as
(K,Na,Ca,Fe,Mg,Al,Si)-mineral + H+-ions
-.:. K+ ,Na + ,ea 2+ ,Fe 2+ ,Mg 2+ ,AI 3+.
_ -lons + H4Si0 4
+ secondary minerals
- 380-
values secondary minerals are precipitated from the
solution whose kind (aluminium hydroxide, aluminium silicate,
quartz, mica) again is determined by the ionic ratio.
3.REACTION RATES
Equilibrium conditions in complex systems are determined by
manifold complex compositional parameters. For a simplified
feldspar alteration reaction, the most important chemical
variables are the ratios ~, ~, and silica activity. Provided
that the latter is fixed at a certain value (e.g. quartz or
"amorphous silica" saturation), the former two determine the
equilibrium conditions of coexistence. Data have been
calculated by Helgeson et al.(l). Reaction rates for the
alteration of minerals in this system depend on the extent
of deviation from equilibrium conditions. Reactions have been
found to be the faster the greater is the departure from
equilibrium ratios. Absolute concentrations are not as im-
portant as these parameters.
- 381 -
The concentration of dissolved materials in a solution in
contact with a mineral at disequilibrium changes in a
definite manner: first, it increases rather quickly but then,
as equilibrium concentration is approached, increase becomes
less strong. In a closed system at temperatures between
1200 and 2S0 oC equilibrium is reached within a coupe of days
(2 to 10). In an open system in which the transport fluid
is moving equilibrium usually is not attained (fig. 1). It
depends on the ratio between the velocity of fluid replace-
ment (i.e. flow rate) and reaction rate how closely
equilibrium conditions are approached. If reaction rate is
great compared to flow rate, concentrations are closer to
equilibrium values than in the case of high flow/low reaction
rates. This is a very important difference. In the former case
the contact area is of very little importance for the total
concentration of solutions and for the amount of altered
material whereas in the latter the extent of reaction
definitely is dependent upon the area. It has been found
experimentally that feldspar reactions are of the "area
dependent" type but reactions with olivine and diopside are
"concentration dependent" (Kronimus, 2).
4. ENERGY EFFECTS
Many reactions can occur between rocks and fluids. Those
including feldspar, diopside, and olivine are - with few
exceptions - generally exothermal. This chemical energy can
be added to the physical energy obtainable from cooling down
the reservoir. The rate of exploitment of chemical energy
is directly dependent on the reaction rate. For high efficien-
cy high reaction rates are necessary. Energy production is
determined, therefore, by fluid composition. The criteria
mentioned in the preceding section for area or concentration
dependent reactions hold true for energy production, too.
- 382-
s. MINERAL DEPOSITION: SUBSURFACE ROCK COATING AND SCALING
From solutions formed by primary attack of convection fluids
secondary minerals can be precipitated for which the solutiom
are supersaturated. Under static conditions equilibrium phase
associations are likely to form, but not in a dynamic system.
Products of deposition are, in these cases, dependent more
strongly on nucleation than on stability. Metastable formation
of gibbsite in the presence of quartz and of muscovite in
the kaolinite stability field have been observed. The ionic
ratios mentioned before again are of primary importance. They
determine the kind of a newly formed mineral as well as its
amount. The experimental investigations habe shown that at
low PH metastable phases are more likely to occur than at
neutral. The amount of precipitates is determined by
variations in composition and temperature.
6. EXPERIMENTAL RESULTS
,
Experiments have been performed on alkali feldspar, sanidine,
olivine, diopside, and samples (small drill cores) from
basement rocks of the Urach borehole. The mode of reaction
of the minerals was very similar: At flow rates (simulated by
exchange of solutions in equal intervals) higher than reaction
rates congruent dissolution prevailed (fig. Z, typical
example selected from many different run series). Dissolution
was faster if departure from equilibrium was greater. Since
ionic concentrations varied only within 1 order of magnitude,
H+ concentrations, however, within about 5, PH is a sensitive
measure to determine deviation from equilibrium. Its influence
on the amount of dissolved material is shown in fig. 3:
increasing acidity drastically increases reaction rates.
-383-
It can be concluded that at 2000 C reservoir temperature a
feldspar rock with 1 km 2 surface area yields more than
6000 kWh/day with a fluid acidity equal to 0.1 n Hel at a
flow rate of about 850 m3/h. This energy production is,
however, coupled with the production of vast amounts of
dissolved matter (fig. 4); the load added to the transport
fluid in the latter case increases to about 7S tons per day.
7. CONCLUSIONS
8. ACKNOWLEDGMENTS
- 384-
log [No-)
[ I"f+)
12?-------------------~
10
500
LowAlbite
8
0.1nHCI
5 1--_ _ _ _ _ _ _-( 200'C
l. No-Mont . 1.00
2
~,
,<' Fa M
o ~+ 9 K-Mont . us.
FILO \ 200
FLU
-2
Kaolinite
-l.+--+--+---+-~~--~~
-4 -2 o 2 458
[K+[
log iH+i 2 5 10 14 18 doys
10-3 moles Si O2
per mm 2·doys
75
50
50
30 200'C
10
25
2 3 4 5
- 385-
9. LITERATURE
(1) Helgeson, H.C., Brown, T.H. & Leeper, R.H.: Handbook
of Theoretical Activity Diagrams Depicting Chemical
Equilibria in Geologic Systems Involving an Aqueous Phase
at one Atm and 0 to 300oC. - Freeman, Cooper & Co., San
Francisco 1969
-386-
GEOTHERMAL APPLICATIONS OF THE GEOCHEMICAL STUDY OF
HOT SPRINGS IN EASTERN PYRENEES
- 387-
About 120 complete analyses of Pyrenean hot springs were performed by
the LGE during this contract. A first report on estimated temperature at
depth by classical geothermometers and smme specific approach was presen-
ted in 1977 (BOULEGUE et al.1977). Now. at the end of the study. we pre-
sent a synthetic report based especially on the Pyrenean results; but. we
still use results obtained in Corsica. Massif Central and comparisons
with data on different geothermal fields.
In our project. we proposed:
(i) to achieve a g~obal modelling of the chemistry of the springs.
(ii) to compare the behaviour of some trace elements with the
behaviour of major elements
(iii) to search for potential trace elements geothermometers
(iv) to attempt a new discussion of temperatures previously
obtained in Massif Central.
We shall briefly present each point.
1. GLOBAL MODELLING OF THE CHEMISTRY OF THE HOT WATERS FROM PYRENEES AND
CORSICA.
In our p~i~report. we show that all the hot springs are roughly
at equilibrium with kaolinite at emergence temperature. We proposed a
modelling of the cooling of the waters from the deep reservoir to the
surface at equilibrium with kaolinite. A first calculation method was given
which allows us to determine deep pH. then un-ionized silica concentration
and deep temperature. A more complete calculation method was in progress;
.it is now achieved: the saturation degree of the solution for many common
minerals can be calculated at each temperature (MICHARD et FOUILLAC.19BD)
The results are shown on figure 1. It is quite clear that the solution beco-
mes at equilibrium with kaolinite. quartz. albite. adularia. calcite and a
Ca alumino-silicate at the temperature previously estimated by silica and
sodium/potassium geothermometers. The same mineral association is present
in the reservoirs of all thermal stations. The variations in chemical
composition are related with differences in temperature and inputs of solu-
ble anions. The latter depends on the distance from the sea and can be con-
sidered as a rain input. We achieve a calculation process allowing the de-
termination of the concentrations of all species in the reservoir waters
from only two data: temperature and soluble anions concentration (A ).
-388-
log Q
-1
-2
50 100 150
-389 -
Table 1. Alkali ions concentrations in the different spas.
Spa Label nO of li Na K Rb Cs
springs 10- 5 10- 3 10- 5 10- 7 10- 7
~,
- 390-
2. RELATIONSHIPS BETWEEN TRACE ELEMENTS AND DEEP TEMPERATURE.
T (Fig. 2c), and less with (A-); Rb+ depends only on T (Fig. 2d); CS+
doesn't exhibit correlation neither with T nor with (A-).
For Pyrenean and Corsican waters Na/Li and Na/Rb can be regarded as
temperature functions as well as Na/K. This result suggests us to look at
Na/Li and Na/Rb ratios in many geothermal fields. This compilation leads
us to the definition of a new geothermometer: the Na/Li thermometer (FOUIL-
LAC and MICHARO. 1979, 1980).
Log Na/Li 1000/T - 0.38 for low salinity waters
The Na/Rb ratio at the contrary, offers no definite advantage over the
Na/K one.
The As concentrations exhibit a correlation with deep temperature,
but the springs of some area have As values much higher than the general
trend. Any attempt to derive a general As-T relationship has been unavai-
ling.
- 391-
N. LI
10- 1 ",'"
" A
®
•
[JJ
v · ro.-------
~~
~
~OO>A~ ~
~ D ~
Q
5
°
80 100 120
i. 100 1l.
Fig. 2a Fig. 2c
K
1O'"
" Ii!
to
IIil
O~~----~~------~IOO~--------±,~~~
Fig. 2b Fig. 2d
- 392-
3. TRACE ALKALI IONS-MAJOR ALKALI IONS RELATIONSHIPS.
A correlation study between (Li, Rb, Cs) and (Na, K) shows that (Rb+)
and (K+) on one hand, (Li+) and (Na+) on the other hand are very strongly
correlated.
+ +
For Rb and K (Fig. 3), the correlation is still higher if we consi-
der separately the three different regions: Eastern Pyrenees, Aude-Ariege
and Corsica. In the two former regions, the coefficients correlations are
respectively 0.99 and 0.999 and the line fits also the superficial waters
of the region. For Corsican waters, a strong correlation appears after cor-
rection of rain-brought potassium. Thus, in all cases, thermal waters keep
the memory of the incoming superficial waters. The relationships can be ex-
plained by a model involving superficial inputs, rock deposition and new
formed minerals deposition. We have thus an insight on the changes of the
rocks at contact with water (tentative applications on porosity euolution
can be considered). But we observe also a strong Na-Li correlation (Fig.4)
which seems not to be explained by this model. We need for a more complete
~nowledge on trace element behaviour during rock-water interaction.
The first attempt was done in Le Boulou and Aquacetosa, two CO 2 -rich
springs. The former one is very near the sulfurated springs of Amelie;
the latter one is near Pietrapola. Na/K and silica thermometers give simi-
lar temperature for Le Boulou and Amelie on one hand, for Aquacetosa and
Pietrapola on the other hand. In all case, Na/Li gives a fair estimation
of the deep temperature.
The Na/Li geothermometer has been still used for a discussion of the
geothermal interest of the Western boderline of the Limagne and the Saint-
Flour area. In Limagne, silica and Na/K indicate high temperature estimates,
but Na/K indicates in many cases unresonable high values. For all these
waters, Na/Li ratios give temperatures of about 200°C; the very high tem-
peratures obtained by Na/K geothermometer in some places can be attributed
to exchange of K with clay minerals (FOUILLAC et MICHARO, 1980). The inte-
rest of the Saint-Flour area has been also confirmed by the Na/Li value
(180 to 260°C). Recently, we observed, for the first time in France, a
strong shift of 180 on the Cl-rich water of Coren, 6 km North of St-Flour.
- 393-
fig. 3
U
I~'
A
fig. 4
-394-
PROPOSALS FOR FURTHER WORK.
REFERENCES.
-m-
TRACE ELEMENT GEOCHEMISTRY IN THERMAL WATERS FROM PLOMBIERES
AND BAINS (VOSGES)
Abstract
The thermal waters of Plombieres and Bains can be des-
cribed as mixtures of Sondage Neuf water (Plombieres) and cold
water.
The "soluble" elements (e.g. Li, Na, Rb, Cs, W, As, F)
are useful for calculating mixing ratios between the hottest
water of the region{Sondage Neuf) and cold water, both at
Plombieres and Bains.
A new quartz-scheelite geothermometer is described which
allows to calculate the fraction of hot water in a given
source, and at the same time its temperature in depth (before
mixing). Some limitations of this approach are discussed.
For a number of elements in Sondage Neuf water the con-
centration is controlled by the solubility of their (hydr)-
oxides (Sc, Ga, Zr, Th) or fluoride (La) at the temuerature of
emergence. The Mn, Fe, Co(?), Ni, Zn, Ge, As, Cd, inC?), Hg
and Pb content appears to be controlled by the solubility of
their sulfides at a temperature of about 100°C. At that
temperature the dissolved sulfide is oxidized to sulfate, so
that the concentration of these trace elements does not de-
crease further during cooling.
The suspended matter originates from the precipitation of
the above elements, accompanied by coprecipitation or adsorp-
tion of other elements.
- 396-
1. SAMPLES, ANALYTICAL TECHNIQUES AND CHEMICAL INFORMATION
-397-
From the contents of Li, Na, K, Rb, Cs, Si, S (sulfate), As, W
and F, one calculates that the fraction of the hot component
(I, II, III, VI) is 82 + 5 % in source VII, and 50 ±5 % in
sources IV and V.
The low Mg and tritium contents of sources I, II, III and
VI suggest that their waters are not mixed with cold water,
contrary to the conclusion of Tardy (1). The pure thermal
nature of sources I, II, III and VI also appears from the
following considerations.
Let us assume that Sondage Neuf (I) water is a mixture of
some unknown hot water and of cold water, then one can write,
for a soluble element,
Cmix = X,C hot + (1 - X). Ccold (eq.l )
where X is the fraction of the hot water (unknown);
Chot is the concentration of the element in the hot water
(unknown);
Ccold is assumed to be identical to the concentration
in surface water.
In order to determine X and Chot ' one obviously needs two
equations of type (1).
In the case of silica, Ccold = 5 ppm or 10- 4 . 1 M/I and
Cmix = 100 ppm or 10-2J8 M/I (Sondage Neuf) , hence:
10- 2 • 78 = X.[ Si hot ] + 10- 4 . 1 • (I-X) (moles/liter) (eq. 2)
Poss.ible solutions of eq. (2) are shown in Table III.
Also indicated are the temperatures in depth (column 3)
corresponding with the silica concentrations listed, using the
quartz solubility data of Fournier and Rowe (2).
If applied to tungstate, eq. (1) reduces to:
10- 5 . 9 = X. [Whot ] (moles/liter) (eq.3)
whereas for calcium one obtains
10- 3 • 84 = X. [Ca hot ] + 10- 4 . 22 . (I-X) (moles/liter)(eq.4)
The allowed combinations of [Whot ] , [Cahot ] and X are also
shown in Table III. Let us assume furthermore that, in depth,
there exists a chemical equilibrium between calcium and tung-
state ions in solution, and a solid phase CaW0 4 (scheelite).
- 398-
Then the ionic (molar) activity products [ca hot ] [Whot ] re-
present the solubility products of CaW0 4 (s) in depth (Table III,
co:lumn 6), and thus contain a temperature information (column
7). Contrary to silica, the scheelite solubility decreases
with increasing temperature, at least above 70-S00C, i.e.
above the temperature at the emergence. The temperatures in
column 7 of Table III were calculated using the most recent
scheelite solubility data of Atademir et al. (3), extrapolated
from 100 to 200°C taking enthalpy data of Naumov (4).
-402-
Table I: Sources sampled in the Vosges (France)
Plombieres les Bains Bains les Bains
I Sondage Neuf, 67°C, pH 8.1 IX Robinet de Fer 48°C, pH 7.1
II Romaine 59°C, pH 8.0 X Arteria 49.5°C,pH 7.4
III Stanislas 58°C, pH 8.0
IV Savoneuse 9 47°C, pH 7.7 XI Feconde 48°C, pH 7.3
V Jutier 33 , 5 °C, pH 7. 4 XII Casquin 49°C, pH 7.3
VI Vauquelin 70°C, pH 8.1 XII Cendix (cold) 11°C, pH 6.0
VII Ste. Catherine 63°C, pH 7.9
VIII Augronne (river) 11°C,
pH 6.1
- 403-
Table I I ( continued)
Element Plombieres Plombieres Bains (IX, X,
(I,II,III,VI) (IV, V) XI, XII)
Mo - 7.49 - 7.50 ...- 8.3
Ag (- 7.56) ...- 9.3 (- 7.7)
Cd - 6.88 ...- 9.05 - 7.00
Sn - 6.96 n.d. - 7.25
Sb ...- 9.4 n.d. -
... 9.4
I <;- 7.1 ...- 7.1 .....
7.1
Cs. - 6.09 - 6.42 - 6.28
Ba .. 6.05 6.29 - 6.31
La - 8.44 - 8.33 - 8.48
Ce ...- 9.15 -
... 9.15 - 8.69
Ta ...- 10.0 ...-10.0 - 9.35
W - 5.92 - 6.27 - 6.55
Au ...- 11 - 10.1 ...- 11
Hg
Pb
- 8.04
- 9.04
- 8.47
n.d.
...-
-
8.6
9.00
-404-
Table IV: Trace elements in Plombi~res thermal waters (sources I, II, III, VI).
Main species in solution at ca. 25°C, pH = 8.1, and insoluble compound controlling
trace element content, with corresponding saturation index n, i.e. experimental
total concentration divided by calculated total concentration in equilibrium with
solid phase.
Hg(SH)~
180
VOrt
200 160~ , ~
140~-n~e~ ________ _~Q::
I
100 I
,
~
.....
~:, 1201- ~:
I
I
\
100 1097
I .
04<a-:";' 40 6'0 80 100 Si02 (mg/liter) I I • X
0.6 0.7 0.8 0.9 1.0
Fig • 1 r Arsenic, cesium and tungsten content Fi~~ 2 : Derivation of the temperature in depth of
of Plombi~res thermal waters as a function of Son age Neuf \.ater (source I, Plombieres).
silica content. x represents the fraction of pure thermal water
A average of sources I, II, III, VI; (see text and table III)
B source VII; C ~ source IV;
D source V;
E sample VIII (river water)
A
I F( :100)
·~
~ IJgli Cs
]
100
~
~ /j N,"'OOOI
I~: 80
I
60
I
I 40
o
.
CO
I
I
20
. ~~
1<f1 I I , ' I ,I
10"" Icf 20 40 60
(Cal (moles/liter) 80 T.",erg. ('C)
Fig. 3 : Solubility lines of precipitated Cawo4 Fig. 4 : Lithium, sodium, cesium, arseniC, fluorine
at various temperatures. and tungsten content of Plombieres source I (A),
A = average Wand Ca values for sources I, II, III of Bains sources IX to XII (B) and of cold water (C),
and VI; B = source VII; C = average values as a function of temperature at the emergence.
for sources IV and V (all at Plombieres) ;
D average value for sources IX to XII (at Bains);
E = average value for fourteen sources at Amelie
(Eastern Pyrenees);
P, Q, R: see text
REFERENCES
-409 -
GEOCHEMICAL SAMPLING AND ANALYSIS OF GEOTHERMAL FLUIDS
IN SEDIMENTARY BASINS
W.M. EDMUNDS*, W.G. BURGESS*, A.H. BATH*, D.L. MILES* and J.N. ANDREWSt
Summary
Techniques are described for the recovery of formation waters from
deep sedimentary aquifers which enable the maximum information to be
obtained from individual boreholes drilled for geothermal exploration.
The scale of contamination involved in two recovery methods is illustrated
and it is suggested that most reliable samples are obtained by centrifu-
gation of interstitial water from core material. Methods used in the
analysis of small volumes are listed and the need for modification of
methods usually used for dilute solutions is stressed, and their limita-
tions illustrated. A method designed to determine the ~He content of
interstitial water from recovered cores, the problems of oxygen and
hydrogen stable isotope analysis of brines, are discussed. Results from
exploration boreholes in the U.K. are used as illustrations.
-410-
1. INTRODUCTION
Drilling has been carried out by rotary mud flush; core recovery using
a 20 m barrel has usually beer. high. To check on mud filtrate invasion of
the cores, a lithium tracer has been added to the mud at a level approxi-
mately 20x the expected lithium concentration in the formation water. The
tracer and pore fluid levels for one exploration borehole in the Wessex
Basin in which five aquifers were investigated are shown in Figure 1.
Background levels of Li+ for the formation waters may realistically be
taken as the lowest observed level of concentration, substantiated by
regional data. The scatter towards higher values shown is due to the
variable amount of contamination by drilling-mud filtrate: less than 3%
for the Chalk, 5% for the Greensand and up to 50% for the Triassic Sand-
stone. In the case of the Upper Lias Sands, the uniform level is ambiguous;
either it is the true level of Li+ in the formation water, or the core has
been contaminated uniformly.
Account can therefore be taken of the contamination in all cases
- 411-
.. _.
I
. .. ••
.m".I
,,.. . . .
. ..
" ".fH
LowI' C"-11rI
(C'•• IC:.ouS)
Figure 1.
t
Lithium tracer used as a
. control on mud filtrate
.•. .
G, ... nund
invasion (Marchwood bore-
Up~.r
"'"
,...
(C'.'Ic.eoua)
-
IJrldpor I S,1ld1 formations
f
: (U~rL.i . . )
.-
. ..
"..
.,'0
Ttl.sale Safldslon••
I
•.. .. • ".
_JNI.",
~!:.-'
I
- .
L.ITHIUM (mg/I)
The overall scheme used for processing core material and interstitial
fluids is summarised in Figure 2. Fine grained lithologies are preferen-
tially sampled, and checks for mud filtrate invasion are made in the field
using a micro-pH electrode inserted into 2.5 mm diameter holes drilled
radially across the core . The core is then sub-sampled, waxed and shipped
to the laboratory for processing in the shortest practicable time. Experi-
mental studies were carried out both by the gravity drainage method (1)
and also using heavy immiscible liquids (2, 3).
Possible chemical fractionation during centrifugation has been examin-
ed for various lithologies and results for the Lower Chalk are given in
Figure 3. A progressive decrease in solute concentration with increasing
- 412-
. ."-,..
. .\ .
I
)
.. t
cvn 1)0 c-ol
.wroll 1 . Figure 2
IlIt.II:",.l.
1U-1.n ~eot'Mal.c.l
-114' .t...
~!lr.r.le. ' p,~rtl . .
WWI·"'T~ 10'
ll'l.~'01\
Generalised
scheme used
~1 .. - 'lf"""'.anCI
"Ill lei l.abor.. \.OI:y for core
---
processing and
--..... OOIot., ) e- .... 1110.
to C) e- I ...... IIIU
crvsh
fluid analysis
'h't.1MhtlU'.
':sn....'nt\lOll during the
U,K. geothermal
programme.
N:!IlOUl:
O....c.,. ;rI:-r.r.
atne ... IOflC'.t _lrUe
Na'
1100
1000
20
.. _ ..-......... ../
..
..
1('
16
"- .",,/.. Figure 3 •
16 a __ Chemical fractionation
observed at high
mg / I
Mg"
""-./a centrifugation speeds.
a
2 Results are for the
40 Lower Chalk; using
immiscible heavy
~.--./. liquid displacement.
20
7 a - a
Sr··
"-./•
5
160 0 0
CI -
o~ 0 .......... 0
/
150 0
o 20 40 60 80
per cent yield
- 413-
50
fO
40
50
40
II:
11130 II:
III
:I II
~
z ~ 30
o z
Z
<It
.... 20
VI
le .. ,
con/emln./ed
] semple
10 ,enl
con/emlne/ed
l .emple
Pr.~ur.
Chember _. c •
Po, .... Ie, compolilloni Semple '"
• T ,~,---- ;'0 i - - -1 8
pH pH
10 9 8 7
LI ~~ ~ ~ ~
70 80 50 40 30
Sr 1;0 1iKt 0'0 ,~ 7~ eb sb .'0 30
-415 -
Table I. Comparison of major element results for interstitial and DST
fluids from the same interval.
Wessex Basin
An trim Basi n
worcester Basin
Kempsey: Triassic Sandstone
Inters ti ti a1 936m 7.35 1320 20 500 140 420 3130
DST 936-942 m 8.05 1800 80 340 90 820 2840
6. FLUID ANALYSIS
-416 -
appropriate to the analysis of potable waters cannot be applied directly
to more saline fluids without some degree of modification; the poor quality
of many geothermal water analyses has been well illustrated by a recent
international collaborative study (7). These errors stem, most probably,
not from an incorrect choice of analytical methods, but from a lack of
appreciation of the limitations of those methods and the complex nature of
the samples. The effect of the presence of a saline matrix can be so
severe, in some cases, as to degrade the analytical signal to a point where
it is impossible to estimate with any degree of precision the level of
analyte present in solution; this is illustrated in Figure 5, using the
example of barium in a Triassic Sandstone DST water.
Li 0.2 5.0
Na 25.0 - 400.0
K 0.5 5.0
Rb 0.03 - 3.0
Mg 0.5 6.0
Ca Flame atomic abosrption spectroscopy. 10.0 40.0
Sr Samples for alkali and alkaline earth 0.05 - 1.5
Ba determinations are typically diluted 0.1 1.0
Fe up to 200-fold and ionisation buffers 0.05 - 2.0
Mn added to both samples and standards to 0.02 - 0.5
Ni minimise interference effects. 0.1 2.0
eu 0.05 - 1.0
Cd 0.05 - 1.0
Pb 0.2 1.0
Zn 0.1 1.0
Cl 2.5 50.0
50 4 Automated colorimetry 5.0 - 200.0
I 0.01 - 0.06
Si Colorimetry 0.5 4.0
Br Neutron activation 0.5 40.0
F Ion-selective electrode 0.05 - 10.0
HC0 3 Titrimetry 100.0 - 400.0
- 417-
J
- - ; - Ii! ~hermal 1------- Figure 5.
I wol.,
I
_~ Sa __ Recorder trace showing the
dramatic increase in noise
-l i __553~6nm levels in atomic absorption
signals for the determination
-,J--
I
of barium in a Triassic
I -- -
1______ ._
Sandstone geothermal water,
both undiluted and diluted
tenfold, compared with those
---
It j
from normal calibration
- 4---ut. -, 2-01191
_ l
solutions. A nitrous oxide-
acetylene flame was used
and the calibration solution
-r ~
I contained 2000 mg/l potassium
_ IL to suppress ionisation.
I
1-0 mg t l
f-
- - - - -+11-1-1-------- - - - - ---
Analytical methods are therefore also being developed to determine the ~He
content of interstitial fluids.
Sections of core are selected after contamination checks (Figure 2)
and are coated with paraffin wax at the drilling site as soon as possible
to avoid diffusive loss of ~He and evaporative loss of fluids. In the
laboratory, central sections of these waxed cores are broken out, weighed
and then re-coated with wax. These are placed in a vacuum extraction
system, cooled to -800 e
with a solid e0 2/ethanol bath and the system is
then pumped to better than 10- 2 mbar in less than 45 seconds. This pres-
sure is sufficiently low to remove all but 0.001% of the air surrounding
the core and contamination by atmospheric ~He is then negligible. More
prolonged pumping is avoided as it may extract ~He from the core. A meter-
ed volume of 3 He tracer is then admitted to the extraction apparatus and
the core is heated to 2000 e for 45 minutes to release ~He. Any residual
nitrogen and oxygen are adsorbed on a titanium getter at 800 0 e and argon
is removed by adsorption on a charcoal trap cooled by liquid nitrogen.
-418 -
The remaining gases are analysed mass spectrometrica11y and the 4He relea-
sed from the core quantitatively determined by isotope dilution.
Core analysis may yield low 4He contents as a result of diffusive
loss of helium during sample preparation or of invasion by drilling fluids
during core cutting. Generally, several samples must be analysed and the
maximum observed 4 He content is taken as representative of the formation
at that depth.
The maximum 4 He contents of two core samples are compared with the
results of corresponding DST fluid analysis in Table III. The results for
these two intervals show good agreement, although taken individually the
core or DST results may sometimes be anomalous within a single borehole.
Although opportunity to compare DST fluid analysis with core analysis has
been very limited to date, further development of the technique is in pro-
gress to assess the use of 4 He profiles in geothermal studies.
- 419-
b1.o
f-'-,,·· Figure 6
f Isotope fractionation
-,u -1-0 effects observed at high
..
-
speed centrifugation in
samples of Upper Chalk .
..
-
-so
o 20 40 00 00
per cent yield
-420 -
Figure 7
Comparison of stable
isotble results for
interstitial waters and
DSTs for the same horizon;
data all from Permo-
Triassic sandstones •
• CST
o COre
0.5"1..'" scale
REFERENCES
(1) Edmunds, W.M. and Bath, A.H. Environ. Sci. Technol, 10 (1976),
467-472.
(2) Kinniburgh, D. personal communication.
(3) Batley, G.E. and Giles, M.S. Wat. Res., 13 (1979), 879-886.
(4) Riecke, H.H. and Chilingarian, G.V. Compaction of argillaceous
sediments. Ebevier. 1974.
(5) Miles, D.L. Unpublished data.
(6) Cook, J.M. and Miles, D.L. Methods for the analysis of groundwater.
Rep. Ser. Inst. Geol. Sci. London (In press)
(7) Ellis, A.J. Geochim. Cosmochim. Acta, 40 (1976), 1359-1374.
(8) Sofer, z. and Gat, J.R. Earth Planet Sci. Lett. 15 (1972), 232-238.
(9) Andrews, J.N. 1977. Radiogenic and inert gases in groundwaters.
Proc. Symp. water Rock Interaction, Strasbourg.
p. 1334-1342.
-421 -
INVESTIGATION OF THERMAL AND MINERAL SPRINGS
IN THE EIFEL MOUNTAINS REGARDING GEOTHERMAL INDICATIONS
Summary
As a preliminary study for planned geothermal investiga-
tions in the Eifel mountains the hydrogeochemical data pub-
lished until now were preliminarily interpreted. By applying
discriminant analysis four hydrogeochemical groups of springs
could be separated. Three of them represent independent types
of mineral water whereas the last group represents a type of
mixed water. The best separation in discriminant analysis was
obtained by the HCOl-content, the S04-content, both in mval %,
and the Ca/Mg-ratio. Within the different groups those springs
were pOinted out which show most clearly the character of the
respective group.
To obtain a first idea of the geothermal situation in the
investigated area Si0 2 - and Na-K-Ca-geotemperatures were cal-
culated. Areas with higher geotemperatures were found in the
following regions: the region of Aachen, the lower part of the
Ahr valley and the area of Daun-Dreis-Gerolstein.
-422-
1. INTRODUCTION
2. CLASSIFICATION OF SPRINGS
In order to recognize the chemical relationship between
the different springs a PIPER-diagram was drawn for the anions
(fig.1). According to their regional accurrence the waters are
represented by different symbols.
The classification of the springs according to their
location correspond only partly with their chemical character.
Especially the springs form the WESTEIFEL region show non-
uniform compositions of their anions.
-423-
By applying discriminant analysis four hydrogeochemical
groups of springs could be separated (fig.II). Three of them
represent independent types of mineral water:
-424-
• ARDENNEN
~ WEST - EIFE L
A AHRTAL u SIEBENGEBIRGE
~ ~
----------------HC0 3---------------.·- FIG.I
3. APPLICATION OF GEOTHERMOMETERS
- 425-
..
, ,
..
DISCRIMINANT ANALYSIS
,.
, .. - f.-
• chloride water in the region
of Aachen and Cologne
..
'"
160
• 79
, .- .. a
o
_
oo ~
--....
......
• bicarbonate water in
the larger Eifel and
Ardenne region
.
:0 ~
-
00 :0
2 -S2 7 - - f - ... ~ o~o _
... 0
. districts of the Ahr volley. the
Siebengeb, athe Neuwied bosir
~ • 0 0
"8. 161 2 - ,
•
01--- i- • sulfate water ofthe Permian
C basin of Wittlich
0-271 7 _
C
~ f-- i- f--
o • gravity centre of the groups
- -
U
f- -
.
N -ll' 2
.494 e - - -- . - f-
• -
' I-- - - f- -
•
.'I!) el - -
' f- t -- 1- - -
-
---- I -
.
0
0
.. " ' I-- - - - -
_ el6J . 66O!t _ 4'" . l29O
1 c anonIcal vorloble
.1 'll 002' 16&' FIG. II
- 426-
Tab.1: Emergence temperatures and calculated subsurface
temperatures of selected springs
temperatures (OC)
location measured calculated
Temerg. TSi02 TNa-K-Ca
Corneliusquelle Aachen 45,0 - 175,4
Kaiser Quelle Aachen 53,9 114,7 176,0
Nuerburgbrunnen Dreis 13,5 116,8 203,7
Brunnen 3 Gerolstein 11,0 - 248,3
Ludgerus Sprudel Bad Breisig 28,0 - 227,9
Angelikaquelle Bad Tonisstein 12,0 78,0 183,5
Burgquelle Plaid 12,8 86,1 175,3
Vulkania Heilquelle
Dreiser Weiher - 130,6 177 ,6
Nuerburg Quelle
Dreiser Weiher - 125,1 185,0
Gisela-Quelle Daun - 111 ,0 144,7
REF ERE N C E S
-427 -
Some Remarks on the Application of Geochemical
Techniques in geothermal exploration
F.B. Tonani
Introduction
The present paper is intended to furnish an overview of the original
contribution of geochemistry to geothermal exploration. As the consequence
of time constraints on lectures, a few subjects are highlighted:
geochemical thermometry, detection of geothermal steam leaking from
reservoirs, geochemical logging of drilling mud.
Pitfalls
The considered geochemical exploration process has been designed primarily
to avoid certain pitfalls, that appeared to be stumbled on frequently.
Some of them will be discussed. However, one has to realize that informal
records must be depended upon most, if the historical process leading to
incorrect predictions has to be analyzed.
Pitfalls that have been come on most are, e.g., the notion that chloride,
or boric acid, had some close magmatic source, or the fact that low Na/K
ratios may result from factors other than high temperature. Similarly, wells
have been drilled simply on patches of high near surface temperature
gradient, and/or electric low resistivity. These conditions can be
associated with geothermal resource, but they result from other factors
as well.
-428-
Most case histories appear to have some common feature, regardless of which
specific indicator is considered.
Figure 1 depicts the general case. The set of a priori possibilities and
SOme distribution of probability must be defined. Correlation describes how
likely entailed consequences are, e.g. how likely is observing the selected
indicator given the geothermal resource, or, for that matter, given any
plausible effective natural factor.
Thus, considering one specific indicator and one specific set of natural
factors correlated with it, involves defining the exploration target. It is,
in fact, an operational definition of the target based on the concept of
"anomaly", regardless of whether we give also an explicit definition of it
or we don't. Exploration will be as effective as the implicit definition of
geothermal resource comes close to the real resource. Each specific indicator
points to the complementary set of the set of conditions or factors F1, F2•••
F , inasmuch as it appears "anomalous", and that wi II be the operation-
n
defined target. Such operation-defined targets depend on the overall
characteristics of the study area, given the indicator. Whilst usually one
indicator implies targets only broadly related to geothermal resource,
combinations of indicators can be made to match the real resource fairly
closely.
The concepts represented by these terms have been basic steps in the
advancement of geochemistry. Nowadays, one may like saying geochemical
potential better than saying affinity. Nonetheless, the underlying concept
represents specifically some best effective way of investigating natural
systems, in between extreme reductionism to primary atomic properties and
sheer empirical descriptionism.
For example, the emanative process in the Earth's crust is specially relevant
to geothermal exploration, inasmuch as it involves hydrothermal systems and
volcanoes, as well as broad process of mass transfer that escape non-chemical
observation. Overall eman'ative process are involved in the 'short' geochemical
cycle of such elements as chlorine, and other halogens, mercury, hydrogen
(as both water and other hydrogen-bearing molecules), boron, nitrogen,
carbonium etc. Geochemical potential in mobile geo-phases as opposed to solid
phases describes chemical factors in the outward flow that keeps volatile
elements in the outer and cooler sections of the Earth. What we have, is some
broad, far-reaching view of precisely the same reactions that appear as
geochemical indicators, such as geochemical anomalies resulting from
leakages of geothermal steam and geochemical thermometers.
-432 -
I am suggesting that individual capability to make correct inferences and to
optimize decisions depends on the knowledge outlined above and on the
perception of reality that it entails. I suggest that this is part of the
original contribution of geochemistry to geothermal exploration, too.
Additional comments
As of twenty years ago, at the beginning of the current wave of geothermal
activity, scientific knowledge of hydrothermal and volcanic emanative
process was poor. Worse, it was dominated by prejudgement as opposed to
observation of facts. Among people offering theories, people who actually
observed thermal and volcanic emissions at one given time could be counted
On the fingers of one hand.
Under the circumstance, most widely accepted opinions were based on evidence
from solid materials and involved hardly any direct geochemical evidence.
- 433-
Geochemical thermometry.
I will limit this expose to well-tested geothermometers, such as silica and
metal ratios. Let me only mention en passant that, e.g., a belief that
isotope geothermometers do not depend on the same assumptions and problems
as other geochemical thermometers appears to be a major pitfall in their
application.
Factors and process must be understood in each specific case, and reducing
it to a list of assumptions applying to all possible cases only conceals the
usual pitfall, for factors to be considered must. be evaluated for the
specific sample under study •
..llii£!
Figure 3 depicts correlation between temperature and silica with reference
to different mineral assemblages.
Width of possible field measures a priori uncertainty, and evaluating which
solid phase has to be considered is nr. 1 problem in application.
Amorphous silica and quartz have been found to be in control in many known
cases. The former case occurs when fast precipitation is the controlling
reaction, whilst the latter case appears to occur when w~ter interact with
quartz-bearing rocks.
Figure 5 depicts the basic correlation between metaL ratios and temperature
(fuLL Lines), as obtained from the same data as reported by Fournier and
TruesdeLL in their cLassical 1973 paper. Data refer to rhyoLitic and more
generalLy, acid feLdspar rocks. Data regarding basaLts are insufficient for
empiricaLLy determining appropriate correLation lines. Tentative correlation
Lines have been calculated based on pLagiocLases being ideaL soLid soLutions
see dashed lines in Figure 5.
ConcLusive remarks
Comparing observed and calculated temperatures, and trying to reconcile them
as welL as chemical data, is of use in scientific studies aimed to better
understanding the overall correlation between temperature and chemicaL
composition of water.
The advantages of this approach are of different nature. In the first place,
-436 -
it may correspond to somewhat more reaListic definition of geothermaL
resource, inasmuch as it invoLves pressure as weLL as temperature.
Massive transport of boric acid with the steam phase can onLy occur at
fairLy high temperature, see Figure 6 reporting avaiLable data. Ordinate is
the distribution coefficient of boric acid in liquid and steam, abscissa
is reciprocaL absoLute temperature.
Resul ts
The experimental set up operated with grab samples, discrete automatic
analyses, manual transfer of data. Other than the keying-in of data the desk
top computer on board was not used much.
Features are rapidly evolving towards what is likeLy to be a commerciaL
out fit.
Unexpected process of mud contamination have been found to take place.
The most significant resuLts obtained with current techniques are depicted
in three graphs: 1) Content of clay in formation, 2) Apparent porosity,
3) Density of fractures.
-438 -
Estimates of enthaLpy of formation water, as weLL as some assessment of
permeabiLity, become more and more frequent as the technique is upgraded.
However, most current effort is aimed to making the technique cost-effective
i.e. to changing the outfit from an experimentaL set-up into an industrial
tooL.
-439 -
GEOTHERMAL
OTHER FACTORS
RESOURCE
~
G E 0 C HEM I CAL
I N D I CAT 0 R
FIG. 1
,k \~A1"~l~HE""R_ES"",="~___
x. ~f-l.,\"1 L -
8,F
- ~I
HETAI10RfHIC ___ _
, /' ROCKS \....
JUvElJ1{.e MATTER LOSSES TO MANTLE
FIG. Z
-440-
... ..
·.' +--____~---~---~----_--_
... >.0 3.'
T
I
__l
4 .0.10- 3
FIG . 3
/
5i0 2
mM/1l
/
4 .0
2 .0
/ ~-
O_~
- V-
:z',2:'!'k2
'" ~ ........
CORR£I. ... TIOff
~:~'"// .
'i t. _ _ .... ~"""
.-
i
o ~. _-- ",,\ . t 0 ~
~~
.~~~ sol'l· os, Co £. 0
---
60°_-- ,..",tt
c. " • 0
~
20
- ----------
80 80 100 120 ,~o c
FIG . 4
-441 -
833.3 _ 273 1096.7 _ 273
N.K t - 19 N./K .. 0.55 C.N. t - '!Ca + 2.37
'9 No
1930 _ 273
C.K t - ¥C... 2.92
19 K
\\
I
\\
"
\\\\ 1\
b'\ I
I
\
I
\
N
.. <~\ \
10' I
t '\:'J \
I
..
\~ \
\
....
"1°
1000 \~
\\
\
\ I~
0
3. ,o-J
1
."
"" e
@ (3
:: l • ::!t·!·~.~·· w Ii)
FIG. 7
SESSION IV - Temperature, Heat Flow,
Downhole Instrumentation
-ffl-
HEAT FLOW PATTERN OF THE UNITED KINGDOM
Abstract
The mean U.K. heat flow is a very normal 59 mWm- 2 • yet the values
vary systematically over the country between 40 and 100 mWm- 2 • There is
no geological or geophysical indication that this variation is caused by
recent thermal events in the mantle. A firm correlation between heat
flow and 'basement' heat production s·trongly suggests that the variation
originates in the upper half of the crust. Belts of higher heat flow
are underlain by granite-enriched upper crust. roe granite bodies are
of approximately the same vertical thickness as the low-grade metamorphic
rocks which they intrude: both are underlain by dense lower crust
relatively depleted in heat production.
These relationships imply that near surface heat flow anomalies
should persist to many kilometres depth. Thus high surface temperature
gradients in granites should continue down to all drillable depths.
Geothermal resources exist in such bodies, particularly where they are
insulated by overlying low conductivity sediments.
-w-
1. INTRODUCTION
The assessment of geothermal resources within the U.K., which is a
relatively stable continental area devoid of present-day or very recent
volcanic activity, falls naturally into two parts. (i) A description of
the temperature field, and understanding of the underlying factors which
control it, is of paramount importance in directing attention to the most
easily accessible parts of the resource and their size. (ii) The method of
exploitation must be controlled in part by the temperature but also by the
permeability and mechanical properties of the rocks in which the resource
is si tuated.
The present study has been concentrated upon understanding the
distribution of temperature within the U.K. subsurface. The approach taken
has been to make measurements leading to the calculation of heat flow. An
understanding of the geographic variation of heat flow has emerged from
consideration of these data. The temperature field is approximately [1]
given by the product of heat flow and thermal resistance to the depth of
interest: b.z.
T=T+qr-'
z 0
i ki
where the difference from surface temperature is (Tz-To)' the heat flow is
q and the thermal resistance is given by summing the thickness b.Z of each
discrete geological unit (i) divided by its characteristic thermal
conductivity k.
It is not out of place to stress once again (see [1]) that the
variation of thermal conductivity between common rock types is of the same
magnitude as variation in heat flow itself - a range of 3 times in each case.
The variations are quite independent, so that variations in thermal gradients
of 9 times are encountered. It is quite clear that measurement of thermal
gradient by itself is but a poor guide to the size of geothermal resources
at depths below those of the observations, or, in some cases, laterally.
Measurements for the determination of heat flow have been carried out
under the contract at a total of 26 sites (Figure 1). Of these, 12 have so
far yielded reliable values. With previously published values [refs 1-5]
these results bring the total number of U.K. determinations to 92 and the
area-weighted mean for the country to 59 mWm- 2. This mean value is close
to the European (64 mWm-2) and World (59-64 mWm- 2) averages for continental
regions [6].
-448 -
8· .. ·.~ 2·
o· 2"
OXFORD
HEA.T flOW SITES
• tak«l OYIi Gutlf'lg conllICt
.. laken
"J.Jt . . . .
CJIIIW ckJring conlracl
...
-r--~..... 111-""2 111M
11 Stl. number
... 100
kilometres
I
200
I
...
... ...
00·
.. o·
"".
- 449-
I
$ EJ
I"' +
' /~
Quartz .L C
! ....
E...
..
+. . . I
I 1
Silica glasa Wlll2J
-4 -2 0 +2
Percentage deviation flOm tha mean
."
.
/
/
/
•/ °
/0
0/
.
~
o<t" •
/
/
Plllbo• • Olwlded ~( • 181
Y
/.
/.
O~O----~
2 ----7
4 ----~e'-~
04"ided b.r W M'" K·'
-450 -
repeated measurements on samples of silica glass (4a) and a machineable
ceramic (4b). The ratio of their conductivities determined with this
instrumentation is 1.21 ± 0.18 whereas with the divided bar, the ratio is
1.19 ± 0.08. The needle probe has been used by normalising the results for
specimens of unknown conductivity to those for 5i0 2 glass which is therefore
used as a standard, as in the divided bar method. Clays undergo irrevers-
ible dehydration within hours of removal from a borehole and the use of the
needle probe at the site is essential to avoid systematic errors in
conductivity as great as 10-50%.
20~------~--------~------~
15 a.
10
OL-~~~-L-- __-U~~~~__~
1·2 1-4 1~ 1·8
- 451-
3. TEMPERATURE
Temperatures have been measured using laboratory calibrated thermistor
resistance elements previously described [4] as having precision to better
than ± O.OlK and accuracy to ± O.OSK. Holes deeper than 1.5 km have been
logged using a single conductor tool driven by a DC power source at surface
with an earth return. The tool generates a signal in the range 5 x 102 -
2 X 104 Hz proportional to the temperature of a thermistor element in an
oscillator feldback loop. This signal can be read at surface. The tool
is laboratory calibrated; and details of its construction and performance
are being prepared for publication elsewhere.
8' S'
.' ~
46
o
0' 2'
58'
100
I
200
!
56'
kilometres 56'
54'
54'
52'
50'
6'
.' 2' 0'
Figure 5. Heat flow values contoured at 45, 60 and 75 mWm- 2 • The two belts
of higher heat flow are delineated here by the 60 mWm- 2 isopleth.
-452 -
4. HEAT FLOW
The results obtained on the present contract are combined with previous
measurements [1-5) and the geographic pattern of heat flow in Britain is
shown in Figure 5.
There is a clearly defined regional pattern to the heat flow. A general
background of values mainly in the range 40-60 mWm-2 is crossed by two belts
of variable but higher flux locally reaching as much as 130 mWm- 2. The two
higher heat flow belts are delineated by the 60 mWm- 2 contour in Figure 5.
It has been argued elsewhere [10] that this pattern of variation
cannot easily be ascribed to variations in the mantle heat flow because
the scale width and sharpness of the anomalies suggests no great depth of
burial for their cause. Although, locally, heat flow refraction due to
bodies of contrasting thermal conductivity and convective transfer by water
in deeply buried aquifers may each be of importance, the continuity of
anomalies across surficial geological structures suggests that neither is
the ultimate cause of the pattern. Considerations of scale thus suggest a
crustal source but one whose characteristic depth is greater than that of
the geology exposed at surface.
120
• Hercynian granites
of SW. England
• Lower palaeozoic
granites
Miscellaneous
• cleaved basement
OL-~ __ ~ __ ~ __ ~ __ ~ __ ~ __ ~
o 2 4 6
Heat production Ao p.W/m 3
-453 -
The variation in heat flow is interpreted as being due to variation in
crustal heat production. Heat flow (qo' mWm-2) is correlated with the heat
production (Ao' vWm-3) of rocks sampled in boreholes where these are either
granites or low metamorphic grade cleaved rocks (Figure 6):
qo = 27 + l6.6Ao (correlation coefficient R = 0.96)
The granites mainly have higher heat production than the slaty basement.
Thus it is thought that the crustal rocks underlying the higher heat flow
belts include a higher proportion of granites than elsewhere. This
conclusion is in part supported by direct observation of granite occur-
rences but over much of the U.K. the basement is obscured by a cover of
younger sediments.
That the same regression line should fit qo-Ao observations in granite
and slaty basement is thought to be due to bodies of both rock types having
the same vertical thickness (Figure 7) of 16 ± 4 km. This could have come
about if the slaty basement overlies a lower crust of denser rocks, depleted
in heat production. Granite magmas would be too bouyant to crystallise in
the dense lower layer and would rise into the upper crustal layer. There
may also have been some exchange of heat producing elements between the
intrusions and their country rocks.
16 km
Average
Dense""--
Depleted -----
~
~
~
____
~ LOWER CRUST
..., , , , , _
-
~~
--::;-
..-"'2..--
-:> ~
~"----.
REFERENCES
[1] Oxburgh, E.R. et a1, 1977, Seminar on geothermal energy, vol 1 155-17~
[2] Pugh, D.T., 1977, Limnol. Oceanogr. 22,581-596.
[3] Wheildon, J. et ai, 1977, Seminar on geothermal energy, vol 1 175-188.
[4] Richardson, S.W. &Oxburgh, E.R., 1978, J1. Geol. Soc. Lond, 13~323-337.
[5] Bloomer, J.R. et ai, 1979, ref 6, 293-300.
[6] Cermak, V. &Rybach, L (eds.) Terrestrial Heat Flow in Europe (1979).
[7] Birch, F., 1950, Bull. geo1. Soc. Amer., 61, 567-630.
[8] Sass, J.H. et ai, 1971. J. Geophys. Res., 76, 3391-3401.
[9] von Herzen, R.P. &Maxwell, A.E., 1959, J. Geophys. Res., 64,1557-63.
[10] Richardson, S.W. &Oxburgh, E.R., 1979, Nature, 282, 565-567.
-455-
EXPLORATION AND INTEiPiETATION OF THE
SW ENGLAND GEOTHERMAL ANOMALY
Geophysics Department
Imperial College of Science & Technology
London SW7 2AZ
Abstract
The high magnitudes of the fir.st reported heat flows in SW England led to
serious speculation that deep convective circulation was a possible contribu-
ting fa~tor to the measured surface heat flow. Were convective circulation
to exist, the consequential heat flow enhancement would be localised. The
strategy ~f this research was tharefore to extend the coverage of heat flow
over the entire Cornubian. batholi.th., and its .environs. This has been
achieved through two phases of contract support from the Commission and the
UK DepaTtmeftt of Knergy. In the first phase eleven new heat flow sites
were occupied, and in tha .second phase twenty-five new sites have been
completed, bringing tha total coverage for SW England, including six earlier
published values, up to a total of forty-two. The uniformly high values
associated with the ~ranite (around 120 mWm-2) contrase with normal heat
flow (around 60 mWm- ) at sites remote from the granite. The uniformity
of the granite values suggests that convective transfer is not a significant
mechanism. Furthermore using model studias it is shown that the observed
significant contrasts in thermal conductivity and heat production, combined
with the likely space-form of the granite, reasonably account for the
observed heat flow distribution. Th.is study therefore reaffirms the
Cornubian batholith as the most favourable UK site for BDR development.
- 456-
1. INTRODUCTION
2. EXPLORATION
-457-
arisen at Predannack and Rosemanowas, sites of 300 m boreholes, in which
thermal conductivities are reasonably uniform, .to test the corrections
we apply for these recent climatic events.
3. INTERPRETATION
GRANITE SITES
GRILLIS FARM CM-A SW 6795 3846 100 20 33 92.2 112.9 119.9
POL GEAR BEACON CM-B 6927 3663 100 22 23 100.6 121. 7 128.6
MEDLYN FARM CM-C 7083 3404 100 8 32 98.3 113.6 120.6
TREVEASE FARM CM-D 7185 3180 100 20 33 91.5 111.9 118.8
TRERGHAN FARM CM-E 7353 3033 100 18 32 94.5 112.9 119.7
BRAY DOWN BD-A SX 1907 8177 100 18 31 88.9 113.4 120.2
BLACKHILL BD-B 1835 7820 100 20 34 97.0 119.0 126.0
PINNOCKSHILL BD-C 1892 7450 100 13 33 102.9 120.7 127.5
I BROWNGELLY BD-D 1924 7247 100 21 32 87.1 108.4 115.4
... GT. HAMMET FARM BD-E 1885 698-6 100 20 34 97.7 118.8 125.6
I
'"""
NEWMILL LE-A SW 4608 3435 100 23 32 102.7 123.8 130.7
BUNKER'S HILL LE-B 4022 2726 100 23 31 104.5 123.9 130.9
TREGARDEN FARM SA-A SX 0553 5945 100 20 32 105.8 125.8 132.6
COLCERROW FARM SA-B 0679 5763 100 20 32 102.8 126.5 133.4
WINTER TOR DM-A SX 6117 9156 100 29 34 78.6 107.4 114.2
BLACKINGS TONE DM-B 7850 8593 100 31 34 85.5 105.5 112.4
SOUSSONS WOOD DM-C 6733 7971 100 27 34 123.4 132.2 139.3
LAUGHTER TOR DM-D 6562 7549 100 31 34 90.0 114.2 121.0
FOGGIN TOR DM-E 5663 7334 100 31 34 89~0 110.9 118.0
SITES ADJACENT GRANITE
MERROSE FARM CDD-1 SW 6559 4351 100 23 23 72.2 79.2 84.1
KESTLE WARTHA CDD,..2 7533 2579 150 47 41 82.2 96.4 102.5
CALLYWITH FARM CDD-3 SX 0886 6783 150 43 47 91.2 101.1 106.2
GAVERIGAN GAV SW 9316 5916 325 105 30 97.2 98.1 105.7
TABLE I HEAT FLOW [N S. W. ENGLAND
B. OTHER BOREHOLES - SUMMARY COMPILATION
GRANITE SITES
GEEVOR SW 375Q 3450 403 7 31 128.6 128.6 134.8
TROON 6570 3677 122 36 40 109.1 122.7 129.8
SOUTH CROFTY 6680 4105 650 7 57 128.9 128.9 137.7
ROSEMANOWAS A 7352 3456 303 99 52 102.8 1005.5 113.8
ROSEMANOWAS D 7352 346'P 292 97 52 103.4 106.4 114.6
LONGDOWNS 7368 3462 t82 51 50 105.2 111. 7 118.2
HEMERDON SX 5733 9849 128 42 12 93.3 107.9 114.8
I
... SITES ADJACENT GRANITE
a-
0
I WHEAL JANE SW 7849 4380 230 164 49 125.2 125.2 132.9
NEWLYN EAST 8146 5390 103 34 34 90.5 104.6 111.1
BELOWDA BEACON 9788 6254 141 20 31 78.1 85.5 91.5
LANIVET SX 0216 6413 86 29 0 79.4 93.1 99.7
WILSEY DOWN 1797 8890 726 200 42 67.3 67.3 74.5
MELDON 5676 9220 61 17 25 103.7 114.1 120.1
BOVEY TRACEY 8271 7929 95 35 33 78.7 94.6 100.6
COUNTRY ROCK SITES
PREDANNACK SW 6901 1634 304 100 61 60.4 61.5 68.7
KENNACK SANDS 7325 1647 152 50 22 68.4 73.1 79.4
HONEYMEAD SS 7990 3930 290 46 15 54.0 54.0 57.4
CURRYPOOL FARM ST 2270 3871 182 58 24 52.8 60.7 67.6
CANNINGTON PARK 2470 4010 760 234 159 40.1 45.1 50.0
BLRCKINGSTONE s.w', fNlJlRN'O H[Rol FLOw
CEOPHYSICS DEP'.:IRTtI£NI
InPfRIAl couleE LONDoH
.'
.' .,
10
""
20 MERN CO NDU CTIVI TY
•,
~
: 3 . 10 W./M.OEG ·
30 , "'"
""
11ER I fLOW, UNCORREC lEO]
\ =85.5 MW . M· ?
'0
" " liE RT PROOUC T I ON
"'::: " " GRANJI =~ .9 10·6 . W. M 3
" ""
SO
0.:"'
j,::
\oJ ...
CORRE Cr IONS
c~ 0
60
0
0
10POGR APHY 3·5
'0 "" CLJI1ATE 20.4
""
"
80
"
" CORREC1ED
90
" HERT FLOW
]001 """ -- 112.4MW.M z
" "
FIGURE 1
50 mWm-2 60 70
C"rrectlon "
7'0
100
IPREDANNACK BOREHOLE I
150
250
FIGURE 2
-461-
200
SOUTH - WEST ENGLAND
HEAT FLOW COVERAGE ~ LUNDY ISLE
EEC CONTRACT NO 586-78-1EGUK IGS, Curry Pool Farm
UKAEA CONTRACT NO E/5A/CON/1)5
KEY TO BOREHOLES
Contract Sites Other Sites
• Granite o Granite
• Country Rock D Country Rock
.s::
-o 1-100 .00
Z IBOREHOLE LOCATK>N JAAPl
E
, ::.::
~
.,., "0
N ·c
C!'
cti
c
.9
~
z
Geevor
MI..
~
, ~
: ,~ ~ ~ ~
" ~ ,~ ~
~~
KEY TO BOREHOLES
Contract Sites Other Sites
• Granite o Granite
• Country Rock o Country Rock t
£
0"100 100
Z [HEAT FLOW VALUES 1
E Ccxrected lex tOpography and recent
~ cltmate only Heal tlow In mWm-2 _
I ~
... ....
W
(.!)
'"
I
(ij ~ .- -.
c: , .",.,/.
0
~ ",/ r--.
Z
CARNMENELLIS/
~C'l lt
LIZARD PENINSULA
200 300 March. 1980
FIGURE 4
National Gr id Km East
110
SURFACE HEAT FLOW Qs
~Modelled Cls
ro
90
80
o.10 J?·2
8:
7hi'=::: I I I I I '1- ~.,.....
731t,K@m 7?1
0
rod
61·5
P km
NORTH K 15 10 5 0
p
5 Ao.2-o" Ao • 2·0"
I K Xy ,2.5" Kxy I 2·5"
.,.
~ ·10
FIGURE 5 CONDUCTION & HEAT PRODUCTION FINITE ELEMENT MODEL ACROSS THE
CARNMENELLIS GRANITE PLUTON. March . 1960
special country rock boreholes have provided.the first opportunity of
properly evaluating. these contrasts between these rocks and the granites.
REFERENCES
1. TAMMEMAGI, H.Y. and WHEILDON, J., 1974, Geophys. J.R. astr. Soc., 38:
83-94.
2. TAMMEMAGI, H.Y. _and WHEILDON, J., 1977. Geophys. J.R. astr. Soc., 49:
531-539.
R. GABLE
Bureau de Recherches Geologiques et Minieres
Departement Geothermie
B.P. 6009
45018 ORLEANS CEDEX
ABSTRACT
-466 -
1. INTRODUCTION
-~-
To measure with accuracy temperature profiles as deep as 2,500
meters, we have conceived a specific equipment. A great sensibility of
temperature measurement has been obtained by using thermistor probe under
specific conditions. This great sensibility can be observed on the tempe-
rature profile at every change in lithology (fig. 3). About 10 profiles
have been measured in deep wells drilled generally a year before. The
heat flow values determination in these wells gave good results (fig. 3).
Mining holes are not very deep, generally about 200 m deep.
But they are generally drilled in basement rocks and continuous drill-
core is recovered. These holes are cased with PVC and temperature profiles
are measured one or two months after the end of the drilling. This delay
is the time needed to reestablish thermal equilibrium in the hole. Some
examples of results are presented : borehole of Saint-Yriex, SW of Massif
Central (fig. 4), borehole of Champrobert, NE Morvan (fig. 5).
4. RESULTS
Distribution of data
The new determinations of heat flow complete (annex 1) those
of the first map (annex 2). It added to the data already existing for
sedimentary basins and provided the information for basement regions, and
in the case of the Massif Armoricain, the very first data. But data are
scarce for the north of France and montaneous regions (Alpes, Pyrenees).
Distribution of anomalies
Generally high heat flow values are connected with recent geo-
logical features (graben) : "Fosse Rhenan", "Couloir Rhodanien", in rela-
tion with the elevation of the Mohorovicic discontinuity (annex 2). High
values also appear, such as in Lorraine, where thermal activity is known.
-468 -
SOULZ BPR 5 NANTOUIUET'
TEHPEil.ATURE "C Tf.HP'EiATUR.£ 'C
50 0:>
01 •, ,, • 0
,,
,,
,,
,
1000
1000
~ ! "...-.
<: <:
jj.
g Q m7m
lOOO
lOCO
'HEAT FLOW DtnlUoUKATION
FIG. 1 - HEAT FLOW DETERIIIBATIOll IN DEEP HOLES FIG. 2 - HEAT FLOW DETERIIINATI(!( IN DEEP HOLES
(CORRECTED TDIPERA'RJBE PROFILE) (CORRECTED TDIPERA'lUBE PROFI LE)
Profit d~ te-~a1ur~ Saint MAUR 112
(11.10.79 )
0~~____~r-____~.-____2~5____-.~~__~~~,,____- .__________~
50
\
---- - - - --- -- - - ~"'-"!-,,...L,I
100
SENONIEN
CENONANEN
-------------.-~~
..... " .
WEALDIEN
. \ ,',.
600
KIMMEFaOGIEN
650
HEAT FLOW DETERMINATION
---
()n;tuct~ tfit fbw
-
D~pth
700 m JffNrcJ;in ,""Iml SEQUANEN
MARNEUX
ft4,~7 25,m 4;7.6
FIG . 3
- 470-
<;} YRIEIX SJO CHAMPROBERT S 59
o r ',' ? ~ ':' "",,,,,run 'C oI P lJ 11 TM¥*'altllw 'C
•
M
13.2
III
Z1,1
. !I)
<2
]:" ,6
155.
I
...
;j C)()
81,
16.
II'
91
'iIl.S 150
HEAT FUM' OE","NlNATKJN
106 D~P(h Ccn~crMry c;......" H~.r- Flow
(m) mw/·c/cm 'c/rtJOm mN . ,"-2
51 s,n f,CO 71 . I
]8.'5 1 .SlJ C, D IOJ, J B~ S.77 I, ZO 69, 1
16.30 J.'1 l.n NIS.' III 6,31 I,ZO 75, 7
96. J5 ~51 1, 13 'HJ.9
/2( 5,10 1,'0 71, 8
IH. 'J t.m 1,1 '09.9
,~ 109 Im=73
----
FIG. 4 - HEAT FLOW DETERMINATION IN MINING HOLES FIG. 5 - HEAT FLOW DETERMINATION IN MINING HOLES
-472 -
Finally the level of heat flow in France is high compared with
that of 2urrounding countries as observed at the time of the first flow
determination (1).
REFERENCES
-473-
A CRITICAL STUDY OF HEAT FLOW DATA IN FRANCE
Summary
More than 160 heat flow data with different origins have been combi-
ned in order to outline the major trends of heat flow in France. The basic
data are separated into different groups according to the various charac-
teristics of the measurement and a systematic comparison between each
group is attempted. In particular, classical measurements obtained from
conductivity determination and temperature logging in shallow boreholes
are compared with measurements derived from bottom hole temperature in
deep oil exploration wells.
As a result of this investigation, a 30 % difference between the two
types of data is suggested. This bias could result from paleoclimatic
temperature variations. Its potential effect upon the heat flow pattern of
France is briefly discussed.
1- INTRODUCTION
The first heat flow data in France (Hentinger and Jolivet, 1970),
obtained from temperature and conductivity measurements in boreholes at
thermal equilibrium, have given values higher than the continental average.
From 1975 to now, The Institut National d'Astronomie et de Geophysique
(INAG) has proceeded to new heat flow measurements in France, confirming
this previous observation (Groupe Fluxchaf, 1978). Seven boreholes have
been drilled for this purpose and a number of experiments was carried out
in exploration boreholes. About 40 new heat flow determinations have been
obtained, most of them in area of basement outcrop.
At the same time, an indirect method of heat fiow evaluation using
file data from oil industry was developped at the Bureau de Recherches
Geologiques et Minieres (BRGM) (Gable, 1979). In this method, the geother-
mal gradient is obtained from temperatures measured in deep boreholes du-
ring formation testing or bottom testing (with a correction), whereas the
conductivity is deduced from well logs using tables of physical constants.
About 90 heat flow evaluations were thus obtained in the sedimentary
basins.
Because of their respective locations (roughly basement outcrops and
sedimentary basins respectively), these two types of data are complementa-
ry. However, since the characteristics of these measurements are so diffe-
rent (depth range involved, method for estimating temperature gradient and
conductivity), fr is necessary to check their consistency. In a previous
study (Vasseur and Nouri, 1980), it was shown that the mean value of the
two groups of data were not significantly different. However, because of
their complementary distribution, the coherency could not be assessed defi-
netely.
Since then, new data have been obtained which enable more critical
tests for consistency. It is the purpose of this study to discuss further
the coherency of the data. The data included are those obtained by INAG and
BRGM described above as well as new measurements obtained in the course of
the present contract (e.g. Gable, this issue).
- 476-
(though important scatter appears around the line ~A = 1.35 ~B)' For
Parisian Basin (B. P.), this ratio is even larger (1.53).
A comparison between data of groups A and C on 3 pairs gives a ratio
of 1.51. On the contrary, nearby data of groups A and D seem to agree clo-
sely with a ratio ~A/~D around 1 (except for one pair). Only 2 pairs could
be obtained for comparison of group D and B and no conclusion can be drawn.
Another type of comparison can also be obtained using data belonging
to the same geological unit. The whole area is divided into six regions
(Fig. 3) : Massif Armoricain (M. A.), Parisian Basin (B. P.), East - Jura
(E. J.), South East (B. R. A.), Massif Central (M. C.), Aquitan Basin (A. P.)
(E. J. and B. R. A. are not geological units but were chosen in order to
obtain a sufficient number of data). Fig. 3 gives the histograms of the
data for each group with the mean value and standard deviation. Although
the number of data is often small, it is clear that, in most cases, data
of group A give, for the same region, values higher than data of group B
(and C). For Parisian Basin, this result is obvious.
4- TREND ANALYSIS
The results obtained above suggest that data of groups A and D give
systematically values larger than data of groups Band C. This would not be
so troublesome if only one type of data could be used to trace out the
pattern of regional heat flow variation. In fact, because of their geogra-
phical repartition, all the groups of data must be used simultaneously in
order to display this pattern over the whole country. In this paragraph,
classical contouring methods are used in order to map the trends of regio-
nal heat flow variations.
As the data are scarce and scattered, a severe smoothing is necessary
in order to point out the trends. In the simple method used, the function
(heat flow) is locally approached by a quadratic polynomial computed as
follows : M being the point where the function is to be evaluated, select
the data existing in a circle C with center M and radius R I then compute
the 6 coefficients of the polynom G(x,y) so that E Pi ( G(xi'Yi) - ~i)2
6t. ,y.)e:C
is a minimum. The weight factor Pi is defined Ey ~ Pi =«t_-{)/(~ +n 2 »2,
d i being the distance between M and the data point (xi'Yi) i the parameter
n defining the smoothing degree. It is important to calculate the standard
deviation of this smooth approximation ~(x,y) by
-477 -
The representation is acceptable if~ is of the order of the assumed error
on the gata. For Rand n, values of 200 km are found to be convenient be-
cause the heat flow can be evaluated over most of the country with an accep-
table standard deviation.
On Fig. 4 isolines of heat flow are plotted when accounting for data of
groups A+D only (Fig. 4a), B+C only (Fig. 4b) and A+B+C+D (Fig. 4c). Their
characteristic features will be discussed later. The standard deviation a
of this approwimation is 11 mWm- 2 for the case B+C and 15 mWm- 2 for the
case A+D, which is reasonnable compared to the accuracy of the data. In
contrast with this, when all the data are used together, the standard devia
-2
tion reaches 19 mWm • An attempt to use this standard deviation as a cri-
terion for deducing a possible correcting factor for the data is now de-
velopped.
-478 -
6- DISCUSSION
As a result of the above study, it is conjectured that the data of
group A (and D) obtained in oil exploration boreholes give values higher
by about 30% than those of group B (and C) obtained in shallow boreholes.
This assumption cannot be asserted definitely because too few data enable
a direct comparison ; only hints for such a difference can be deduced from
the above statistical study.
Moreover, it cannot be inferred that the actual value for the region&
heat flow is 0.7 ~A for group A and ~B for group B ; data of group A could
also likely correspond to actual values and, in these circumstances, values
of group B must be increased by a positive correction of about 30 %.
Several physical explanations for such a correction can be proposed
such as the peculiar location of boreholes (shallow mining boreholes as
well as oil exploration boreholes are drilled above special underground
structures). Another possibility is the paleoclimatic variation which does
not affect shallow and deep temperatures with the same intensity. For
example, a decrease of 12° C of the ground temperature during Pleistocene
between 10 000 and 70 000 years before present results in a decrease of the
temperature gradient of about 0.007° C/m (over 20 % of its value) for
depth less than 400 m (such as boreholes used in group B) , whereas the
bottom temperature of boreholes deeper than 2 000 m is hardly affected
(less than 5 %). Therefore the proposed correction could be related to the
paleoclimatic variation ; in this case a high positive correction needs to
be applied to data of group B. Further study is on the way for the determi-
nation of paleotemperature variations and of the relevant correction.
At the present stage, we are just able to speculate about the validity
of the heat flow map presented on Fig. 6. The relation of this heat flow
pattern with hercynian directions emphasizes the possible role of hercynian
granites lying in the Southern and Northern part of Massif Armoricain, over
Massif Central and Vosges. Numerous heat production measurements performed
in these granites indicate values of 2 to 5~WID -3 • For a 7,5 km thickness,
this corresponds to a superficial contribution to the heat flow of 15 to
37,5 mWID -2 which is not far from the amplitude of the regional variation
observed on Fig. 6.
-479-
..
.
.
.. ,
•
?0
A
C
..
.. .. ,
.:< B \ o
•
..
I'
-480 -
A B C 0
A
... . / ,
MA I··I ~ "'
/ / " It
I" I
/ II "
/ I •
• / .t
100 f I" >00
/ I " L
/
I / / II
1/
" ."
/8' 1. 31
A/c -1 . U
·-L,Ll:: I
I I n .~ ~ i>l 1 rn I
O/e - 1. 03
'0 1.0 . "/0
ent
Fig. 2 Compa rison of nearby measur ements of differ
tL-
groups and result ing ratio.
accord ing
Fig. 3 Histog rams of heat flow data separa ted
n
to their group (A,B, C, D) and to their locatio
(6 units define d in the upper part of Fig.).
0"=11
Fig. 4 Isoflux contours obtained from the trend analysis using data A + 0
(Fig. 4a), B + C (Fig. 4b), and A + B + C + 0 (Fig. 4c). 0 gives
-2
the standard deviation of each representation in mWID
- 482-
Aa
"
,. ,. ,. 17'1 '1
r·
4> • .I A I{J A (Petrol)
¢J •
OPB
o
4> - A.; 'Pc tOOl)
I. d . otAA AC)
.7
•
Fig. 5 Isolines of the standard deviation when a correcting factor AA is
applied to data A and D, and AC to data c. The cross indicates the
minimum.
Fig. 6 Isoflux contours using the "optimal" correcting factor for data A
and D (A A '" • 7) .
-483 -
ACKNOWLEDGEMENTS
We greatly appreciated the help of Mr Gable (BRGM) for data exchange
and fruitfull discussions. We are indebted to Mr Goguel who suggested the
use of trend analysis in order to look for a potential correction.
REFERENCES
-484 -
GEOTHERMAL TRENDS OF DENMARK
Laboratory of Geophysics
Aarhus University. Denmark
Summary
- 485-
1. INTRODUCTION
2. GEOLOGY
-486 -
'6
\
\
\ l000m
~
~'~'N
'"
'\ \
"\ \ \
.. \
\
\
\
\
\
..... d
o -2
8 C temperature step increase at 10,000 BP yields 12-13 mW m
the model of Bech (6) for glacial and postglacial times yields
-2
14-16 mW m ,whereas the whole model of past 120,000 years,
including the temperature decrease of gOC at 65,000 BP results
in corrections of 8-9 mW m- 2 •
-488 -
drilled by ano~her agency for its own purpose. Needle probe
and divided bar thermal conductivity equipment and data inter-
pretation techniques have been improved. Mainly supported by
the Danish funds, a temperature logging equipment has been
built to measure to a high precision temperatures and tempera-
ture gradients in deep boreholes. Measurements are being car-
ried out in a more than 3 km deep borehole. Furthermore, ther-
mal conductivity and heat production have been measured on
core materials from 27 deep boreholes. The most promising geo-
thermal reservoirs seem to be found in North Jylland, and our
main activity is therefore concentrated in this area (Fig. 2).
-489 -
brium gradients. Results from borehole Jebjerg are shown in
Fig. 3. Final results from the other holes will appear as they
reach thermal equilibrium.
Jebjerg
50
E
.c
~
"
100
-490 -
Gradl.nt ,o"teUon ImK m-I) Fig. 4. Model climatic temperature
-W 0 W ~
gradient corrections calculated for a half-
o t--"'--..,.,..........-r::,.;i
space of constant thermal diffusivity (K).
Curve 1 is calculated from the climatic
500 model used by Beck (6) of past 120,000
years for latitudes 40-60 0 and K = 10-6m2 s -1
Curves 2, 3, and 4 from a simplified surface
1000 temperature model containing two temperature
steps -10 K at 70,000 BP and +10 K at 10,000
e BP and with K = 1.0, 0.5, and 1.5·10- 6m2 s- 1 ,
~ 1500 respectively. At shallow depth (~500 m) sig-
i
c nificant corrections associated with climatic
temperature variations of mainly past 1,000
2000
years are superimposed on the simplified
models.
25
'000
probe several times during the year may yield valuable results.
5. TEMPERATURE-DEPTH RELATIONS
200 Ars
400
800
800
1000
1200
1400
! 1800
:-
~
Q 1800
2000
2200
2400
2800
2800
3000
3200
30
10
O+---~---r--~--~~~r--L+-~,
o 2 3 4 5 8
Thermal conductivity k IWm- 1 K -1)
-492-
n .. 28
10 i. • 0.36 ~W m- 3
• D.22 pW m- l
c:J Sandston..
~ l l m,s-tone
~ Oolomlle
o 0 1 02 Q3 O. OS 06 07 08 09 10
Radioa'live hfOt production AIJoIW m ~3l
m Hallie and ant>ydlll.
soo
1000
ISOO Mod.l B
"'0<1,1 A
'to _ I'"C I To .6~ I
sooo
.
,.
SIMlO
- 493-
view to obtaining the general minimum and maximum temperature-
depth functions. The area is likely to contain considerably
greater temperature variations than indicated by the BHT valu~.
6. CONCLUSION
---
REFERENCES
-495 -
THE SHALLOW SUBSURFACE TEMPERATURE FIELD IN THE NETHERLANDS
W. VAN DALFSEN
Summary
Shallow subsurface temperatures in The Netherlands were measured in
groundwater observation wells. Based on the measurements, temperature maps
have been assembled showing temperature intervals of one degree centigrade
for depths between 25 and 250 m below ground surface.
The relatively low-temperature areas at a depth of 25 m are predomi-
nantly groundwater recharge areas. This pattern of relatively low-tempe-
rature areas persists to at least a depth qf 250 m.
The relatively high-temperature areas are situated in the river val-
leys and in the east and south of the country. The relatively high tempe-
rature gradients in the latter areas should be ascribed to the relatively
low thermal and hydraulic conductivity of the rock in these parts of the
country.
The relatively high temperatures in the river valleys are ascribed
in first instance to the convective heat transport by upward groundwater
flow. Some tempertature logs, however, do not show a convexity along the
depth-range between ground surface and the less pervious base of the water
bearing strata, where groundwater flow is most-likely to occur. This fea-
ture as well as the lowest temperatures down to a depth of at least 250 m
in groundwater recharge areas stresses the influence of groundwater flow
on the subsurface temperature field.
-496 -
1. INTRODUCTION
Ten temperature maps of The Netherlands (scale 1 : 600 000) have been
assembled, showing temperature intervals of one degree centigrade at depths
of 25, 50, 75, ••••• 225 and 250 m below ground surface. Three of them, for
the depths of 25, 125 and 250 m are represented here (Figures 1 - 3).
On the maps the locations of the measured temperature entries have been in-
dicated.
2.1. Features of the observed temperature field
Comparing the temperature maps to each other, one learns that tempera-
ture, as is to be expected, generally increases with depth, indicating an
overall upward heat flow.
-497 -
Less self-evident, at first glance, is the pattern of relatively low-
temperature areas at a depth of 25 m (shaded in Fig. 1). These relatively
low-temperature areas are The Veluwe, The Utrechtse Heuvelrug/Gooi, the pro-
vince of Drente with the adjacent southeastern parts of the provinces of
Friesland and Groningen and the southern part of the province of Noord-Bra-
bant along the border with Belgium. This pattern persists and even becomes
more pronounced at greater depths.
The pattern of relatively high-temperature areas at depth of 25 m
(8 > 11 °C) is less persistent. The relatively high-temperature area in the
southwestern part of the country fades out at greater depths, where another
pattern of relatively high-temperature areas appears. At a depth of 125 m
these areas (8 > 13 0 C) are situated in the IJssel valley, the central, the
eastern and probably also in the southernmost part of the country (shaded
in Fig. 2).
The highest temperatures at a depth of 250 m are expected in an area
north of Venlo in the southeastern part of the country, where temperature
probably exceeds lS o C. This has been based mainly on the temperature mea-
surements to a depth of 200 m in observation well Californie and also on
those in observation well Homberg, the latter however reaching a depth of on-
ly gg m. Another high-temperature area is situated in Twente, where in the
shafts of three rock-salt exploitation wells temperatures of 17.5, 17.6 and
18.1 °c were measured. With the available temperature data no more areas
with temperatures exceeding 17.0 0 C at a depth of 200 m could be indicated.
The lowest temperature, observed at a depth of 250 m, is 12.3 0 C in ob-
servation well Hoenderloo in The Veluwe. Extrapolation of the temperature
logs of two other observation wells in The Veluwe, indicated an area with
temperatures even less than 12 0 C. Thus temperatures at a depth of 250 m in
The Netherlands cover the range of approximately 12 - 18 0 C.
-498 -
ground surface (2) with the temperatures (9 < gOC) at a depth of 25 m (Fig.1)
in The Veluwe, it appears that here subsurface temperatures are less than ave-
rage air temperatures. This area constitutes a heat sink that with the usual
simplified physical model of heat-exchange between atmosphere and solid earth
cannot be explained satisfactorily. Noting that The Veluwe, being an ice-pus-
hed ridge, is a groundwater recharge area, plate VII-3A of (3), the effect of
heat transfer by infiltrated water should obviously be taken into account.
In this area the total annual precipitation surplus almost totally infiltra-
tes into deeper permeable, unconsolidated sands. Since the precipitation
;urplus occurs mainly in the cold season (October - April), the low tempera-
tures at a depth of 25 m in The Veluwe can be ascribed to the infiltration
of large masses of water with a less than average air temperature.
Having recognized the influence of groundwater recharge in the cold
season upon the shallow subsurface temperature field of The Veluwe, the ex-
planation of the other relatively low-temperature areas at a depth of 25 m
is straightforward. The Utrechtse Heuvelrug/Gooi, another ice-pushed ridge,
is also a groundwater recharge area as well as large areas in Drente and
along the border with Belgium. Comparing the relatively low temperatures
in the northeastern part of the country with the regional average air tempe-
ratures, 8.5 - gOC, it appears however that here the effect of groundwater
recharge on the temperature field is less pronounced.
With the above considerations about the relatively low temperatures at
shallow depths in groundwater recharge areas a start~ng point has been pro-
vided to describe the temperature field at greater depths. Firstly a few
pertinent notions should be made pertaining to groundwater flow in the
hydrological cycle, which are essential in the subsequent description.
-499 -
groundwater fluxes through horizontal planes in recharge areas have a
downward direction and decrease with increasing depth of the planes, and
conversely in discharge areas have an upward direction and increase with
decreasing depth.
Usually geohydrologists indicate a less pervious stratum below which
- in their view - groundwater fluxes are negligible. The depth of the less
pervious base of the water-bearing strata in The Netherlands increases from
a few tens of metres below ground surface in Twente, the Achterhoek and in
two southern parts of the country to over 250 m in the Central Graben and
the northwestern part of the country, plate VII-3D of (3).
-500 -
Having identified upward groundwater flow as a cause of relatively high
subsurface temperatures, the relatively high-temperature areas in the Maas,
Rijn, and IJssel valleys (shaded in Fig. 3) and in the low western and nor-
thern parts of the country, are to be ascribed in the first instance to up-
ward groundwater flows.
It should be clear that the temperature anomalies, due to upward or
downward groundwater flow both depend on the amount and areal extent of the
groundwater fluxes and on their depth ranges.
-501-
...
.....
. ...
.. :. , 8 £ (
o I IJ"
:...-...--,'" ...
- 502 -
-I'
~c== r;::;:>.
o
/
- 503-
FIGURE 3 TEMPERATURE DISTRIBUTION AT DEPTH OF 250 m
-504 -
4. REFERENCES
3.
Atlas of The Netherlands, 1963-1977
Government Printing and Publishing Office.
-505 -
TEMPERATURE AND HEAT FLOW PATTERNS OF ITALY
Abstract
A preliminary reconstruction of the underground temper-
ature regime in Italy has been made, using terrestrial heat
flow values measured during pure and applied research,temper-
ature data from oil and geothermal wells and the values meas-
ured in oil and geothermal reservoirs. ~
Due to the fact that the data were not always reliable
or distributed uniformly, the resulting maps (terrestrial heat
flow and temperature at different depths) are of a semi-quan-
titative nature. Although representing the regional trend there
may be a certain margin of uncertainty on a detailed scale.
-D-
1. INTRODUCTION
The work described in this paper represents a prelimi-
nary attempt at reconstructing Italy's underground temperature
regime from terrestrial heat flow data, bottomhole temperature
measurements and other information capable of indicating tem-
perature values at depth. The work was conducted under the
CNR-EEC Contract No.088-76-EGI (1).
- 507-
An attempt has been made by LODDO & MONGELLI (2) to
correlate terrestrial heat flow distribution and geological-
structural conditions.
3. UNDERGROUND TEMPERATURES
The distribution of the underground temperatures was
reconstructed from temperatures measured:
a) in 1052 wells drilled by AGIP for oil research (7) and in
some others drilled by other companies;
b) in 68 productive oil reservoirs (7);
c)in wells drilled for geothermal fluid research (8),(9),(10),
(11),(12),(13),(14) and (15).
- 509-
Consiglio 1f.. .,,,,,le delle Ric.ehe
_
..' ___ f-r " Ist ituto '"te(nazionl~ per Ie Ricerche
Geotermiche - Pisa
-
ITALIA
..
Carta del flusso di calore
... ..
' t- -
[mw/m']
• misura
mW/m'
-
Rappresenlaziona convanzionare
11' . ,. 1' 1
-
X ~ Long · 82, 85450
II I ,4, . 2.
Y = 6376, 10387 ' Sin ( lat - 42 0 )
."in km
" I Gradi sessadecimali
", Areo di parlilielo II 42 0 N
r4. Raggio . li ssoid8~ a 4~ N
"
... I... ,.. or .1.- .. -
- 510-
Consiglio Nazionole delle Ricercho
ITA LI A
Carta delle tel'!lperature
sotterranee ( in"C) a 500 m
sotto il piano campagna
41 1 l al I" ~
X - Long· 82, 85450
111 14 + la l
y ~ 6376,10387 · Sin I Lat - 42°)
"tin kin
.a. Gr.di sessadecimali
." Arco di par.nelo • 42· N
•.,. Raggio elissoidaMi a 4r N
."---_.....
o 100
_-......... 200 km
...
...
.. ... ... ..
- 511 -
Consiglio NlZionele delle Ric ... c'"
ITALIA
Carta delle telTlperature
~tterranee ( in"C ) a 3000 m
sotto il piano campagna
Aappresentazione convenzion.le
XI~ lOr:o~ 82, 8545Ct'
1' 1 'c. 11 1
Y = 6376, 10387 ' Sin (Lat - 42' )
Illin km
12. Gradi sessadecimali
,;. Arco di par.llelo • 42°N
.. ... ,,.
- 512-
Consiglio Nazionale delle Ricwche
IlALIA
Carta delle tel!'perature
sotterranee (in"C) a 5000m
sot to iI piano campagna
Rappresentazione convenziolMlle
I I' la l _S f
X - long · 82. 85450
11' ,.. 111
Y = 8376. 10387 ' Sin (Lat - 42 0 )
'hln km
'2 .Gradi sessadecimall
Il.Arco di parallekt a 42 00 N
.""Raggio elissoidale 8 42" N
o....._ _......
100 200 km
' _ _---0.
..
>t.
".
... .....
' ,,. ... ...
-513 -
3) temperature interpolation and extrapolation depends criti-
cally on geological knowledge and/or assumptions on the hyd-
raulic state of rock bodies at depth. Subsurface complications
thus affect the reliability of thermal trends in some areas;
4) geothermal anomalies are better evaluated on a local scale
whereas on large denominator scales a strong horizontal grad-
ient turns into an effective discontinuity. This may lead to
a serious misrepresentation and reading difficulties;
5) these maps should be taken as semi-quantitative, valid with-
in a regional context, but with a high margin of uncertainty
on a local level.
References
- 514-
ture.Proc.lnter.Congr.Thermal Waters,Geothermal Energy
and Vulcanism in the Mediterranean Area,Athens !,56-70.
(9) P.BALDI, G.CIVITELLI, R.FUNICIELLO, G.LOMBARDI,M.PAROTTO,
L.SERVA (1976) Study of the stratigraphy and minerali-
zation of the deep wells in Cesano geothermal field
(Rome,Ita1y). Proc. Inter.Congr.Therma1 Waters,Geo-
thermal Energy and Vulcanism in the Mediterranean Area,
Athens !,71-86.
(10)A.CALAMAI, R.CATALDI,M.DALL'AGLIO, G.C.FERRARA (1975) Pre-
liminary report on the Cesano hot brine deposit
(northern Latium,Ita1y). Proc. 2nd U.N.Sym-p.Deve1op.
Use Geothermal Resources,San Francisco !, 305-313.
(ll)R.CATALDI & M.RENDINA (1973) Recent discovery of a new
geothermal field in Italy: A1fina. Geothermics ~(3/4),
106-116.
(12)G.MINUCCI (1961) La perforation 'rotary' pour 1es recher-
ches d'energie endogene. Proc.U.N.Conf.New Sources
Energy,Rome l,223-233.
(13)F.PENTA & B.CONFORTO (1951) Risu1tati dei sondaggi e di
ricerche geominerarie nei'Campi F1egrei' per vapore,
acque terma1i e 'forze endogene' in genera1e. Ann.Geo-
fis. !(3), 35-51.
(14)F.PENTA & B.CONFORTO(1951) Risu1tati di sondaggi e di ri-
cerche geominerarie ne11'Iso1a d'Ischia da1 1939 a1
1943 ne1 campo del vapore, delle acque terma1i e delle
'forze endogene' in genera1e. Ann.Geofis. !,1-33.
(15)F.PENTA & B.CONFORTO (1951) Sulle misure di temperatura
del sottosuolo nei fori trivellati in presenza di
acqua e sui relativi rilievi freatimetrici in regioni
idrotermali. Ann.Geofis. !(1),43-93.
- 515-
HARMONIC PICK-UP:
PRELIMINARY· MATHEMATICAL MODEL FOR INTERPRETATION
Abstract
-516 -
1. DESCRIPTION OF THE MODEL
P (r=Rp, t)
V(r=Rp,t)=Vo.sin(2nft)
Fig. I
2. EQUATIONS
The pressure P(r=Rp,t) at the entry of the fissure is to be determi-
ned. To do this it is necessary to determine the fields of pressure and
velocity which result at all points of the fissure. The assumptions above
result in values of pressures and velocities depending only on the radius
r and time t : v = V(r,t) P = P(r,t)
Firstly, the local conservation equation for mass and momentum at a
given point and secondly, the comportment and state equations of the fluid
lead, after linearization, to the following wave equation for pressure :
- 517-
c being the wave velocity in
c
2 -l o
-
o P
(r -)= 0
the fluid
r or or
with the following boundary conditions
-
oP (Rp,t)=-pVo21Tf COS(21Tft) 9P '
-tRf,t) = 0
or or
We are only concerned with the steady state solution at the same frequency
as the flow applied to the entry of the fracture.
- 518-
Fig. II
109
I
I
10B /
I
\ ./ f
\ ./ I
\ /' I
101 .\ I
1 ~----'-~_~--r---'--"--'---'--"..-l..-'-'--"--~-~-~--1
10:5.1.-
o 2 4 6 8 10 12 14 16 18 20 22 2' 26 26 30
FREQUE'JCE(HZ)
-519 -
4. IDENTIFICATION OF THE PARAMETERS
The experimental signature obtained using the harmonic pick-up makes
it possible to determine the radius of extension Rf and the thickness e of
the fracture by making comparisons with the set of theoretical signatures.
Rf is obtained from resonance frequencies by means of the formula given
above. Using this determination ~f Rf, it is possible to plot a reference
signature corresponding to reference thickness e • The signature corres-
o
ponding to the true thickness of the in situ fracture is deduced theore-
tically from the reference signature by a vertical translation 0 , rela-
ted to the thickness by the following formula:
o
e = 10- /4 (e and 0 in em) (Figure III)
5. CONCLUSION
The main purpose of this preliminary interpretation model is to show
the possibility of obtaining the geometrical parameters of the exchanger
by measurement. These models.,remain to be refined in two ways:
- by taking into account the deformation property of the rock matrix;
- by studies of geometries which are closer to real cases, notably
by examining fissures which have ellipsoidal or other sha~s,and
Fig . III
\
donnees experimentaies
\
\
\1 10- 2 , 5/4 0,24 em
- 520-
THE PERFECTING OF A HARMONIC PICK-UP FOR THE SIMULTANEOUS
DETECTION OF THE THICKNESS AND EXTENT OF A FRACTURE CROSSED
BY A BORING
Abstract
The survey of fractured media by non-steady water tests requires the
creation of dynamic signals at the entry to the fractures. Apparatus desi-
gned for such field survey is described. This apparatus can create a sinu-
soidal pulse flow inside the fracture and measure the pressure. Between
two packers, set above and below the fracture, an alternating piston ap-
plies pressure through a membrane to fluid in the fracture. A sensitive
oil pressure detector measures pressure and temperature. Signals are trans-
mitted to the head of the boring by a frequency modulation device.
This pick-up is designed for borings as deep as 2000 m, 165 mm in
diameter, and with temperatures up to 150°C.
- 521-
1. PURPOSE OF THE APPARATUS
The measuring device described has been named a "harmonic pick-up"
and is designed for surveying of certain geometrical characteristics of
underground reservoirs, in this case the extent and thickness of a frac-
ture cau~ed by hydraulic fracturing at a great depth, taking into account
the general assumptions regarding the shape of this fracture.
-522 -
fig. I
f F
<:>
U')
~
f
D
fonction F <:>
U')
M
1<0
,Ir--rr-.--.<:>
<:>
d'
", I
I
~I
.!l
....<0I-<
<0
:>
- 523-
FUNCTION D : flow measurement
This is effected by calibration in the laboratory using the cylinder
capacity and the measurement of the r.p.m. of the turbine by means
of a rotation detector (5).
FUNCTION F : Packers
The measuring chamber is limited in the boring by two packers (4) and
(14), operating by expansion by means of hydraulic jacks powered from the
head of the boring. The first version of these packers will be for a boring
diameter 165 rom but are interchangeable, allowing the harmonic pick-up to
be adapted 00 larger diameters of boring. It is also possible to operate at
the bottom of the boring using only the upstream packer.
4. PERFORMANCES
In its present version, the harmonic pick-Up can be used for frac-
tures located as deep as 2000 m using borings with a minimum diameter of
165 rom. Fitted with a turbine and gas bottles, it is approximately 12
metres long and can withstand differential pressures of approximately 200
bars. Operating temperature can reach 150°C. The cubic capacity of the ex-
citer piston is approximately 800 cc; the excitation frequency can vary
between approximately 0.2 Hz and 20 Hz. A patent has been applied for to
cover the apparatus, and it is currently planned to design versions which
are slightly different to adapt it to other test conditions.
- 524-
IN SITU DETERMINATION OF THERMAL
CONDUCTIVITY IN CASED DRILL HOLES
Abstract
Geothermal investigations on the Schwabische Alb (FRG)
ha¥e been made in shallow drill holes of about 30-40m depth.
To get the heat flow, the temperature gradient and the
thermal conductivity of the surrounding rock must be known.
If no core is available in situ methods for the determination
of the thermal conductivity in cased drill holes have to be
used. The developed method for in situ determination of
thermal conductivity uses the principle of an infinite cy-
lindrical heat source. A separated section of the water-
filled drill hole is heated up with a constant rate of heat
input. The temperature rise of the well-mixed water section
as a function of the heating time allows the calculation of
the thermal conductivity. Assumptions and advantages of the
used method are discussed. The construction of the probe and
first results of in situ measurements carried out with the
described device are presented.
-525-
1. INTRODUCTION
-526 -
T temperature (K)
t time (s)
Q heating power per length (W/m)
A. thermal conductivity (W/mK)
C constant
-527 -
4. RESULTS
-528-
o~ ____ ~
550 000
- 529-
Casing
Packer
ContrOl Unit
Heater
constant cunent
source
DigItal
Ohmmeter
Calculator
In cluding
clock generator
Thermal
Sensor
Printer
Tape Recorder
Circulation
Power Supply
_, ."'_.L-Circulation
Unit
-530 -
'"
to
ZI
..
F
_11
I.'
2U.D IiiI1D 100\0
TIME
'"
..
II
II
U
!..11
i:!
~
~
.. Hea.t1ng ~ver :
Thermal conductivity.
24.0S 'II'r.
1,6) ,,/0>1<
'"
e IS
11
0' o·
THill la,
- 531 -
IS
II
..
II
, 0
..
..
"'
,t
II
0' 0'
TIME (al
-532 -
2 .5
:.::
@
E
~
>-
I-
>
I- 2 0
U
::J >- I 81 z=17.5m
0 (;
..
z 2 81 z =27.5 m
0
u 0
:j 82 Z= 7.5m
0
...J .D
« 0
<1 82 Z = 7.5m
L ...J
5 86 z=21,9m
cr 15
UI
I I 6 87 z =15,5 m
~
I-
7 87 z =26.0 m
!j 87 Z = 26.4 m
9 88 Z = 15,5 m
10 88 z = 21,5m
10
II 88 Z= 29,S m
12 89 Z = 20,S m
13 89 z=30,5m
5. CONCLUSION
The first results abtained with the new probe in shallow
drill holes in the area of Urach are very encouraging because
of the good fit between laboratory data and in situ values.
The advantages of this cylindrical source method can be
summarized as follows:
ACKNOWLEDGEMENT
The authors are indebted to Dr.R.Hanel, G.Zoth (both
NLfB, Hannover), B.Geiser, U.Schenkluhn, M.Schmarsow and
U.Topper (all TU Berlin) for their valuable help and advice
during the investigations and to U.Cramer typing the manus-
cript. The work was kindly supported by the Commission of
the European Communities (Brussels) and the Federal Ministry
of Research and Technology, FRG.
REFERENCES
1) Beck, A.E., Anglin, F.M., Sass, J.H.: Analysis of heat
flow data - in situ thermal conductivity measurements.
Can. J. Earth Sci.B, 1-20, 1971.
-534 -
DESIGN AND TESTING OF DOWN HOLE PROBES FOR
OPERATION IN DEEP AND HOT ENVIRONMENTS
Abstract
During 1977 the research work conducted as part of Contract 166-76 EG-I led
to the construction of all the parts of the electrical device for measuring
temperature and pressure in geothermal boreholes. Further research directed
at improving geothermal measuring techniques, as part of the new ENEL-CEE
Contract 585-78/7 EG-I, led to the development of two probes, one for measu
ring the diameter of geothermal boreholes (CALIPER) and the other for dete-
rmining fluid characteristics at wellhead (ISOKINETIC PROBE).
With the first probe one can measure the diameter of the wells along two
orthogonal axes and define fluid velocity in a longitudinal direction du-
ring production or injection tests.
The isokinetic probe represents a new instrument for studying fluid charac-
teristics at wellhead. By taking an isokinetic sample of fluid along the
well radius one can trace profiles of the flow in the various phases.
CISE (Centro Informazioni Studi e Esperienze) of Milan carried out the
design and construction of both probes.
-535-
1. INTRODUCTION
trical measuring device already prepared for the temperature and pressure
probe. Using this new probe one can measure the diameter of the boreholes
along two orthogonal axes and determine the longitudinal fluid velocity.
this was impossibles until now because the market was unable to offer ade-
quate electrical instrumentation. Apart from the temperature and pressure
probe the device also represents a new means of attaining the following
main objectives:
1) individuation of the formation intervals most suited to stimulation ope-
ly, the velocity parameter attainable with the mechanical devices offered
phases: for example, one obtains a profile of the steam, water and gas
flow. Integrating each profile on the whole section of the well one thus
obtains: steam, water and gas flow-rate and, consequently, the composi-
-536-
2. CALIPER PROBE
The caliper probe is lowered into the wells and operated by means of
the 7-wire electric cable and the coupling section used in the temperature
and pressure probe. A description will now be given of the caliper and re-
two pairs of orthogonal arms, each of which measures one of the axes of the
transducer and one signal only per pair. The arms are spring-thrust against
the bore-walls and the measurements taken as the prG~e is pulled slowly up-
wards. The opening and shutting mechanism for the arms is adjusted by a DC
motor directly controlled from Lne surface. The fluid velocity is measured
integrated magnet to a coil in the fixed part of the probe. The propeller
revs are determined by counting the impulses and its velocity is equal to
the average of the number of impulses in a given unit of time.
2.1 Construction
The overall design of the caliper is shown in Fig. 1, the main parts in
2.2 ~~~~~_~~~_~=~~~~~~~_~=~~
The D.C. permanent magnet motor operates the opening and shutting mech~
tures is about 10,000 hours. The permanent magnets are in ALNICO 8. The
graphs in Fig. 2 show the characteristic curves of the motor. The motor mo-
vement is transmitted from a magnetic coupling (that separates the gas en-
epicyclic reducer. The latter, a normal two-stage reducer modified for tem-
peratures of 2S0oC, operates the trapeze-thread screw that opens and shuts
the arms. The maqnetic coupling also acts as a torque limiter to avoid over-
-537-
loading the motor. Should the shutting system break down the probe can still
be extracted.
2.3 ~~~~~~~_~~~_=~~~~~~~_~=~~~~~=~_!~~_~=_~~=
Each pair of arms is connected mechanically so as to have one transdu-
cer and one signal only per pair. Separately operating arms would have meant
one transducer per arm and more than seven cable conductors to carry twice
as many signals above ground. Where the cross-section is no longer ellipti-
cal the area calculation is slightly less accurate than in a system with
four independent arms. The arms are spring-thrust again the bore-walls and
the measurement taken as the probe is pulled slowly upwards. One of the two
pairs of arms is thrust against the bore-wall by a much stronger spring
than in the other pair; the probe is thus inserted in such a way that each
pair of arms lies along an axis of the ellipse representing the bore sec-
tion, in order to measure the area. The two stronger arms will consequently
tend towards the longer axis and the two weaker arms to the shorter one.
As the cable is too long to use a transducer with differential transformer
a group of D.C. potentiometric transducers has been fitted. The arms meas-
uring the major axis have a maximum expansion of 740 mm, with 660 mm maxi-
mum for the arms of the minor axis.
2.4 ~=_~=~~~~!~_~=~=~~~~~_=~=~=~
Fluid velocity is measured by a ducted propeller that transmits the im-
pulses through two integrated magnets to a coil in the fixed part of the
probe. The propeller revs are measured by counting the impulses and its ve~
-538-
the approximate values of the electric parameters, calculated for a 4000 m
wires, provide the signals corresponding to the opening of the arms. As the
cable is fitted with seven wires only the rev counter was connected in se-
ries to the motor operating the arms, which does not run while the measure
ments are being made. A 330CL resistor was connected in parallel to the m~
tor so that the signal from the coil would not pass through the latter (and
tact) .
2.6 ~~~~~~~~~_~~~~~~~!~~~_r~~~!
The hydraulic plant for calibrating the caliper (Fig. 4) was construct-
ed in such a way that the central part into which the probe is inserted can
urements, for each diameter, of the characteristic curve for propeller velo
3. ISOKINETIC PROBE
led at wellhead during measurements (no.l in figure). Fig. 5 also shows the
sampling equipment used during these operations. The isokinetic device con-
sists of two types of measuring probe: the mobile probe scans the well
along one radius and the fixed probe measures flow on the bore walls and in
the immediate vicinity, where the mobile probe is unable to reach. Both the
mobile and the fixed probe have a static pressure tube fitted at the inlet.
positioned precisely along the bore diameter. The isokinetics of the sample
inlets is equal to that measured on the bore walls in the same section; the
in fact give a zero reading. This situation can only be obtained, of course,
if the pressure in the sampled section of the bore is high enough to carry
- 539-
Teflon Cable
Motor
Reduct I on Gear
Aru
Prop.llar
-540 -
~otor
1d1\ng
Speed
(rpt)
1000
500
10 30 ~o
Voltage supply (V)
~O .-------------...,.50
Yol tage Cou pl
Supply
30
( V)
20
10
o
Propeller .Ignal for two rota Characteristic curves recordsl on the motor.
lion direction.
Fig. 2
..
L.
2
..(
c :>
"
""oc:
L.>
~------------------N
.,
c:
260 {i
.
L.
600n .:?
CD
....
><
~ __________________________-L___ ~
rl111~____~2~~5~r~1______~
Central conductor
L -___________________________________ Arlour
- 541-
CALI PER
Inl.rchanq.able Tube
10 .easurl.lnt tank
10 lusuruenl
- 542-
"ob11, Probe
flied Probe
Hu l Exc hanger
T
- 543-
the mixture into the probe and the flow-meter instruments inserted at the
outlet. Once isokinetic sampling conditions have been achieved a steam, wa-
ter and gas mixture is usually obtained at the probe outlet. The flow-rate
values of each separate phase can be obtained by providing the mixture sam-
pled by the mobile probe with enough heat to totally evaporate the water
content (probably very small). This heat is provided electrically. The wa-
ter flow-rate value can be calculated from the amount of heat used to eva-
porate it, once the evaporation heat value is known. In the case of the
fixed probe, the water flowing along the bore wall has a higher specific
flow-rate than that carried with the steam and gas so that the heat content
than evaporation. After condensating the steam and cooling all the water
with a special cooler, the fluid passes to a separator where the uncondens-
able gases are removed. The water flow-rate is then measured with a burette
and the gas with a rotameter. Values are thus obtained for gas, steam and
All the component described above have been constructed and tested in
the laboratory. The ground instrumentation for the two probes has been se-
-544 -
DIRECT HEAT FLOW MEASUREMENT PROBE PROJECT.
ABSTRACT
-545 -
I) - Mathematical Approach.
Let us consider an infinite isotropic and homogeneous me-
dium, thermal conductivity A , in which exists a constant
vertical temperature gradient Y .
Temperature 9 is everywhere definite by the relation:
e & 9(z) =Yz
+Q
0 Y
Fi g. 1. -Q
we find
-546 -
with: ~:. '( A
SO Q, heat quantity produced and absorbed by the discs, is
proportionnal to the preexisting heat flux ~ ,the coeffi-
cient of proportionnality being a geometric constant factor
having the dimensions of a surface and which, in this theo-
rical case, does not depend upon the thermal properties of
the medium. For these reasons, we shall call "equivalent
surface" the ratio Q/~ of the heat quantity absorbed and
produced by the discs to the normal flux in the medium
before their introduction. In this very simplified theori-
cal case, this ratio is independant of A and always grea-
ter th an 1.
K
h/o K
20
0 1
1 2.6
2 4.56
5 10.5
10
10 20.5
2
o
012 5 10
Fi g. 2.
II) - Realization.
It seems natural, in order to obtain production and absorp-
tion of the same heat quantity per unit time at two points
separated by a known distance, to appeal to Peltier effect.
So in fact, the two discs will be constituted by the sol-
ders of several thermopiles made of numerous doped bismuth
tellurure barrels.
Peltier effect is a reversible thermoelectric effect occu-
ring in a elementary circuit made of different conductors,
which consists in the production of a certain heat quantity
at one of the solders and in an equal absorption at the
other one each time an electric current flows through the
circuit.
- 547-
A thermopile is the amount, in series, of a great quantity
of such elementary circuits, allowing the accumulation of
the thermoelectrical effects and the relative diminution
of the parasitic Joule effect.
Doped bismuth tellurure presents a thermoelectric power
about ten times greater than that of the others materials.
Besides, its thermal conductivity does not differ signifi-
cantly from the mean value of rocks conductivity.
We have previously shown that such thermopiles could be
constructed, could produce the Peltier's strenght necessa-
ry to the measurement of geothermal flux and that our first
realizations had succeeded in front of hard experimental
conditions. These points will not be reviewed again.
We shall only mention again that one of the thermopiles is
used, like a differential thermometer in order to evaluate
the quantity a (h) - a (-h), (by making use of its Seebeck
effect). Current adjustment through the others will be made
in order to have this quantity equal to zero by the varia-
tion of the so producted Peltier's strenght.
So, such a probe will have a cylindrical configuration,
whose central part, strongly laterally isolated, will be
constituted only by the thermopiles (hight 2h, radius a),
while the terminal parts (hight 2H, radius b) will assume
the best possible diffusion of the Peltier perturbation
through the external medium.
But in reality this external medium is not the homogeneous
isotropic one we have dealed with in the preceeding para-
graph: the probe must be set down a hole (radius c), fil-
led with a liquid whose thermal properties vary from one
hole to another and are not very well defined. Sometimes,
the hole iss ur r 0 unde d by a cas i ng, 0 r by a tem en t, and so
on, and in the best case, the real system geometry is that
shown in fig.3, table 1.
+
+ +
+ +
+ +
+ +
+ terminal
+ + ports CU
+ + AI
+ + isola t i!'g
+ + rna tee-rial
+ . . +
+ ' . ' .,. oJ .--yt--.:...._--I-.:!:,;,_ dc?PCC?d
+ bismuth
+ tee- Ilu ruree-
+ +
+ + boree- hoi C?
+ + liquid
-
+ + ee-nvI ronrTlee- nt
+ +
+
Fig. 3.
- 548-
Come back to the Theory.
Naturally, it does not exist any analytical approach able to
account for the thermal behaviour of such a system with suf-
ficient accuracy. So, we must appeal to computer methods in
order to obtain approximate numerical results, (the accuracy
of the method we have used is about a few per cent).
The different occuring parameters in this study are:
Geometric Parameters.
2h = hight of thermopiles.
a = module radius.
al = external radius of the isolated module.
b r a diu s 0 f the te rm ina 1 par t .
c = hole radius.
2H hight of the terminal part.
Environmental Parameters.
A3 thermal conductivity of the hole.
A4 = thermal conductivity of the external medium.
Constants.
A1 th erma 1 conductivity of Bi 2Te 3 .
A5 = thermal conductivity of the te rm ina 1 parts.
A 2 th erma 1 conductivity of the isolating material.
What about limit conditions?
In the plane z = 0, because of symetry, temperature is nought.
8(f'o)=O
(we use cylindrical coordonates, so 8 = 8 (f'z).
When F becomes very great, the perturbation caused by the
probe trends towards zero :
8 (.,0 ,z) = '( z
The same when z is very great
8 (F ,00) ='( z
- 549-
for the general equation of the system,
with the conditions of table I and the limit conditions we
have just defined.
It can be shown that i f 9 (r,z) is a solution of equation
(1), we may write:
9- (p,z) = 91 (f'z) + 92 (p, z)
with
div ( A grad 94 ) 0 div (A grad 92,) =Q
9A ( f,O ) ,. 0 9,t, ( FJ 0 ) = 0
9.(aO,z) ;:. ,(z 9 t (cO,Z) ,.. 0
-550-
if the equivalent surface of the probe is a constant quanti-
ty or if it depends upon the thermal parameters of the envi-
ronment. It permits also to study the influence of the sys-
tem geometry upon its comportment. For a better understan -
ding of its importance, it may be preferable to speak r.ow in
terms of thermal resistance.
When the probe is active by its Peltier effect in an origi-
nally isothermal medium, a temperature difference fl8 2 .• p-
cing a flux
fig. (4).
1
pears between the active surfaces of the thermopiles, produ-
whose distribution in the medium is complex
+
+
+
+ +
. . . . . . . ,'. .'."
+
...+
+
+
+
+ 3 4
+ +
+
~ ...., - .,::"
+ ....:< ... ,' ... - .- +
. ..
' ~. ............ '
+ .. .. :
~
+
.......... +
+
Fi g. -4.
A first part J. of the flux goes directly from one active sur-
face to the other through the thermopiles barrels (1). BeR
the corresponding equivalent thermal resistance of this path.
In the same manner, a part J"
of the flux goes through the
isolating cylindrical part of the probe, whose thermal resis-
tance is R2 (2).
Another parti of the flux flows directly from one surface
to the other through the liquid of the hole (thermal resis-
tance R3 (3).
Finally, the remainding part j~ of the flux flows from one
surface to the other through the terminal parts (thermal re-
sistance R4 ), radially through the liquid of the hole (ther-
mal resistance R5 ), then through the external medium (ther-
mal resistance R6 ) and once more radially through the hole
(R 5 ) and the other terminal part (R 4) (4).
The equivalent diagram, in terms of thermal resistances, is
shown fig.(S).
- 551 -
R~ Rs
( ~A ~~ ~) ~~
il 9 1 R~ Rz. R.3 R6
\ R~ RS'
\
F i g. 5.
~= R,
At,
» R4+ Rs (1)
-552 -
Then ~o would be equal to ~h , the dimensions of which
-553 -
Preliminary results are shown fig.6.
d...
Rc.
H
\Or---+-+-~------~----------------~---
2,5 5 10 20
Its appears that the independance of the probe Viz the ther-
mal properties of the external medium is greatly dependant
upon the size of the terminal parts. Whatever h may be,
there is an optimal value for H (or an optimal range of
values, roughly comprised between 5 and 10 cm), beyond
which ~ diminish suddenly. Apart from the fact that
Rc.
when their lenght increases, terminal parts become thermal-
ly resistant, we have no found at the present time any
quite satisfactory explanation to this unexpected variation.
On the contrary, the variation with h is much more regular
and our present conclusion is that the best possible probe
seems to have h between 20 and 40 cm and H between 5 and
10 cm (having taken account of some other factors as time
constant and fabrication procedure). Such a lenght for h
is only possible with a composite probe, made of two modu-
les separated by a copper barrel.
We have already seen that contact resistance depends upon
the difference b - c and to have sufficiently small values
for (R4 + R5) necessitates that b be very close to c. A
good means for having this without disturbing the thermal
state of the hole when setting the probe down it is to make
composite terminal parts, with a central copper core and
a weak proportion of peripherical cylinders of the same
metal. Such a system greatly improves the efficiency of the
probe.
The casing of the hole, if present, is equivalent to a ther-
mal resistance in series with Rc and may be taken into
account as a parasitic element.
-554 -
Excentration of the probe trends to increase the ratio ~ .
Rrock
Therefore, all measurements are made with the probe tangent
to the borehole side.
Conclusions.
At the present state of our investigations it appears that
with a short probe ( h < 5 cm), condition (1 ) is never
realized. Nevertheless, such a probe could give us results
within 10-20 % accuracy, if we had any informations about
environment thermal properties. This is often the case.
Besides, such a probe may be very usefull for differential
or punctual heat flux measurements.
But for current measurements a longer probe is necessary.
The method of measurement is simple. The probe is set down
the hole at a predetermined depht and remains at rest till
the stationnary thermal state is established, which is
ascertained by the constancy of the registered thermal dif-
ference ~e1 between the two surfaces of the central modu-
le, in case of a short probe, or between the two modules in
case of a composite probe. Then Peltier's strenght is put
on and adjusted till the condition ~et.. = - ~e1 is satis-
fied. ~!e can then deduce ~ from the knowledge of the ap-
proximate apparent surface·of the probe for the measurement
conditions.
We hope that the study of the variable state, when Peltier's
effect is put on, as well as when the passive probe is set
down the hole, will bring us supplementary informations
that will permit us to obtain more accurate results.
- 555 -
AN IN-SITU THERMAL CONDUCTIVITY PROBE
Abstract
It was the aim of this investigation to develop and test a probe
for in-situ thermal conductivity measurements also applicable for rocks.
First a new theory for a spherical probe for absolute measurements was
developed. A sensor including the spherical probe (heater) and a down-
hole hydraulic drilling mechanism for drilling well defined special side
boreholes was constructed and tested in limestone clay and slate.
A first application of the sensor was done in flat drillings around Lake
Laach (Eifel) in addition to temperature gradient measurements.
The data evaluation of the temperature and thermal conductivity measure-
ments are presented. High gradients compared to the Eifel area could be
detected from 2 out of 5 flat drillings close to Lake Laach.
-556-
1. SPHERICAL PROBE, PRINCIPLE OF OPERATION
In~his case we produce a spherical temperature field using a nearly
spherical heater with constant surface temperature. We measure the time
dependent temperature field at different radial distances and the power
injection of the heater.
~ shows the configuration. A side drilling with a diameter of
about 13 mm is brought very smoothly into the rock with about 30 0 inclina-
tion. After this the sensor (diameter is 12 mm) is pushed into this small
drilling. A nearly spherical heater at the front end of the sensor pro-
duces the spherical temperature field. 8 thermistors at different radial
distances from the heater register the temperature wave up to a distance
of about 10 cm. If the constant surface temperature of the heater and the
heater radius are known, the distribution and velocity of the temperature
waves including the power loss of the heater depend only on the thermal
conductivity and thermal diffusivity of the heater environment. Because of
the thin disturbing layer of mostly water between the sensor and the sur-
rounding rock, the sensor parameters and heater temperature, heater-
radius cannot be assumed to be constant, but all these parameters can be
determined from the in situ temperature registration. No calibration con-
stants are used, only temperature and heater-power have to be registrated
precisely.
-557-
[l-;C~)]
R To
T (r, t) = (1)
,.
r
with
~ (x)
je-T
x t2
dt
0
The power loss through the surface of the probe is: (Kesse1s, 1977)
R heater radius (2)
L (R,T) = 4TI To RA [1 + _R ] To surface temperature
=
a Iiif r = distance from heater-center
a2 = thermal diffusivity
A = thermal conductivity
3. TECHNICAL DESCRIPTION
Fig. 1 also shows a sketch of the principle of operation of the
sensor and drilling mechanism. The central body is fixed in the main
borehole. Inside we have two different systems: one for the side-drilling
mechanism under a fixed angle of 30 0 and a maximum depth of 30 cm and a
diameter of 13 mm, while the second mechanism drives the thermal conduc-
- 558-
tivity probe. The axis of the drill device can be rotated with different
speed. Both systems are operated from the surface using hydraulic pressure.
Using this pressure either the first or the second cylinder can be pushed
forward and backwards for a fixed length. A flexible part, different for
the drill and for the sensor, is able to follow and drive the drill in its
given direction.
The depth of penetration is monitored on the surface. When the dri11-
hole is completed the drill can be returned into its initial position.
Then the total inner device of the central body can be rotated by 1800
hydraulically to place the sensor head in front of the drill-hole.
The sensor is pushed into the drill-hole without rotating. A strong flexi-
ble tube connected between sensor and hydraulic axis contains the harness.
The downhole position is again monitored on the surface.
-559-
Warmeleitfahigkeitssonde
fUr in situ Bohrtochmessungen
f
~
I1
f
~~
1~
if
oS
i=
I
i
.c
I
~
E
.8
Ci
::;,
ro
i
I
•
i
Fig. 1
II!
-560 -
t 1.0 L(~) fur kleine Zeiten
-
~
...J 0,8
0,6
0,1.
0,2
OT----r---~i---.----~I--~----~i--~--~i
o 0.2 0,1. 0,6 0,8
1 - 1. J
Vi [min 1 -----
Fig. 2
- 561 -
IUSUII; UIII 13. I I 1f18
TIEfEMEII£IOI ".!500
AIIZAII. In I'[SSPIJICTE
T[~J
10 11 12 13 l' I~
.IOEO()I
.9OEtOO
.8OEtOO
.1OEtOO
.6ot:tOO
...... .................
• JOEtOO .. '
.90[-01 .18£+02
.8Cl£-o1 .18(+02
............................ .12£+02
............... .6ot:+01
.IOE-OI
.................................................. .20['01
.............
2 10 2 10
A[nl~(J~11 A[n/~(J~'1
Fig. 3
- 562-
+ . .L GEOELECTRIC
MT MAGNETOTE LLUR I C
11 IFN 0
0 IGM 1
0 IF 2
0 BGR
* DRILUNGS
o 1 2km
Fi g . 4
-563 -
evaluation was done for S flat drillings (SO m) around Lake Laach, the
location is shown in Fig. 4. The results are as followed:
Drilling No.1: (see Fig.4) (12 m depth)
no gradient possible due to Basalt
average value for thermal diffusivity a2 6,1.1Q-3cm2/sec
i for 4-6 m 2,31·10- 3cm2/sec
TO 1l,40 C
thermal conductivity (S m) 2,7S·10- 3cal/cm K sec
a2 from temperature registration (S m) 2,6.1Q-3cm2/sec
a2 from power registration (S m) 2,9.1Q-3cm2/sec
These values of temperature gradient compared with other Eifel area values
shown an anomaly for drilling No.2 and 4. (Topographical corrections are
under investigation).
6. REFERENCE
Kessels. W., Theoretische und experimentelle Untersuchungen an einer
Kugelsonde fUr in-situ Warmeleitfahigkeitsmessungen, Diplomarbeit.
Institut fUr Geophysik. Technische Universitat Braunschweig. 1977.
-564-
SESSION V - Geophysics I - Active and Passive Seismics
-565 -
SEISMOLOGICAL INVESTIGATIONS IN URACH
G. Schneider
Abstract
-567 -
1. INTRODUCTION
Qs = 75 + 25
-569-
From measurements ouside of the anomaly:
Qp = ~OO Profile Schwabische Alb-NEI,Fig.l.(8).
On travel paths from the Upper-Rhinegraben area, running as
SMS through the crust, the following values were obtained:
QS S = 150 ~ 25 Profile Kaiserstuhl-MSG-Gle
The last va~ue after reducing the "normal" part of the ray
between MSG and the western border of the Urach-Kirchheim
volcano is:
QS S = 60 + 20
M
When all values are taken together it can be seen that in-
side the anomaly the Q-values are only 50% of that observed
outside. Different models for the relations between absorp-
tivity and temperature indicate a temperature increase of
about 100 K inside the anomaly.(l).
Acknowledgements.
- 570-
Study of the Kirchheim-Urach area. Special acknowledgements
are directed to Miss J. Turnovsky M.S. for the handling of
magnetic tapes and first evaluations of the records, to
Dr. J. Wieck for maintenance of the PCM-stations during the
two year interval of observations and to Dr. V. Valius for
calculation of theoretical travel-time curves.The monthly
bulletin of the German Central Seismological Observatory Er-
langen-Grafenberg was an very important tool for evaluations
of the material made shortly after the events.
E9
BFO
EB
RAV
a 50 100
!
km
- 571 -
Beu I'n t [sJ
J Urach
./ Volcano
10
a. b.
Oberschwaben
- i
tsediments
--
./
,.-
~
5
10
-~ -
- 572-
IR - t P [5) _h_:}_8~~.,; _ !p: ?3 _k_mj~
2
IJ. [km]
o 100
Alb Obs
BFO TUB
STU
+NEI
STUT b.
- 573-
W .1 Kirchhqim -Urach Volcano 14 E
1
MSG l Gle l Beu 1 Ubk
A [km]
0 20 40 60
a.
...
10
20
Schwabi5che Alb .
z [km]
Upz
Btu (52 ) Gle {51l
+
Z b.
D.ls
4
51 - s2 = 10km
2
vp =6km/s ~c.
Op= 200
o 5 10 15
v [Hz]
Fig.4 Determination of Q.
a. Travel paths through the crust of the studied area
(Full lines: ray paths of distant earthquakes;
dashed lines: ray paths of local earthquake phases
from the western part of the Schwabische Alb and the
southern part of the Upper Rhinegraben).
b. P-pulses from the Schwabische Alb recorded at the
stations Beuren (Beu) and Glems (Gle).
c. Theoretical amplitude relations for P-waves ob-
served at two different hypocentral distances 5 1 ,5 2 ,
- 574-
Bibliography
I
(1) CARLE,W., 1975: Mineral-und Thermalwasser von Mittel-
europa. Wissensch. Verlagsges.m.b.H., Stuttgart, 643 pp.
gart, 5 pp.
- 575-
STRUCTURAL STUDY OF THE URACH AREA BY DEEP REFRACTION SEISMICS
Abstract
This paper summarizes the available results of the 1978-79 refraction-
seismic survey in the geothermally anomalous Urach area which relate to the
depth structure and the velocity distribution of the Variscan basement as
derived from first arrivals of compressional waves. Application of the
MOZAIC time-term analysis resulted in an detailed map of depth isolines of
the basement which outline the extent of the SW-NE striking Rotliegend
trough with maximum depths of 1.6 - 2.0 km around Urach and rather steep
flanks in the NW and SW but gradual rises to shallower depths in the NE and
SE. A uniform velocity distribution within the basement has been derived
with the velocity placed at 5.66 + 0.02 km/s. No significant decrease in
velocity within the anomaly as compared to outside could be detected. No
evidence for velocity anisotropy was found in the area. It is suggested
that any high temperature source for the anomaly has to be located at
depths well below 3 km.
-576-
- 578-
The possibility of lateral velocity variations was tested by calculating a
velocity for each observation and then plotting it as a function of posi-
tion at the midpoint between shot and receiver (Figure 3). Simple con-
touring was then attempted to identify eventually present systematic chan-
ges of velocity within the anomaly as compared to outside. However, no such
changes could be found.
The delay-times obtained in the MOZAIC analysis were converted to depths by
using a mean overburden velocity of 3.6 km/s which is in basic agreement
with the results obtained in the deep borehole URACH III (19) and earlier
measurements carried out in the area (6). In Figure 4 the depths below sur-
face to the basement are plotted in the form of isolines graded in 200 m
intervals. Allowing for an uncertainty in the delay times of 0.02 - 0.07 s
corresponding to a depth uncertainty of 100 - 300 m, the depths are in rea-
sonable agreement with available borehole data (9, 19). The contour map
clearly outlines the extent of the NE-SW striking Rotliegend trough (6, 16)
which reaches its maximum depths of 1.6 - 2.0 km around Urach. A rather
steep rise of the basement in the NW and SW and a more gradual upward fle-
xure in the NE and SE characterize the borders of the trough. The small
uplift between shotpoints U 2 and P coincides with the position of the
Hohenzol1 erngraben. It should be noted, however, that the maximum sedimen-
tary thicknesses do not coincide with the maximum values of geothermal
degree. A more elaborate discussion of the methods and results is given by
Jentsch et al. (14).
c) Ray tracing:
Figure 5 shows a cross section along profile 8 1 which runs through the cen-
ter of the anomaly and passes the borehole URACH III. The basement sur-
face was derived by using the two-dimensional ray tracing technique of
Cerveny et al. (8) and supports well results obtained with delay-time ana-
lysis. The basic agreement with borehole data is illustrated by the sonic
log plotted in Figure 5 (19). Although the mean overburden velocity of the
sonic log is around 4 km/s the 3.6 km/s mentioned above is considered to be
a better estimate as the sonic log does not comprise velocities in the first
340 m of the overburden which are probably very low.
5. DISCUSSION
These results lead to the following conclusions:
As the maximum basement depths do not coincide with the temperature maxima
of the anomaly a possible mechanism to maintain the anomaly, i.e. the sedi-
ments storing or shielding the heat, over a long period of time, is there-
-579-
fore considered to be unlikely. If any velocity changes within the basement
in the Urach area due to thermal effects are present they are probably
smaller than the maximum 2 %which, on the basis of laboratory measure-
ments (11, 12, 18), could be expected for the temperatures thought to
prevail in this area and would be just about resolvable with the methods
and data used here.
A possible high temperature source for a hydrothermal convection system
will certainly be located well below ",3 km depth or it most probably
would have stronger effects on the seismic velocities.
Any crack or fissure systems that can serve as aquifers for hydrothermal
convection within the basement are likely to have random distributions,
eventually connected with fractures associated with the Upper Miocene
volcanic activity.
Regular crack patterns should have had a bulk anisotropic effect on the
seismic velocities (4).
As to the question to what extent standard refraction-seismic methods can
be used as an exploration tool for areas of geothermal potential we think
the following answers can be given: With the type of experiment and the
methods described here a straightforward and comparatively cheap way is
given to derive refractor velocity structure and topography as related to
compressional waves over a large area and thus provide the basic informa-
tions to aid and to define where other, more detailed, methods should be
applied.
In low grade geothermal areas such as in Urach the changes of seismic para-
meters due to temperature are just at the limit of the resolving power of
these methods. Although these would eventually just about be able to re-
solve such small effects they would most probably fail to detect areas of
low thermal potential not known prior to the investigation. As other stu-
dies in high-grade thermal fields have shown (15), however, refraction seis-
mic methods can well be used to outline the extent of such areas.
Hence well controlled seismic-refraction experiments should be used as a
first step in assessing an geothermally anomalous area as an aid to define
areas of special interest where more detailed methods, e.g., reflection
seismics, should be applied and where other geophysical methods such as
electric and magnetic are likely to add significant information.
-580 -
8' 9' 10' II'
49'15' 49'15'
e,
.. .,
f10
~~--
\
~4/
""-- ,-
48'30' 48'30'
.,
OIll..H
lCI-lQ~
I 8QREH41£ ~ACH m
- 0 PrtOFU.E
'('l- [S(UNE OF GEOTHERMAL
,. OEGRE:E t..,.n-cl
o f8£1III4li SHOTPaNT
aU.t.A~'"
0 'aN t
..
__ FAUll E
201<>4
"-'" VOlCANIC AA£A
4.,.45' 47'45'
8' g' 10' II'
- 581 -
, REDUCED TINE IN SEC
<N o N 0'1
,.,
"0
I.·•. I I I
MOtO" "01
C
·i"Frl'.~A~·
"
AlOlO" S04
. ••101041. Iltt
0'1
10.041 CHit
gs
.u. •n
•
· I·UI "t2
1010., l.at
•••,.'01. CMI'
:>< 101041" eM"
:x
,- ,otOtOI. (MIO
(1)0
••
,.1011 CHot
o .·,.,r," •
•
1OIon
lotllU
Cl'tO •
tHOl
CJl -:.~
--j
.r.'l~:" •101011 tHOt
D
-:r.'!'1:'J!' :( :1- ............... • 10tDit CHOI
ZN
no ,..- ',r ",
:~ .,olon 11111,0
I"TI
j,
Z -r,I.' .,01011 .,.01
.I
.r
IIl0101t 111101
.
w I-
a A)OIOn 1(1101
~
. ,
.
.,01014 kiM,.
-I "'·"'(fl·
.1 .IL .,olOn 11.01
'" '
•• 11 ,., u. ..............
,
AlOia" MOl
II I ','" .'~lr" I I I
-582 -
rlO' r 9"lO' 10·
....;:..::------i------t-=---
.... HW[!n
-------t" .
'f(l1Xllln
.. .. 56 "-"lIS[(
• U-u ~
( :I'!U ~
/~ :::;:~:t..~"(o:on«/If)I
Q ~tII'01"IS
O>IlillliW
""'
..·+..,..
- - - ----+..- - -- ---+..-,..----~10."..
• H-U,·
1"1 "' - , • •
I 't·,.· oUllmOlll
!I. '.-:0-
• SMO'I'OI"1fS
ISQ.IM:S OF
CKOIIolUlNAL 0U>ft1Et
IN ...,,.(
- 583-
10 30 50 60
V;3·6kmi.
V.5.66krnl>
9.5-66 kml.
vlkml., Ii 5 i. 3
"",Ie-log
'"
~
REFERENCES
(1) Bamford, D.: MOZAIC time-term analysis, Geophys. J.R.astr.Soc.,44,
433 - 446, 1976
(2) An updated time-term interpretation of Pndata from
quarry blasts and explosions in Western Germany, ~ Explosion
seismology in Central Europe, data and results: P. Giese, C. Pro-
dehl, A. Stein, eds., Springer Verlag, Berlin-Heidelberg-New York,
215 - 220, 1976
(3) Pn velocity anisotropy in a continental upper mantle,
Geophys. J.R.astr.Soc., 49, 29 - 48, 1977
(4) Bamford, D., Nunn, K.: Insitu seismic measurements of crack aniso-
tropy in the carboniferous limestones of northwest England, Geo-
phys. Prospect., 27, 322 - 338, 1979
(5) Bamford, D., Jentsch, M. Prodehl, C.: Pn anisotropy studies in
northern Britain and the eastern and western United States,
Geophys. J.R.astr.Soc., 57, 397 - 430,1979
(6) Breyer, F.: Ergebnisse seismischer Messungen auf der sUddeutschen
GroBscholle besonders im Hinblick auf die Oberflache des Vari-
sticums, Z.deutsch. geol. Ges., 108, 21 - 36, 1956
(7) Carle, W.: Die Warme-Anomalie der mittleren Schwabischen Alb, in
Approaches to taphrogenesis: J.H.Illies, K.Fuchs, eds., Schwei-
zerbart'sche Verlagsbuchhandlung, Stuttgart, 207 - 212, 1974
-585-
(8) Cerveny, V., Langer, J., Psencik, I.: Computation of geometric
spreading of seismic waves in laterally inhomogeneous media with
curved interfaces, Geophys. J.R.astr.Soc., 38, 9 - 20, 1974
(9) Geyer, O.F., Gwinner, M.P.: EinfUhrung in die Geologie von Baden-
WUrttemberg,Schweizerbart'sche Verlagsbuchhandlung, Stuttgart,
1968
(10) Hagedoorn, J.G.: The plus-minus method of interpreting seismic re-
fraction sections, Geophys.Prospect., 7, 158 - 182, 1959
(11) Hughes, D.S., Maurette, C.: Variation of elastic wave velocities in
granites with pressure and temperature, Geophysics, 21, 277 -
284, 1956
(12) Variation of elastic wave velocities in basic igneous
rocks with pressure and temperature, Geophysics, 22, 23-31, 1957
(13) Jentsch, M.: A compilation of data from the 1978-79 Urach, Baden-
WUrtemberg, seismic-refraction experiment, Open-file report 80-1,
Geophysical Institute, University of Karlsruhe, 1980
(14) Jentsch, M., Bamford, D., Emter, D., Prodehl, C.: A seismic-re-
fraction investigation of the basement in the Urach geothermal
area, Southern Germany, submitted to Geol. Jb., 1980
(15) Majer, E.L., Mc. Evilly, T.V.: Seismological investigations at
the Geysers geothermal field, Geophysics, 44, 246 - 269, 1979
(16) Schadel, K.: Die Geologie der Warmeanomalie Neuffen-Urach am Nord-
rand der Schwabischen Alb, Proc. Seminar on geothermal Energy,
Brussels, December 6 - 8, 1, 1 - 80, 1977
(17) Timur, A.: Temperature dependence of compressional and shear wave
velocities in rocks, Geophysics, 42, 950 - 956, 1977
(18) Volarovich, M.P., Gurvich, A.S.: Investigation of dynamic moduli
of elasticity of rocks in relation to temperature, Bull.Acad.
Sci.USSR, Geophys.Ser., engl. Transl., 1 (4), 1 - 9, 1957
(19) Wohlenberg, J.: Geophysikalische Untersuchungen in der Forschungs-
bohrung Urach, Bericht des Niedersachsischen Landesamtes fUr
Bodenforschung, September 1978
-586-
COMBINED REFLECTION AND REFRACTION MEASUREMENTS
FOR
INVESTIGATING
THE GEOTHERMAL ANOMALY OF
URACH
R. MEISSNER, H. BARTELSEN,
T. KREY+ and J. SCHMOLL+
Abstract
The combined seismic reflection and refraction measure-
ments in the Urach area have revealed many details of the
geothermal anomaly. Horizontal velocities show a minimum at
the center of the anomaly in the uppermost part of the
crystalline basement. (Vertical) stacking velocity show this
velQcity minimum to extend down to the bottom of the crust
showing up to 5 % velocity variations at deep crustal levels
as compared to the average value along the profile I. This
effect is stronger than predicted from theoretical calcula-
tions of the thermal effect of a temperature difference of
50 or 1000 C. The center of the Urach anomaly coincides with
a relative maximum of crustal thickness and with a change of
dip of deep crustal reflectors. Some indications of faults,
possibly important for magma breakthrough, could be mapped
along the profiles.
- 587-
1. INTRODUCTION
The goal of the combined seismic investigation of the
geothermal Anomaly of Urach was the derivation of a detailed
picture of the subsurface and the determination of correlations
between temperature and seismic parameters, such as velocities,
absorption, and reflection coefficients. In order to achieve
this goal three basic approaches were made.
(1) Theoretical calculations based on laboratory experi-
ments.
(2) The field experiment performed between 22 July and
5 August 1978.
(3) A basic interpretation of the refraction and reflection
work of the field experiment.
Results from the three approaches turned out to be so volumi-
nous that only a small part of them will be presented in this
paper. Moreover, all data are not yet processed, and a final
interpretation must be postponed to 1981.
-588-
and on the average temperature level. For one Kb or about 3.5
km depth -1and for 1500 C the value for AVp I 4 T is about
- 100msec/1000 C. For 2 K~ and 2000 C average temperature
-AV / AT is about 60rnsec /1000 C a value similar to that re-
p
cently published by Christensen (1979) for 2 Kbar and sialic
rocks. ~4 gives another representation of the temperature
influence on velocities, showing the accuracy which is needed
in the determination of velocity values for varying tempera-
tures at constant pressure (depth) in order to measure tempe-
rature effects. As seen in the left diagram, a resolution of
50 m/sec is needed for a change of 50 0 C, a situation repre-
senting the Urach geothermal anomaly. The whole temperature
effect gives only a 1 % change in velocity (for AT = 50 0 C)
which is certainly small compared to effects of 16-20 % caused
by inhomogeneity and anisotropy. There is some hope that late-
ral variations in temperature by their very nature cause
smooth and continuous changes in the isotaches (= lines of con-
stant velocities) whereas materials and anisotropy may change
more discontinuously and abrupt,preferably along fault zones,
and stay constant over large areas. The pure thermal effect
on the other hand, may be increased by hydrothermal alterations
generally also directed towards a decrease of velocities for
increasing temperatures. These effects, however, are very
hard to control. For the investigations in the Urach area, chan-
ges of about 50 mls representing 500 temperature variations are
expected from the pure thermal effect in the center of the
Urach anomaly for the first 3 to 4 kilometers. If the anomaly
in temperature would stay constant also at greater depths,
velocity variations of less than 50 mls are expected for the
lower crust. Such a small figure is very hard to measure with
a normal field set up. It provided the preconditions for the
most concentrated field work existing so far for monitoring ve-
locities in a crustal cross section.
3. THE FIELD EXPERIMENT
In order to measure the expected small variations in velo-
city due to thermal effects with the highest accuracy pOSSible,
- 589-
both the horizontal and the vertical velocities must be gathe-
red by the field set-up. Measuring the two directions in velo~
city on two perpendicular profils would also provide some limits
on the anisotropy and possibly on lateral changes in anisotro~
py. The field set-up on line I consisted of a geophone line of
about 23 km length consisting of three 48 trace independent re-
flection units. An 8-fold coverage for reflection and a 16-fold
coverage in refraction was achieved by arranging shot points
at 1.3 km intervals. For the picking of the first refracted arri-
vals and some wideangle arrivals the reflection line was ex~
tended to more than 100 km by 27 portable refraction stations.
A slightly different arrangement was made for line II crossing
line I nearly perpendicularly. Fig. 5 shows the positioning of
the two lines with regard to the known heat flow information.
The whole field arrangement is described in more detail in
the "Summary and Half Yearly Report" (4). Also the very diffi-
cult and time consuming correction problem and first results
from the refraction evaluations are described in (4). Also for
the geologic setting of the experiments reference is made to (4).
- 590-
basement with a maximum lateral anomaly at about 1000-2000 m
depth. In this depth range there is a 100 m/sec velocity ano-
maly with regard to the center of the profile which is the
double amount as that calculated before. Averaged over the
whole depth range of 3 kID, the value of about 50 m/sec,
corresponding to 500 temperature difference, agrees with the
theoretical value. As shown in the abstract, horizontal velo-
cities are slightly higher than vertical velocities, as mea-
sured in the Urach III borehole, showing a varying anisotropy
of about 5 to 10 % in the average, a value quite common for
Gneisses.About 7 record sections up to 150 km in length have
been plotted so far from the refraction stations of the insti-
tutes. Fig. 7 gives an example of the data quality between
50 and 110 km distance. Within this range often a double re-
flection from the lower crust and the crust-mantle boundary
appears in the sections. The critical angels are between 55
and 75 kilometers. A small dip towards the ENE is observed
on the western part and on the extention of the profile, indi-
cating the rising flanks of the Rhinegraben system. The central
ata from these refraction studies will be incorporated into
the reflection section, shown in the next paragraph. The velo-
city-depth function obtained from the refraction studies was
already shown in the abstract (5).
- 591 -
in the middle crust is seen at the le~t part of the profile,
probably caused by diapirs. The process of the formation
of the Rhinegraben flanks (8) seems to extend into the Urach
area, thereby causing a tensional stress at lower crustal and
upper mantle levels. This is possibly the reason for the Terti-
ary volcanism in this area. Towards the ENE the quality of
the records deteriorates, most probably because of strong
noise sources in the vicinity of some cities. Figure 9 shows
the same profile in form of a line-plot of clear reflections
without exaggeration, with the data from the long range re-
fraction incorporated. Figure 10 presents the first stacked
record section of profile II. The quality will still be im-
proved by more sophisticated correction processes. Same dip
of lower crustal reflectors, however, is clearly defined.
A change of dip in this profile, too, is seen in the Urach
area, about 8 km SSE from the intersection. More fault zone
indicators are seen along this profile than on profile I.
Velocity ~nvestigations so far are confined to the long
profile (I). From the non-stacked sections with a common depth
pOint arrangement along the line first a rough stacking velo-
city was obtained from the curvature of reflections. These
values were used for first dynamical corrections which later
were improved by controlling the 8 single-coverage sections
along the line. Stacking velocities were also obtained and
controlled by applying a number of different velocities for a
trial stacking. The stacking velocities were then linearly
interpolated along the profile and used for the dynamical
corrections and the stacking of the data, as for instance
shown in Figure 8. The velocity-depth function is similar,
but more accurate than that presented in (5). There are de-
finitely deviations fram the average stacking velocities along
the profile. These deviations were mapped according to a
method by Krey (9). Figure 11 shows these lateral deviations
of the stacking velocity along the profile, separated for
depths-zon~between 5 and 7 seconds and those between 7.5 and
-592 -
are similar for both depth ranges; (ii) ,the low velocity area
near the center of the Urach anomaly reaches values up to
5 % lower than the average at the bottom of the crust, i.e.
much more than a pure thermal effect could do; (iii),the low
velocities below the center of the Urach area at great crustal
depths seem to be a continuation of those in the uppermost
part of the crust; (iv), the method of mapping lateral velo-
city variations from the reflection curvature can be performed
with about % confidence level and therefore provides a
powerful tool for mapping thermal anomalies.
6. CONCLUSIONS
As mentioned in the last paragraph the method of gathering
horizontal and vertical velocities along combined reflection-
refraction profiles provides a resolution of 1 % in velocities.
It is therefore well suitable for mapping those small velo-
city variations which may be due to geothermal anomalies.
For the center of the Urach area velocity anomalies up to 5 %
have been found at depths up to 30 km in the crust. Also the
lateral extension of this velocity anomaly could be located.
Moreover, the detailed 8 fold stacked reflection sections
show many items which may be strongly related to the geo-
thermal anomaly, such as a change of dip, indications of
faults, and a relative maximum of crustal thickness in the
Urach area. More refined filter- and correction techniques
applied to profile II, a thorough investigation of absorption
and anisotropy effects, and a complete integration of the wide-
angle data of the portable refraction stations will still
have to be carried out in the near future in order to complete
the evaluation of the data.
-593-
-+--,5,---,--5.... 8 --,---,---,-_6......6_VP (km /sec )
1.--,-_5.....
rn m
P (kb) Q Cl
4 8 12 16
KH (%) 4 8 12 16
KTotol (%)
,
2 .:.4;, ~
,j-
2
#/I I-
~ I
I
I
!
"f
4 I
I
6 6
P(kb) P (kb)
Vz v. -
KTotol = - - . 100
Vz
" Total = - V
_ . IUU
z
200
150
100 +-- -
50
100
-L::.Vp In m/sec
200 2
3
GRANITE
150 150
2 kb
2
100 100 -
'.--_. '3 0 - -
_....
-=-==--===--
10 zo 30 .0 !Okm
<00_
-3000
M
I
50 "
100
24
ISO
32
2DO
40
2SO
41
lOO " 104
o 1-'---+'-
-1
sec
Fig. 8 8--fold stac}:ed. and processed record section of ?rofile I, about 2-fold exaggerated. (in 4 parts)
o, 5 10 15 20
\
25
. 30
,
35 40 km
Wide angle
- rellectlon
~i\tr~~g~\¥: II o o
2
3
I- 4
r 1:
'"'<l'<l
5
--::- --~~ ~..:::-:=-:.. [ 15
- - -----=-.::-;:: _._-
. .....,;;;.~ ~
----- ~-~-- -~ -~ --=.- 20
- ::-'" -=...:: ~ ~~
-L ---,-"--, -- 8
- ~, ~ 25
9
" ~ 110
t30
sec km
Urath Line 2
, 1
0 T oJ
TY f F' ~----
-5
r-4 ,- -5 ....
-;
~ I I I I I I I I I I l
0 100 200 300 '00 500 600 700 800 900 [Dol. Potnl Numbers J
-602 -
S[ISMIC STUDIES FOR THE DETERMINATION OF THE CRUSTAL STRUCTURE IN
THE AREA OF THE GEOTHERMAL ANOMALY I N TUSCANY
P. GIESE and P. WIGGER 1
C. MORELLI and R. NICOLICH 2
Institut fUr Geophysikalische Wissenschaften
der Freien Universitat Berlin
2
Istituto di Miniera e di Geofisica Applicata, Trieste
ABSTRACT
Refraction seismic measurements were carried out in July 1978 and 1979
in Tuscany and Latium with the goal to receive the crustal and upper mantle
structure and the velocity distribution in the geothermal anomaly of
Tuscany.
The main results are: The crust of Tuscany is thin. On reason of the
relative high velocity and the missing of distinct velocity inversions it
must be· excluded that a greater extent of molten material in the upper
crust could be found. However, the transition zone to the upper mantle o
shows layers of extreme low velocity, at least temperatures of 600-700 C
have to be expected in this region.
Besides, in the transition zone under the area of Lardarello a jump of
6-7 km to the upper mantle was found possibly corresponding to the jump in
the basement of that region.
Model calculations which were made for the temperature distribution in
this area are discussed.
1. I NTRODUCTI ON
The goal of this project is to study the crustal and upper mantle
structure and the velocity distribution in the geothermal anomaly region
of Tuscany. The results should show in which way and how far seismic crust-
al studies can help to solve geothermal problems, especially the question
of how to determine the recent temperature distribution down to 30-40 km
depth. Seismic velocity as well as the electrical resistivity depend on
temperature. In the temperature range from 700-800 0 C sialic rocks begin
to melt and the seismic velocity as well as the electrical resistivity are
distinctly reduced in this state. Thus the distribution of the seismic vel-
ocity can give hints to the actual temperatures at greater depth. These
results can be used as basis to considerations dealing with the geothermal
evolution of the area of this study.
Therefore this program comprises two parts
a) The study of the seismic velocity distribution and
b) The study of the electrical resistivity distribution (HAAK, V.
and SCHWARZ, G., 1979)
-603 -
2. FIELD MEASUREMENTS
The seismic field measurements were carried out in July 1978 within
an Italian-German cooperation. From 6 shot points (SP) 4 profiles (07, 08,
09 and 10) have been recorded (fig. 1). These shots simultaneously have
been recorded within the Italian geodynamic program on a line traversing
the Apennines between Elba and Ancona. These data have been available for
our interpretation. Additional measurements were made in July 1979 with
a shot point near Rome on a profile running northwards .
• Bologna
Seismic Relraction
Measurements
Position Map
. 'I
110
~
.. Shot Points(SP)
Recording Lines: + '. ,.,
- Geothermal Prog . 07 Flr enze '" 11
'.
°
,••• Italian Prog .
_ _ _ i
50 100 SP '.\,,, Ancona
L,vorno N.
~
~9 SP 04••••• ~~.A.'05
. " -07 ~~ ••••••••
• La rderell 0 ......... ....
-i
<~
::::;
rt>
VI
rt>
~.
VI
3.
n
o 10 50 km
3
0
C-
rt>
~
0 0 Rome
V> -i>
, Line
M-
::::;
rt> PS<;/= U km(s ~6:2 - :: ) ~16:2 "'",. 10
-i
<::
VI
n
:::5
'"
'<
n
-s
<::
VI
M- 5 7 km/s 753 3 5 53
,zzc:z:::
7i . km/ . • .,.,.
r"I v v Uppe r Tus ca ny Crust
p p o
10 10 I I I I - t-I-t----i 10
Trans i tion Zone
20 20 I I'" ~ E:::4o 120
[TIl 4.0 < VM in < 6.0 km/s
.......
30
f:> 30 f:8l
~
....::::: ..... I---l->----- ---+--+--1130 t:::l Upper M ant le
km SP C km SP C SP S km
SP N
z
J
The special structure of the upper 10 km is explained in the next
chapter.
Before the first deep reflexion no or only a very slight crustal vel-
ocity inversion is visible which is also manifested in the normal values of
the mean velocity. They are between 6.1 and 6.3 km/s. The pMP-groups which
come from the crust/mantle boundary show a very complex structure. It is
possible to correlate separated phases which are caused by alternating high
and extreme low velocities layers thus indicating a wide transition zone
between crust and upper mantle. The top of the transition zone is proved by
the first reflexion of the pMP-group. The velocities reach values between
7.5 and 7.8 km/s. the corresponding depths are at about 20 km. Between
these first reflexioffiand the phase which comes from the Mohorovicic dis-
continuity material with extreme low velocities (until 4.0 km/s) must be
existent. The depth of the Mohorovici~ discontinuity in the region of Tusc-
any shows values of 21 km (prof. 08) to 30 km (prof. 07). East of SP B a
strong increase of crustal thickness to about 40 km can be observed. On pro-
file 07 beneath the area of Larderello a jump down in the crust-mantle-
transition zone to north of 6-7 km is visible. possibly corresponding to a
jump in the basement south of Larderello (fig. 2). In the record section of
profile 12 which is situated outside of the geothermal anomaly the first
discontinuity does not exist. But in this case the crust-mantle boundary is
very significant. Pn-waves which reach the upper mantle could not be de-
tected in most of the cases. This hints to a further velocity inversion in
the upper mantle.
The results were proved by ray tracing and gravimetric model-computa-
tions.
4. NEAR SURFACE STRUCTURE
For the structure of the upper 10 km the following three statements
can be made:
a) In the whole area under investigation the first arrivals come
very early (about 0.5 s after the zero time) and.reach velocities
above 6.0 km/s.
b) From reflexion seismic measurements (BATINI et al., 1978) and
from our own data a reflexion horizon is known in 7-8 km depth.
c) Magneto-telluric measurements (8AAK et al .• 1980) show a cover
with high electrical conductivity up to a depth of 7-8 km.
-~-
These statements, completed by the known geology, are arranged in fig. 3.
l
fill "C\Oeh'f'lonaua napp,t
( L TII •• /P~r,"I." U r, ••• c
r" •• ".
!*JIlIIIi,~!t1:~IJlMllf0'~~i;ii<~
Fig. 3: An idealized model of the near surface structure
The first arrivals were related to the layer of phyllites and quart-
zites of the Lower Triassic/Permian, which is the basement for the gypsum,
anhydrite and dolomite of the Upper Triassic. To explain the high electri-
cal conductivity down to a depth of about 7 km it must be assumed that
either the basement is strongly fractured down to this depth and filled
with saline water or very well conducting sediments must exist at these
great depths. The velocities of the Pg-wave (0.6-6.2 km/s) argue
against this strong fracture down to 7-8 km. As one might infer from the
geology of the Apuan Alps, this "basement" - consisting of phyllites and
quartzites - can be thought as a nappe. In this case Palaeozoic material
and anhydrites of the Upper Trias could be found upon the old antoch-
thonous basement - composed of granites and gneisses. These anhydrite
layers or lenses could be a reservoir for hot water and vapour.
5. GEOTHERMAL ASPECTS
In order to infer temperatures from seismic parameters the dependence
of velocity on temperature can be used. CHRISTENSEN (1979) has measured
this dependence for several rock samples up to a temperature of 500 0 C at
2 kbar. Fig. 4 (solid line) shows the result for a granitic sample. At
temperatures above 350 0 C a strong decrease of P-wave velocity occurs. Be-
cause in the upper crust there is no or only a very slight velocity inver-
sion visible a temperature range of 300-400 0 C for the bottom of the upper
crust can be assumed. A comparison of the computed velocities of the trans-
ition zone (about 5. 0 km/s) with the extrapolated v(T)-plot (dashed line)
shows that for granite material these velocities were reached at 600 to
700 0 C.
- 607-
Fig. 4:
6.2 GRAN IT SAMPLE
• = 2.62 glcm J The dependence of velocity on
p = 2 kbor
6.0 temperature after CHRISTENSEN
5.8
~ _ _L - - L_ _~~ _ _~ _ _~-L~~__
5.6
5.4
5.2
5.0 \
\
vp
km/s
4.8 \
3 4 5 6 7 Vp 8 km/s 3 4 5 6 km/s
0 0
07
z*
o
=22
=10.5
km
,3km
z'" = 17 km
K = 6.0·1O~. D • 10.5 km
0(01 = 3-7 HGU K , 6j)'It?t,zloc
plol, 3- 7 HGU
10 10
20~-----;'~~-~k--
300-400'C-
Z
~ ~
• z V>
Z
- '"
.. :'!
;; ~
UJ
I-
~V>
__J
,!Ci
depth oQ
dePth[:OO~'C~
30
km
__J~~~~~!~
30
km~__r -_ _ +-__.-__4-__' -__-r~
o 400 800 o 400 800
- 608-
The temperatures of 300-4000 C in a depth of 22 resp. 17 km estimated
after CHRISTENSEN are reached by a heat flow of 1.2 resp. 1.4 HFU. This is
in contradiction to the measured heat flow (CALAMAI et al . , 1976) at least
in the southern part with the thinner crust.
From the very "normal" structure of the upper crust on one hand and
the estimated high temperatures in the transition zone on the other hand it
can be assumed that the thermal anomaly on the surface is caused by con-
vection of hot fluids and/or gases which are ascending through fault
systems reaching the transition zone. This possibility agrees very well
with the extension tectonicsin Tuscany. In this frame we have to consider
the discovered fault system beneath the area of Larderello .
In order to find a conformity for the "cold" upper crust and the high
temperatures in the transition zone a heating of the lower crust is postul-
ated. By application of an insteady heat flow and an increase in tempera-
tureby 400 0 C in a depth of 35 km the advance of the heat front was com-
puted. At recent time this front shall reach the boundary of the lower/
upper crust.
depth
30~-------+----+---r- ~--~
km
-609 -
6. CONCLUSIONS
After the given results it is to assume that beneath the very well
known layers of gypsum, anhydrite and dolomite of the Upper Triassic
further tectonic complexes which could be a reservoir for hot water and
vapor, can be expected in deeper tectonic floors.
Deep-reaching fault systems like that one discovered under Larderello
must be present which enable hot fluids and gases to ascend from a amoma-
lous -hot-lower crust which serves as source for the hot material to the
near surface structures.
These structures are detectable by suitable profiling of reflexion
and refraction seismic measurements.
The special structure of the crust/mantle transition zone seems to
be limited to Tuscany. In the east and southeast of Lago Trasimeno - out-
side the Tuscany - the anomalous structure is absent.
REFERENCES
BATINI, F., BURGASSI, P., CAMELI, G.M., NICOLICH, R., SQUARCI, P., 1978:
Contribution to the study of deep lithospheric profiles: "deep" re-
flecting horizons in Lardarello-Travale geothermal field. Mem. Soc.
Geol. Ital.
CALAMAI, A., CATALDI, R., LOCARDI, R., PRATULON, A., 1976: Distribuzione
della anomalie geotermiche nella fascia preappenninica tosco -
laziale. Simposio internacionale sabre energia geotermica en America
Latina. Citta del Quatemala.
CHRISTENSEN, N.I., 1979: Compressional wave velocities in rocks at high
temperatures and pressures, critical thermal gradients, and crustal
low velocity zones, J. Geophys. Res., 84, 6849-6857
- 610-
DANTZ, D.: Ray - tracing in two-dimensional inhomogenous media. Dr. Thesis
Freie Universitat Berlin (in preparation)
EL-ALI, Hussein, 1978: Geothermische Modelle fUr ein Profil zwischen SUd-
adria und Tyrrhenis BerlinerGeowissenschaftliche Abhandlungen, Reihe
B Heft 7, Reimer, Berlin
FIELITZ, K., 1971: Untersuchungen zur Temperaturabhangigkeit von Komress-
ions- und Scherwellengeschwindigkeiten in Gesteinen unter erhohtem
Druck. Diss. TU Clausthal
GIESE, P., MORELLI, C., NICOLICH, R. WIGGER, P., 1979 a. 80: Seismic
studies for the determination of the crustal structure in the area of
the geothermic anomaly in Tuscany, Reports No. 1-6, CEE Contract No.
486-78-1 EGD
HAAK, V., SCHWARZ, G., 1980: Distribution of the electrical conductivity
in crust and upper mant.le in the area of the geothermal anomaly of
Tuscany/Italy. This volume
KERN, H., and FAKHIMI, M., 1975: Effect of fabric anisotropy on compress-
ional-wave propagation in various metamorphic rocks for the range
20-7000 C at 2 kbars, Tectonophysics, 28, 227-244
MORELLI, C., GIESE, P., CARROZZO, M.T., COLOMBI, B., GUERRA, I., HIRN, A.,
LETZ, H., NICOLICH, R., PRODEHL, C., REICHERT, C., ROEWER, P., SAPIN,
M., SCARASCIA, S., WIGGER, P., 1977: Crustal and upper mantle struct-
ure of the northern Appennines, the Ligurian Sea and Corsica, derived
from seismic and gravimetric data. Boll. di Geof. teor. ed appl., No.
75/76
NICOLICH, R., PELLIS, G., 1979: 11 contributo dei dati geofisici per 10
studio delle strutture crostali della provincia geotermica Tosce-
Laziale, Universita degli Studi di Trieste, I.d.M.e.G.A., Rap. No. 38
REUTTER, K.J., GIESE, P., CLOSS, H.: Lithospheric split in the descending
. plate: Observation from the Northern Appennines. Tectonophysics,
in press
REUTTER, K.J.: Personal communication, 1980
RUBER, K.R., THYSSEN, F.: Messung der Schallgeschwindigkeit beim Obergang
fest-flUssig. DFG Kolloquium, Geowissenschaftliche Hochdruckforschung
Bad Honnef, pp. 13-14, 1978
-611-
APPENDIX
SP N SP B
@WNW Profil 09 IV? ESE
6.0----
_ _10 _ _ _ _ _6_., _ _ --~---
a:i
_ 20 1,6.."z;1..::.
o 10 50km
~ g WSW ENE
0::0 SP C SP B SP A
I 02 Q? Profll 03 '9? 04 @ 05
5.5
_,0
6.0
----
_ _ 6.2 - U6.0
- - - 7.5-7.8 -
SP S SP B
SW~ Profile 10 ry? NE
11.0 ' , - , - - - 6.5
-!I.O
_ _ 5.,5 - _
20 -+-- --"oc-'
J5~-~---~--~---r---,~--r-~,r
km 0 2 J·10- 4 OC/cm
- 612-
3 4 5 6 km/s 3 4 5 6
0 0
08 08
z· , 13 km z. : 1L km
0 : 10,5 km 0 : 10,5 km
K : 6.O'1tY,~!OC K , 6.1HtY,Z IOC
piO): 3-7 HGU piO): 3-7 HGU
10 10
20 &OO·C 20
600·C
~
Z z
:;; 9"
<!o ~
.!f
'Ov
!;;
.!i
depth
'"Co depth
'Ov
~
30 30 Co
km km
0 4 0 800 T 12000 C 0 4 0 800
3 4 5 6 3 4 5 6 km/s
O-h~~-=~+---~--~----~----~ O-k--~-==±--~---4----~---'
09
z
20 ~ --~~r----~r--+-7~~----~
:;;
!!
'Ov
bOO ·C
depth
30 HFU
k m /----.----+----.----+-----r----,---J
o 400 800 T o 400 800
- 613-
Vp/VS RATIO AND ITS CHANGES IN THE TRAVALE GEOTHERMAL FIELD
Summary
1. Introduction
-614 -
relationships between elastic properties in a porous medium and the
pressure and the chemical characters of the pore fluid, even for non-homo-
geneous conditions, i.e., the formulae proposed by Gassman (1951) assume a
new generalized meaning.
For this particular area the above mentioned results could have been
applied by bearing in mind the need of bypassing two major difficulties
a) simultaneous recordings of artificial shocks over a sufficiently large
area by means of an extensive seismic network, b) adequate location of
shock sources relative to the network setting, in order to reproduce
successive identical boundary conditions for successive explosions and
detect P and S wave paths as much homogeneous as possible.
The first problem has been solved by designing and assembling a completely
radio-telemetered seismic equipment for 15 sensors, connected by a "Synchro-
propulse" system. The transmitters are of a pulse code modulation type, at
time division, with synchronism guide coming from a repetitor which also
receives the signals from the 15 stations to be transferred to the base
station where the receiving system decodes the signals in both analogical
and disital types. The device perrorms two main advantases ; a) a sinsle
transmitting frequency is practically operating, and b) all transmitters
- 615 -
are interchangeable and rapidly moveable from one area to another (in less
than three days) so that the whole network can be re-arranged in the field
with much freedom in selecting the station points. This is of basic
importance in identifYing the best signal-to-noise ratio and spacing
between stations.
The second goal has been accomplished by taking into account the geological
setting of the Travale area as basically affected by a graben structure
with emerging edges of carbonate to the SW and metamorphic rocks to the NE
(fig. 2). The former is intensely faulted in successive steps, along the
"apenninic" major tectonic trend, parallel to the graben structure and
contains the geothermal systems clustered along the peripheral graben
fractures underlying a thick argillaceous sedimentary layer that builds up
the unpermeable cover.
Given the active fractures extend down to deeper structures, the seismic
network has been placed along three profiles parallel to the graben (i.e.
north-westerly oriented) and spacing about km between Anaue and Montalci-
nello villages (figure 3) that define also the geothermal field extension
by deep wells. The explosion points are aligned on the median axis perpend-
icular to that of the graben at an average distance of 14 km in a quarry
or limestones apparently well massive and geologically rooted in the
Montagnola Senese Hills forming the north-easterly metamorphic edge of the
graben. This configuration has been assumed to allow sensibly comparable
paths in the supposed homogeneity of the refracting layer at higher
-616 -
I , , J.
~
' '41
:
~
"
*
II [I
"
• "
- 617 -
sw NE
T22 R 14 Poggio al Venlo
Lagoni di Travale
I I
'000 .000
"- "- "- "- "- "- "- "- "- "- "- "- "- "- "- "- "- "- "- "- "- "- " t-6000
'"
00
eoootlllrr)-~
'000-1 ...•• 1+,,- "- 1-1000
~
§ "- "- "- "- "- "- "- "- ""- "- ""- "-
D. .. 2
t{~~.iJ
1
.000-1 Q [[[]J \l
4 \ \1 1 I I 1 1 • ....
0 .000
• • ,000 lOGO.
-
Fig. 2 - Geological interpretation of seismic lines through the Travale
geothermal field. 1 : Neogene deposits; 2 : Ligurid nappes; 3 : Tuscan
nappe (Burano formation); 4 : Tuscan crystalline formations; 5 : "deep
reflecting horizon"; 6 : location of calculated velocity functions
• R.di,ondo/i
'0'
.........
- . -:-. -:-. -:-.-:- .-:- :1.
Qu"tt.rnary
UPhoc.n.
r-
'1~
~- - -- --" -
,
t-• • ,
~ \
Phoc.nt
U . W~c.n. c:=J
Eoc.n.
U. JUtl'SIIC
g S..... I,.,. ",.ns ",.,.. .. 1.lon, ... opl'l.olll.t..
"yseh - 1., •• '5 lo, ... ",.on",
"I( ( -_f~ .Il . " ' 111 " 0''''·'
r
Ol19O(tnt
C,.I,ac~ I: : :=I 4,100,.( ... ndl l on.,
rStaql ,a1
t-4.CI9no) oI ntj voi r 1<010\1' .0 "h.lf'''
U Jurlssic
~
AwlOla " If''' "", ,If !!O (1\. '1), 1,1T\ ."IQI\f" "' .ss ..... 1•..,..,,10". ' 1oltat ,' ,."
U. Triassic h""'''slo".,. olnd trI.agr'l .s •• n I'''''"s ton", WI'' ' .An", atlu· ,• .,."
-619 -
,n.
•
TRV-l
I
".
, ...
o
"0
,
tOI,
.
200
.
00 ...
\l
, "., ...
I .-. lAH-6
...
I
'00
I
sw NE
T22 Poggio at Vento
Legoni eli Travat.
I
lOOO
.-
000.
noo
D~@
• • I
D• rnrn
•
'V
• ~--~~----~--~,
.....- -
-620 -
velocity (4.5 km/s) and therefore eventual Vp/VS changes for different
azimuths would have been ascribed almost completely to paths across the
fracture system at different thermo-fluid-dynamical conditions.
REFERENCES
A. NERCESSIAN, A. HIRN
Summary
-622-
1.- DESIGN OF THE METHOD
Due to their mode of genesis, volcanic and geothermal fields are em-
placed in structural contexts which are fundamentally different of the near-
ly horizontal, layered, sequences typical of sedimentary basins. Usual
seismic methods, designed for prospecting such sedimentary provinces are
aimed at detecting interfaces i) across which velocity increases with depth
by a finite jump, ii) which are nearly plane and near to the horizontal
iii) which have a certain continuity in the region, particularly for the
refraction method.
Recently, following Aki (1,2) a method has been developed for the
investigation in three dimensions of the seismic structure of the earth's
lithosphere. Differences in arrival times of seismic waves generated by
distant earthquakes situated in different azimuths and recorded by a two-
dimensional array of seismic stations have been inverted into heterogenei-
ty of structure underneath. The wavelength of these teleseismic signals
and steep incidence on the array are suited for the description of large
scale, decakilometric, heterogeneity in the 200 upper kilometers of the
earth (e.g. under Hawaii (3) or Yellowstone (4).
We have attempted to adapt this procedure of 3D investigation to the
smaller scale and more superficial targets of interest in geothermal pros-
pecting. For this purpose significantly shorter wavelength waves from
borehole shots are used. These shots are emplaced in different azimuths
from the region to be investigated and at a distance of it (of the order
of 80km) suitable tu use the Mohorovicic discontinuity between the crust
and mantle (at about 30km depth) as a mirror which sends the incident
waves back from beneath towards a temporary seismic array. These waves
sample the suspected anomalous zone in the upper crust beneath the array
with an oblique incidence suitable for a resolution of anomalies compara-
ble in depth and horizontal coordinates and for optima cross-sampling of
the upper crust by waves coming from different azimuths. As the shots are
situated outside the perturbated zone and the spreading of the ray bundle
is maximum in the upper crust under the recording array, the region sam-
pling of the structure of interest is adequate.
2.- APPLICATION TO THE MONT DORE MASSIF
The Mont Dore massif (Fig.l) resulting of volcanic activity between
3.5 and 0.1 m.y. is situated within the Auvergne, on that part of the her-
cynian crust of the Massif Central reactivated in tertiary times by graben
- 623-
formation and initiation of an extended volcanic zone. Evidence of deep
perturbation to the hercynian type of crust has been derived from two
North South reversed seismic profiles in the Uimagne graben (Him and
Perrier, 1974) and along the volcanic alignement Mont Domes, Mont Dore,
Cezallier, Cantal (Perrier and Ruegg, 1973) where the crustal thickness is
seen to be reduced to 25km. Although the scarce data available on the upper
mantle indicate that the velocity in the rocks immediately beneath the
crust may be reduced with respect to normal, the crust-mantle boundary is
a clear reflector for seismic waves, at least on a regional scale which
allows to use it as a mirror in the method described above. To send a rela-
tively undisturbed wavefront under the region of interest, thus leaving
perturbations to be attributed to the upper crust under it, reflection on
the Moho at places situated beneath the rim of the Limagne graben or under
the Chaine des Puys in the E and NE of the Mont Dore have been avoided.
Thus only special shotpoints to the N (Gannat), W (Eymoutiers), SW (Auril-
lac), SE (Margeride) were used (with some additional quarry blasts) to
generate waves. The recording array was situated in a 15 x 20km region
centered on the city of Ie Mon Dore. One point of measure for an average
3km2 was obtained for each shotpoint, resulting in a total of about 480
data values of arrival times and amplitudes of the PMP wave reflected on
the Moho midway between shotpoint and array.
Granite + +
+
MetaIlDrphics 'V
,....""
Sediments \III
Volcanics 0
-624 -
3.- WAVEFRONT PERTURBATION AS A MARKER OF STRUCTURAL
HETEROGENEITIES
- 625-
_626-
Figure 3 Residual maps
70
•
• •• •
eo • •
• • • ... .. - .-;+, ..t. __ ~--
• • •
•
•
• ,,,-
GAMMAT
630
70 70
, , -
.
-;~-+'.\'
"B'\''':+• • n-
"
....~ .. ..:-_
. , $1.;
"II
~ i;~"f.
• · )+r-+n
•
...
~It
I
1+":
,-
~"~:
eo
\
eo
- 627-
rences of velocities as large as 20% with respect to a mean value of
5km/s ~e documented; ii) if divided into three classes of values,
one between-0.078and +0.078, one more negative and one more positive,
the residuals may be geographyca11y contoured for each shot as a juxta-
position of irregular patches of the order of 3km in dimensions. This
indicates small scale heterogeneities and sharp lateral changes; iii)
if residual maps from shots in different azimuths are superposed, there
is no general geographical coincidence of patches of the same class,
which would be the case if elevation was not correctly compensated or
if the major part the anomalies were attributable to local station
effects.
aT...
fl T~ E --2:l dn· + K
1 a n. J shot
J
tJ.n = tJ.T
- 628-
where C6T and C6n are the covariance matrices of 6T and 6n respectively.
Since the delays and slowness variations are independent variables, these
last matrices are reduced to their diagonal elements. Moreover, the cova-
riance matrix of the computed parameters e
T -1
~ C6T A + C6n
-1)- 1.
provldes
an error estimation for the solution.
Several inversions are carried out to control the stability of the
results with respect to different discretizations (size of blocks,
thickness and number of layers) and to outline stable features Jf the
structure.
6. - GENERAL RESULTS
On the basis of early geological models of the Mont Dore caldera
which located its bottom at + 500m (6), first inversions included an
upper layer from + 900m to + 50Om. Instead of a concentration of slow
anomalies in this layer, a distribution of these anomalies within the
upper 2km was inferred. Besides, a tentative inversion of the delays in
terms of depth variations of a caldera filled with slow volcanic materials
resulted, like the previous ones, in a bad fit which indicates the hete-
rogeneities to be located deeper in the basement. This leads to the choo-
sen 3 layered model, the main features of which are, for each layer
Layer 1 :from 1 to -lkm. A well defined low velocity zone (L. V.Z.) appears
within the limits: SE of La Bourboule, Le Queureilh, Puy du Barbier,
Roc de Cazeau, Perouse, W of Puy Paillaret, Sancy and Vendeix. The lowest
velocities are concentrated near Vendeix, and on a line Le Mont Dore -
Puy Paillaret. Local low velocity zones are also noticeable below Chas-
treix and north of Lac Guery. Higher velocities clearly bound this L.V.Z.
to the south, SE and ENE.
In the layer 2 : from -1 down to -3km, the central low velocity
zone is reduced to a smaller area between the south of La Banne d'Ordanche
and south of Le Mont Dore. Separate L.V.Z. are still noticeable below
Chastreix, in the area south of Puy Paillaret-Lac Pavin, E of LacGuery
and E of Puy Barbier.
In layer 3 , from 3 to 5km, the low velocities are concentrated on
the western limit of the previously considered central L.V.Z. in layers
1 and 2, i.e. south of La Banne d'Ordanche, south of La Bourboule and
Chastreix. This leads to define several types in the repartition of the
anomalies.
-629 -
Type 1 : s low anomaly from the surface down to Skm. This type is en-
countered locally, W of Puy Paillaret, below Roc Cuzeau, Le Mont Dore,
south of La Banne d'Ordanche, W of Vandeix. and marginally, below Chas-
treix. The points delineate the zone ~ where the slow anomaly is pre-
sent in layer 1 but decreases with depth. This zone extends on the surface
to the SW of La Bourboule and eastwards of the Sancy. ~ : slow anoma-
ly between 2 fast anomalies. The slow anomaly is defined with respect to
the mean velocity of the layer, which does not necessarily mean a reversal
in the velocity-depth function. This type is mainly reported in the NE of
Lac Guery and marginally near Lac Pavin and E of Puy du Barbier. ~ :
slow anomaly below fast anomaly. It is found S of Roc de Courlande, and
marginally, north of Lac Chauvet. ~ : fast anomaly. It is encountered
W of Chastreix and possibly E of col de la Croix Morand.
Type
I~
liD
1[~
:IY1IIillIJ
y~
•
s
;I
/111111.1
-630-
Interpretation of amplitude anomalies has revealed to be more diffi-
cult because measurements of amplitude are not only less accurate than
time readings but they may also be affected by local sub sensor conditions.
Furthermore, amplitude_ in such an heterogeneous medium are influenced by
scattering and focussing as well as by anelastic attenuation. Inversion
of amplitudes in terms of damping (dB/km) results in strong damping con-
centrated around the rim of the caldera in the upper layer. In layer 2
and 3 a weak correlation may be established between low veolocity and
strong attenuation. Further evaluation including that of spectral content
should contribute to refine the image wich at this time seems to associa-
te low velocity with attenuation.
ACKNOWLEDGEMENTS
The idea of the method originated in 1976 in response to dis-
cussions with Yan Bottinga then responsible of "ATP Transfert d'energie a
travers la croute" of INAG. The project was supported by CCE under contract
578 78 EGF through the Institut National d'Astronomie et de Geophysique,
France. Colleagues of the Institut de Physique du Globe, Paris
performed the data acquisition with the aid of Centre Geologique et Geophy-
sique de Montpellier and Institut de Physique du Globe , Strasbourg.
REFERENCES
1) Aki K.A., Christofferson A. and E.S. Husebye, Three-dimensional
seismic structure of the lithosphere under Montana LASA, Bull.Seism.
Soc. Amer., 66, 501-524, 1976
-631-
ETUDES SISMIQUES SUR LA CALDERA DU MONT DORE
(se;sm;c ;nvest;gat;ons on the Mont Dore caldera)
R. FERRANDES ~ A. GERARD:; P. MURAOUR ~ J. PERAGALLO :; G. PETIA~
~ B.R.G.M. - Orleans - :t::t LERGA - Bordeau:x: - :e:t:b eRG Garahy -
RESUME
ABSTRACT
-632 -
I - INTRODUCTION
-633 -
Pour chaque station ont ete effectues 6 tirs, repartis de la ma-
niere suivante
Station I 12 kg + 16 kg 4 + 8 kg 8 kg 24 kg non
12 kg realise
Station II 60 kg 48 kg 24 kg 12 kg 24 kg 56 + 8 kg
-635 -
Station reflexion Station de tir Localite
nO refraction
A S.3 Murat-Le-Quaire
B Le Pessy
C S.4 Le Genestoux
D S.5 Prends-toi-Garde
E S.6 ouest
Telesiege N° 496
F S.6 Est
IV - RESULTATS
- 636-
TEMPS-DOUBLES DES ARRIVEES DUES A DES REFLEXIONS
B 320
C 540
D 280 600
-637 -
IV.2. - Sismique refraction
Difficultes
cette premiere vitesse est assez peu visible sur les dromochro-
niques a cause du trop grand espacement entre geophones (50 a 100 m).
- V2 3 500 mls.
-638 -
Au vu d'une premiere interpretation, etablie a partir d'une hypo-
these de type "tabulaire", on a pu constater, d'une part que les dromochro-
niques theoriques calculees a partir du modele obtenu ne correspondent pas
tres bien aux dromochroniques observes, d'autre part que Ie fond de la cal-
dera propose n'evoquait que de tres loin ce que l'on aurait pu attendre
d'apres les donnees gravimetriques.
nous avons construit une coupe "d'initialisation" tres voisine de celle qui
est finalement presentee sur la figure 4 obtenue apres quelques ajustements
pour etablir une coherence admissible entre les dromochroniques observees
et les dromochroniques calculees par methodes dite des "fronts d'onde".
- 639-
v - ETUDE EXPERIMENTALE DU BRUIT SISMIQUE
- 640-
Disposant de 4 stations, on a done ealeule 6 rapports (dont 3
pour eontrOle).
V.3. - Resultats
Pour la frequence 1 Hz, on a pu mettre en evidence deux anomalies
du bruit sismique qui devient particulierement intense au voisinage imme-
diat de deux fortes anomalies de conductance d'origine profonde cartogra-
phiees par magnetotellurique.
-641 -
Fi 9 u re 1
630
0 2 SKm
!
I I
6. Station d'ecout e sismi Que
4
x POint de tlr refraction
~ Sismique ridlexlon
~ Courbes d'isoaltitude
51AlION I
so I'll
!;--------------------
Elmdcernont. ESPIlC8IrOI 10U
_I
of __ II-____,
L _ _
i
g"'optlOne
C""""'"
t
FIG. 2 : Schema du prafil sismique
- 642-
E',
~---------------------------------------------------------------
DROIIOCHROHlQUCS
• ArrhHS observ&s
AT'E'iv&s ca.lculfu
w E
I ,• • " • •.•••.
..... ,., .
.'. ".' .•....:.:"If .:".::.:,·
,',
.. :.. i;:;
,".
.. ':
.".. - . "'"
.~
. ~ . " ..
.........
..
# • • '" '.
......
:. :I'O(;;~:'~ :;.~:;:.:
~'.
.,." :1:,
..
f •
-'~"""
."'"
..
------
.....~~____~~________________~':~~M~____--------------;-----------~ ~
-
.. . . . . . . ~~.~~. . . . . . . . . ./.;_oh
........ ...,
...... ..........
RU'ncteu.a'" visible
_ t1irOlrs
R6frac:teur vblbltl! probfble
(4'bproh la
Li.1t. ,,1.I-deU de laque.U.. ab.lqua r"6flex!on)
tout r6l'Ncteur q,t posa1blll
4 2 5 1 2 5 1 2 5 1 2 5 1 2 5 1 2 5 T (s)
3
-
2~
- - -
'"........
- 0-
-
~
I I -
II
n- {J-
0,6 11- ~H1-
0.5
0,4
41' A A B A A A A A A A A A A A B A A a ualite
0,3
0,
Figure:5
0,
-
BRUIT SISMIOUE
GROUPE t9-LEGAL
II
,::::
§i'ii'?::[,:
111S1I
r.:'p.:"
T:1 sec.1v/em
FIGURE:6
-645 -
BRUIT SISMIQUE
GROUPE 19 - LEGAL
- 646 -
~I!: S!! ... 11
~
'E;a
(X)
OJ
2
;::0:
VI
.1> 0 ~.
!!!
..0.
!l
<)"
~o· C
tD
:l
ijl
VI
::r:
-e.
n n N
11 0
:l
...,
C
0
:l
i
11
1/1
..
.~
\
... , ,
' ...... _-_._- ..
\
\
.:
\
,,-
U1
:- - , '
", ~ ,,
---~
0.
,,
O,n.
Q.
"0 0 '.
-. :J
1b'Q.
1/1 C
n
3:-
..... 0
.
Vln
::l QI
w ......
rn",
-
ID
)( VI
10 f: ID
- 647-
INVESTIGATION OF THE MICROSEISMIC NOISE OF THE GEOTHERMAL ANOMALY AT
TORRE ALFINA (ITALY).
M. STEINWACHS
Abstract
Between Sept.10 - Oct.5, 1979 the seismic noise was measured at 15 points
in the geothermal anomaly Torre Alfina using 6 mobile seismological tele-
metric stations. In the centre of the anomaly an increase of the spectral
density of the noise was even more clearly observed than in the case of
earlier measurements made at the anomaly at Urach. This increase is parti-
cularly marked in the frequencies between 30 - 40 Hz.
1. The main points of previous investigations that took place within the
framework of the contract 177/77 EGO.
The measurements were taken with equipment of greater bandwidth than the
earlier measurements, and for this reason, apart from the increase within
the frequency range under 10 Hz, a marked effect is observable at around
30 - 40 Hz. In the example given at the bottom left-hand side of fig.l,
the spectral density has a maximum increase of a factor of 10 relative to
the reference point. In the case of other combinations of stations, the
-~-
values deviate only Sllghtly trom tnose already glven. ror example, in
the upper part of fig. 1 parts of recordings at the stations 1, 2 and 3 are
shown. Even though these results are more impressive than expected
and certainly sufficient to back up the suitablility of the microseismic
method as a geothermal exploration method, one should not remain satis-
fied simply with the purely empirical results. The extensive know-
ledge about the parameters of the Torre Alfina anomaly offer comparatively
good conditions to enable one to drawing up a model which should be able
to explain the generation of the geothennal noise.
Further from the centre of the anomaly, the cap rock becomes thicker, so
that the underlying aquifer is in the shape of a dome. Fig.2 shows a ver-
tical section based on a model from BARELLI (1978) and BERRY (1978). A
number of the drillings pass right through the cap rock into both gas and
water-filled regions of the aquifer. In the case of the well Alfina 1
when the gas-filled part of the aquifer was reached a spontaneous escape
of gas took place. The effective gas pressure in this escape was estimated
as 35 kg/cm 2. The gas consisted of 98.6 % CO 2 (3).
The cap rock, which consists of shales with ophiolites and shaly calca-
reous flysch, seals the reservoir off, but also possesses natural leaks,
e.g. in the presence of faults (2), which enable a part of the gas to
escape to the surface. Chemical analyses of gas samples taken from a na-
tural surface escape of gas near the village of Torre Alfina yielded
the same chemical make-up as that of the gas samples that were taken from
the well Alfina 1 (3). These investigations, paricularly the investi-
gations made on the drillings, indicate that migration of gas from the
reservoir to the earth's surface takes place and that the gas-filled part
of the reservoir is continually regenerated from C02 dissolved in the wa-
ter.
- 650-
1
1121121
1121
1
1
1121 2121 3121 4121 5121 Hz
cap rock
Acknowledgement
The author is grateful to the Italian power corporation ENEL for assistance
in the selection of a suitable geothermal testing area.
References
(1) BARELLI,A., CELATI,R. and MANETTI,G.: Gas-Water interface rise during
early exploitation tests in Alfina geothermal field (northern
Latium, Italy). - Geothermics, E., 199-208, 1978
(2) BERRY,P., CATALDI,C.: Geothermal well stimulation at Torre Alfina.
ENEL, Relatione di Studio e Ricerca, 355, 1978
(3) CATALDI,R. and RENDINA,M.: Recent discovery of a new geothermal field in
Italy: Alfina. - Geothermics, ~, 106-116, 1973
(4) DOUZE,E.J. and SORRELS,G.G.: Geothermal Ground-Noise Survey. - Geophys.
37, 5, 813-825, 1972
(5) HAENEL,R. und ZOTH,G.: Interpretation von Temperaturme6ergebnissen aus
der Forschungsbohrung Urach. - Bericht, Arch.BGR/NLfB
Nr. 84 638, 1979
- 652-
(6) STEINWACHS.M.: Microseismic investigation of the Urach geothermal area.-
Seminar on Geothermal Energy. EUR 5920. 1. 339-350, 1977
(7) STEINWACHS.M.: Mikroseismische Untersuchungen geothermischer Anomalien.-
AbschluBbericht Contr.177-77 EGO. Archiv BGR/NLfB Nr.80 041.
1978
(8) STEINWACHS.M.: Microseismic investigations of geothermal anomalies. -
Bericht Contr. 630-78-4 EGO. Archiv BGR/NLfB Nr. 84 719.
1979
-653-
HOW CAN SEISMICS, ESPECIALLY ACTIVE, ASSIST IN
GEOTHERMAL ENERGY UTILIZATION
TH. KREY
Consultant Geophysicist
c/o PRAKLA-SEISMOS GMBH
Postfach 4767, Haarstr. 5
D-30oo Hannover 1
SUMMARY
-654 -
Within the frame of the Seminar on Geothermal Energy, Strasbourg 1980, a series of
interesting papers have been read on the application of seismics to geothermal
problems. A review and some supplements may therefore be in order.
There are two main aspects concerning the application of seismics to the uti Iization of
geothermal energy. The first one ~-efer. to the question: "Where can geothermal
energy be economically exploited ?" The second one refers to planning and con-
trolling the exploitation. Seismic activity attributing to the solution of the first
problem is generally called seismic exploration, whilst seismic assistence in solving
the problems of exploitation has not yet received such a definite designation. Let us
tentatively call it seismic exploitation help.
Now, much to our regret, iso-temperature surfaces are not reflecting beds forseismic
waves in general. An exception exists if such a surface coincides with a phase
transition, e.g. from solid to fluid. But, there is always a certain influence of
temperature on the velocity of seismic waves given a certain constant pressure, which
is approximately equivalent with constant depth. Thus, hints to temperature anomalies
at a certain depth range can be obtained if velocity anomalies can be inferred from
the traveltimes of seismic events.
One powerful tool for the recognition of velocities in the crust is refraction seismics,
either active or passive. This includes the evaluation of first and second arrivals of
the records of more or less conventional active seismic reflection surveys, tailored to
the investigation of the whole crust. This was shown in several preceding papers of
- 655-
this Seminar. Such refraction events, including those from diving waves, provide the
velocities in the horizontal directions very well. But it becomes difficult to recognize
intercolated lenses with reduced velocities and small thicknesses, which might origi-
nate from partial melting. It is therefore interesting to know that velocities in about
vertical direction can be determined well by reflection seismics when applying the
common-depth-point method invented by Harry flAayne. ( 1 ) •
This topic too has been reported on in one of the preceding papers. With such
reflection seismic surveys we should be aware of the following. Very often in the
crystalline upper crust good reflections are rare. So reflections or diffractions from
the lower crust and the Moho have to be used for the desired velocity determination.
With such deep reflections it is mandatory to use rather long shot-geophone distance
ranges in order to obtain a reasonable precision in the desired velocity information.
The multiplicity of COP-coverage should be high enough in order to avoid alias
problems in the determination of optimal stacking velocities. Especially in the case of
Urach, 24-fold coverage, instead of 8-fold, might have resulted in considerable
improvement.
The conversion of observed refraction and diving wave velocities, and of optimal
stacking velocities, into local interval velocities for the mainly crystalline crust
presupposes the knowledge of the sedimentary layering, if present. For this reason
special seismic surveys may become necessary in order to remove the influence of the
sediments on the traveltimes. This operation is sometimes called "seismic stripping ".
(2 ).
For the recognition of more detail on velocities, impedance logs derived from high
quality seismic sections may be helpful. (4). This refers especially to the question
whether the velocity increases or decreases at a reflecting bed or whether laminated
layering exists. This point too was mentioned in one of the preceding papers.
Reflection and refraction events from the deeper crust may point to bodies of reduced
velocity by other indications too, as was reported in this Seminar too. Whenever a
field of traveltime anomalies is observed with different shot and geophone positions,
bodies with anomalous velocities can be determined, as shown at various meetings,
e.g. by G. Neumann. (5 ). Such observations are, of course, not only possible in
active but also in passive seismics, i.e. by using the records of earthquakes, industrial
or military explosions. The difficulty is, at least with earthquakes, to have the
necessary dense grid of observation stations available in the field at the right time.
But perhaps something could be organized with respect to aftershocks.
- 657-
Quite similar, are probably the corresponding curves for shearwaves as long as
Poisson's ratio does not essentially change. But the rock becomes a porous material
filled with a fluid when the melting point of one of the component minerals is reached.
Therefore, at higher temperatures the curves for shearwaves should rise much steeper
than for P-waves. It should therefore always be considered to include shearwaves too
into seismic geothermics exploration. However, strong shearwave sources would have
to be developed, on account of the higher absorption, if reflections from the lower
crust are aimed at. To my feeling this development is still at its beginning, perhaps
with the exception of the VIBROSEIS+)system. (7), (8). The availability of a
sufficient number of horizontal geophones should also be envisaged well in advance.
Perhaps we find better chances in passive seismics as far as shearwaves are concerned.
With earthquakes the shear energy is known to be rather high. The same may be the
case with quarry-blasts.
Up to now our attention has been pointed towards the evaluation of traveltimes
resulting in the computation of velocities and the detection of velocity anomalies.
But regions of increased temperature in the subsurface do not only affect the velocities
of seismic waves but also their absorption. Therefore, comparisons of amplitudes should
be included in the seismic geothermic exploration. As. this exploration is at present
executed mainly onshore there still exist some difficulties, i.e. the various shotpoints
and geophone stations mostly exhibit different energy and frequency characteristics,
which have to be compensated for. Though corresponding data processes exist or are
being developed, relative comparisons should be aimed at, i.e. the ratio of the
amplitudes of two different reflecting horizons should be determined as a function of
the horizontal coordinates x and y, and then be evaluated. The ratio between a
reflection and a refraction horizon may be considered too, however, in this case the
shot-geophone distances should remain fairly constant.
-658 -
need not necessarily point to a change of absorption between the two horizons involved,
as they may also originate from a change in the laminar structure of the involved
horizons.
Now, seismic geothennic exploration is not confined just to the discovery of regions
of increased temperature in the subsurface. Many costs and risks would probably be
avoided if we could renounce on producing artificial surfaces for energy exchange by
"frac-ing". Instead, natural open fault planes or sedimentary layers of sufficient
porosity may serve for the exchange of geothermal energy. Such possibilities may often
exist if the aim is not necessarily the construction of an electrical power plant, but
the utilization of hot water for heating buildings lying in the vicinity of the geothermal
energy source. Here too applied seismics may be extremely useful.
The determination of the structure, or, say, the geometry of porous sedimentary layers
is a problem which applied seismics has had to solve for more than half a century in
order to find oil and gas. No special discussion is necessary. Moreover in many cases
reflection seismics can find out the location where the hot water filling of the pores
changes to water vapor filling, because layers with a gaseous pore filling have a
lower P-wave velocity than those with a liquid pore filling, and water vapor is, of
course, a gas.
Thus, prominent changes of the reflection coefficient may occur which involve the
"bright and flat spots" in exploration seismics for gas. ( 10). Fig. 2 shows such a
bright spot which after drilling proved to originate from a gas-field. Fig. 3 represents
another example showing the frequency changes involved too. Also, when storing a
-659 -
gas in a porous layer we can often control the lateral extension of the gas fill by the
corresponding bright spots. ( 11 ). See Fig. 4. This possibility of applied seismics,
i.e. to take advantage of the occurrence of "bright spots", may become of special
importance in regions where active or declining volcanism is encountered (fumaroles,
geysers ).
The other task, i.e. to discover fault planes and determine their positions is not new
either for applied seismics. However, reflections originating from the genuine fault
planes are observed very seldom when exploring for hydrocarbons. Instead, the
existence and the position of a fault is mainly inferred from the throws of reflecting
beds, from observable diffractions, and from reflected refraction. The detection of
such indications can be considerably improved nowadays by corresponding field geo-
metries of sources and receivers and by special data processing. Fig. 5 shows a
Vibroseis section running at a right angle to the eastern border of the Upper Rhine
graben. Two arrows point to a fault detected by this line, between the arrow at the
surface and that at the largest depth where reflecting horizons are still observed, we
see numerous indications defining the intersection of the fault plane with the seismic
section. This example is part of a seismic survey carried out for Schachtbau, Lingen,
by PRAKLA-SEISMOS, Hannover, and data-processed by DIGICON, London. This
survey contributes to a geothermal heating project for the town of Buhl or environment.
This project is sponsored by the Bundesministerium fUr Forschung und Technologie, Bonn.
I thank Schachtbau for the permission to show this section. In the meantime, a borehole
confirmed the fault in good agreement with the seismic survey.
When discussing the possible help which seismics can attribute to the solution of geo-
thermal problems we must not omit 3-D-seismics, i.e. the exploration in three dimen-
sions right from the beginning. 3-D-seismics is increasingly being applied, especially
when planning the most economical exploitation of known oil, gas- or coalfields. (12).
With this rather new technique the seismic field work has to be carried out on an areal,
i.e. 2-D base. Therefore certain properties of the seismogram, as amplitude or
frequency of a certain horizon, can be represented automatically in a 2-D-manner,
i.e. by mapping in the (x-,y- ) plane. Fig. 6 shows a probe of such a mapobtained
on account of an areal seismic field survey. It represents the structure at a certain
constant reflection time, i.e. approximately a horizontal section.
-660 -
3-D-seismies could also be useful when a system of complicated faults has to be dis-
entangled for some geothermal problem. In this respect a lot of useful information
could be provided for coal mining, see Figures 7 and 8 • Finally, we should not even
exclude the possibility to detect reflections from the deep crust and the Moho by areal
active seismic surveys, yielding velocity information and thus indications of increased
temperature, immediately in a 3-D-manner.
How seismies may assist in the exploitation of geothermal energy has not been discussed
explicitly. However, some hints are already inherent in the preceding paragraphs, for
instance, the application of the "bright spot" philosophy to the question of whether
pores or cavities are filled with hot water or steam. The important chapter of control-
ling the extension and geometry of fraes will be treated in following papers. So I do
not want to anticipate this topic now. I should only like to stress again the application
of shearwaves in this context.
When considering the rather high costs for seismic surveys, we should be aware that
a unit of energy contained in producible hydrocarbons is burdened with considerable
expenditures for applied geophysics, mainly seismies. Why should we hesitate spending
similar amounts of money for the same unit of energy when it originates from geothermies,
especially when the exploitation of energy other than that of hydrocarbons becomes
mandatory ?
- 661-
0/0
6,~
·u
0
Qi
5 r- >
'0
(\)
Ol
4 §
f- .c.
u
3 f-~Cbo
c:
I
a-
:r ~ 0
100 K/kb
a-
N
I
0
~<:::
2 3 4 5 6kb
,,,"....
" '
~..
..Itt ,.'"
t
Fig. 2 Seismic record section with "bright spot" in the upper right part
- 663-
Fig. 3 "Bright spot" with separate representation of amplitudes and dominating
frequencies .
- 664-
o 0061 0085 OIOg 01 33
-0 5
•
(<,.
'"
;It
A
.n 5
1. o
Fig." Seismic section across a gas storage area. Gas is present where horizon A
is outstanding.
-665 -
Seismic Line Crossing
the Eastern Main Rheingraben Fault
Fig. 5
- 666-
Fig. 6 Constant-time section with two different kinds of representation
- 667-
0.05
0.55
0.55
I
0.55
00
'"'"
0.55
(5) Neumann, G., 1979; Ergebnisse eines Verfahrens zur Inversion von Laufzeit-
residuen am Beispiel von Modellvermessungen und Transmission und Reflexion;
Paper presented at the Meeting of the "39. Jahrestagung der Deutschen Geo-
physikalischen Gesellschaft, Kiel 1979"
(6 ) Kern, H., 1978; The effect of high temperature and high confining pressure
on compressional wave velocities in quartz-bearing and quartz-free igneous
and metamorphic rocks; Tectonophysics, Vol. 44: P. 185-203
(10) Backus, M.M. et aI., 1975; Flat spot exploration (Gas Mapping ),
Geophys.Prosp., Vol. 23, p.0533
(11) Schmoll, J. et aI., 1973; Reflection seismic control of the edge water lines
of gas reservoirs; Paper presented at the 35th EAEG tv\eeting, Brighton
- 669-
( 12) Krey, Th., 1978 i Reconciling the demands of 3-D-Seismics with those of
improved resolution ( A research program in the Ruhr Coal-Mining area) i
Paper presented at the 48th SEG-t-Aeeting in San Francisco
( 14) Hubral, P. and Krey, Th., 1980 i Interval velocities from seismic reflection-
time measurements; SEG-Monographs ( in print)
- 670-
SESSION VI - Exploitation of Geothermal Resources
-671-
ACTUAL SITUATION AND FUTURE PROGRAM OF GEOTHERMAL
DEVELOPMENTS IN MEXICO
H. ALONSO ESPINOSA
Summary
Since the year 1955 the first works towards the develop-
ment of geothermal energy to generate electricity were initia-
ted in the Republic of Mexico. As a result of this, in 1959 a
pilot plant generating 600 KW went into operation in Pathe,
in the center of the country. Subsequently, in 1973 the first
commercial scale geothermal plant, with an installed capacity
of 75,000 KW, was terminated at Cerro Prieto, B. C.
Based on the results of this plant, and on the exploration
studies that were performed in such a zone, it was determined
to double its capacity, and on April 1979 another 75,000 KW
started operation, reaching the present capacity of 150,000 KW.
The Federal Power Commission of Mexico continues its in-
ventory on thermal sources of the country, thus in Los Azufres,
Michoacan, 10 exploratory wells have been drilled, with very
good results obtained. At the same time, d~iled studies have
been carried out in various zones and states of the republic,
in an attempt to achive a 40,000 KWh production annually by
the year 2000.
We are aware that geothermal energy alone will not solve
the nation's requirements of electric energy, though its uti-
lization is undoubtely convenient and significant to aid along
with other sources of energy, to diminish the present dependen-
cy on hydrocarbons.
-673-
After 15 years of studies and activities in different
regions of Mexico and after 10 years in the Cerro Prieto area,
commercial electric power generation was initiated on April
1973 in the Cerro Prieto Geothermal Power Plant, with an ins-
talled capacity of 75,000 Kw. On April 24, 1979 the construc
tion of a second 75,000 Kw power plant was completed and
started commercial operation. This fact is specially signi-
ficant to Mexico who requires to diminish, as much as possible
the dependency on hydrocarbons for electricity generation.
-674-
However, this is only the beginning of what must be
done in our country in the near future. Mexico with its geo-
logical characteristics associated with areas of recent vol-
canism, cortical weaknesses and thermal anomalies which are
widely distributed within its territory, offers important po-
ssibilities, presently being evaluated, which have allowed us
to reach the feasibility of generating 40,000 million geothe~
mal Kwh per year. This is a goal for the year 2000 that, once
it is reached, will represent savings of nearly 57 million oil
barrels per year, required by fossil fuel plants with equal ca
pacity and generation.
- 675-
the areas of La Primavera and Lago Cuitzeo drilling has also
been programmed.
-676-
The San Andreas tectonic system, with a NW-SE trend and
from which several faults derive, affects the Cerro Prieto
area. One of these, the San Jacinto, locally known as the Ce
rro Prieto fault, is the main vertical conduit for geothermal
flow to reach near surface zones where they remain stored in
horizontal permeable rock formations.
-677-
Cerro Prieto is a water dominated field in which the
steam-water flow is separated at the surface, using the steam
for electricity generation. The separated water that at
present is not used, contains a variety of chemical products
such as potassium chloride, that in a concentration of 0.37%
might be recovered through solar evaporation and separated
from sodium chloride by flotation process.
-678-
Various North-American companies have requested to the
Mexican Government the purchase of electric energy produced in
the Cerro Prieto Geothermal Field. Such request establishes
that beginning 1983 electric energy would be sold, assigning
by 1986 a capacity of 600,000 Kw during a 10 year period.
-679-
HEAT AND MASS TRANSFER WITH PHASE CONVERSION IN POROUS MEDIA
Summary
Heat and mass tranfer in the presence of phase change in porous media
playa very important role in the production of hydrothermal reservoirs.
After a literature review of this field the equations describing heat and
mass transfer are presented and used to derive dimensionless quantities
governing the phenomena. Recent numerical modelling of these equations are
also analysed and discussed.
A physical model designed to simulate linear unsteady two phase flow
during a self vaporization process without or with reinjection of conden-
sed fluid is presented. In the case of no reinjection the experimental re-
sults have confirmed the validity of lumped-parameter model to describe the
global behaviour of an homogeneous geothermal reservoir during exploitation.
For the case with reinjection only qualitative results showing the possible
interest of this procedure for stimulated vapor dominated reservoirs are
presented.
Concerning the numerical approach a distributed parameter model or lo-
cal model has been developed to simulate the unsteady two phase flow in a
unidirectionnal case. Results showing the evolution of the phase change zo-
ne are presented.
- 680-
1. INTRODUCTION
Numerous studies have been made during the last ten years to unders-
tand the existence of high temperature geothermal fields and natural con-
vection is now recognised as one of the main factors. The effect of convec-
tion is to induce an important upwards heat flux through the aquifer, hen-
ce very high temperature zones in the upper part of the reservoir.
Our research program has been mainly focused on the analyse of these
phase change phenomena in porous media with a particular mention for the
case of the reinjection.
To help provide answers to this question the following point have been
examined during the two last years (initial phase of the study)
- 681 -
2. SURVEY OF THE LITERATURE
Only a few works have been devoted to fundamental studies of heat and
mass transfer in porous media in high energy geothermal fields, i.e. with
an accompanying phase change. The publications examined extensively studied
and presented in (1) can be devided into two categories
• Local analysis: In these largely theoretical studies (2), (3), (4) the
development of vapor phase is analysed in a porous media saturated initial-
ly with liquid. The porous media is considered as a continuum in which the
phenomena are described by local equations. Because of the assumption made
i.e. : latent heat of vaporization neglected (2), (4), permeability of
both phases egal and two-phase zone neglected (the transition between the
liquid and the vapor is considered as very sharp) (3), the results are
not representative of phenomena which appear in geothermal fields during
production.
-682 -
In the publication reviewed there are many common traits. The mathema-
tical models used are basically similar and are based on mass and energy
local balances or distributed parameters. Darcy's equation is assumed to
be a simplified momentum balance for multiphase flow and the viscous dissi-
pation and compressible work are generally neglected in the equation of
energy.
Some authors (11) to (15) have adopted an areal formulation but gravi-
tational forces are not taken into account (two horizontal dimensions).
This simplification obscured a physical aspect very interesting : the pos-
sible formation of a gas-cap at the top of the reservoir. This phenomena
seems to be considered in (16), (17), (18), (19) where the model is three-
dimensional.
3. MATHEMATICAL FORMULATION
When fluids moves through a porous medium two alternative methods have
been used to describe heat transfer. In the first Which is used most fre-
quently in petroleum engineering and hydrology the porous media is conside-
red as a continuum with the different phases at the same temperature. For
this case the concept of thermal conductivity coefficient may be
- 684-
applied. In the second method a distinction is made between the temperatu-
res of the fluid and solid phase. The heat flux from one phase to the
other is expressed by means of a heat transfer coefficient (22).
+
mass e: ~(p. S.) = - v. (p. V.) + I. with L I. = 0
at 1 1 1 1 1 i 1
iC.1
p.
+
v.1 VP
momentum = - Pi
1 J.l.
1
energy
a
at
[( 1-e:)Pr hr + e: L p.1 S.1 h.]
i
1
= -V [AX VT + L. p. h.
1
1 1
VJ +dPdt
state: p. (P,T) ; J.l. (P,T) ; K . (S.) ; AX(S., T)
1 1 rl 1 1
4. EXPERIMENTAL APPARATUS
The apparatus was designed to check the influence of different physi-
cal conditions on the behaviour of a piece of porous medium of simple sha-
pe in which the heat and mass transfer during production are unidirectio-
nal. This experimental model is presented on Figure 2. The porous media is
surrounded by lagging of circular cross section having a diameter of 5 cm
and a length of 2 meters. A metal tube on the outside helps to maintain
the pressure and to control the temperature (the model is heated by elec-
tronically regulated resistors). Pure liquid water is injected in the po-
rous medium and pressured by means of a pressure tank. After thermodyna-
mic equilibrium has been reached,depletion is started by opening a micro-
meter valve at the outlet condenser. Reinjection can be made during the
production by connecting the water tank and the porous media through a
flo\lllleter.
-685 -
During the experiments the foJ.lowing parameter are recorded: T(X,t)
p(X,t) ; Sl(X,t) liquid saturation by gamma ray absorption; Te(X,t) and
Td(X,t) the temperature of the lagging and the metal tube respectively
M(t) mass of produced fluid and h(t) enthalpy of produced fluid.
-686 -
The subscriptsc,i,a,pv and p indicate the instantaneous, initial, entering
lost and produced values of different quantities and, 0, a reference le-
vel. V is the volume of the reservoir; e, the specific internal energy, Q
the heat received by conduction. The others variables have been defined.
a
at
K
rl
-v· [ -(VP
VI
+
- P g) +
I
:r
g
g (VP - Pg~) J+ ~ = 0
Krl-~
h + Kr h + ]
- V. [ --(Vp - P g) - ~(Vp - P g) + q h + q h = 0
V I V g 11. g g
1 g
3
o
6. RESULTS
6.1. Test of global balances model
The global balances model presented on 5.1. has been tested by compa-
ring, in cases of self-vaporization without reinjection, experimental re-
sults with theoretical results deduced from numerical computation of the
equation 1. This comparison which concerns the mass of fluid produced in
function of time for different experimental conditions (porosity, speed
of decompression, initial liquid saturation) is presented on Figure 3. We
can observe that this model describes very well the experimental simula-
tions. For runs performed this agrement justified the validity of the
asumption concerning the thermal equilibrium between the solid and the
fluid phases.
-688 -
6.2. Numerical simulation of the phase change by local equations
In the case where reinjection is made at very low value of the satura-
tion we observe very interesting phenomena : increasing of pressure until
-689 -
critical poin GLOBAL BALAIICFS EQllATION~
Local equations
• Kruger
~lercer - Fa st
OF
EF
1- dim
2-dim
P , 51
P, n
vapor
Toronyi DF 2-dim I , Sl
temperature Garg-Prit hett OF 3- dim LJ , p
Coats DF 3-dim r , T , 51
critical
Thomas DF 3- lm P , T, SI
point
Pin er EF 1- im P ,h
La;;scter OF'! 3- im U p
ent.halpy
FIGURE 1
thernost t.
FIGURE 2
M/M o
exp o result
num . res It
30 60 90
FIGURE 3
-690 -
- - - - __ -1.6_
27 sec .
ap = 0
ax -- --- K 8 . 10- 12 m2
q=75 jr.n
0,5
X/I
0 0,5
FIGURE 4
.'1
gr ,i
gr 8
1
=0
81
reinje ted
?10 3 fluid
recoverea fluid 2 . 10 3
10 3 no reinjection
10 3
t t
mn mn
50 100 50 100
FIGURE 5 FIGURE 6
y y
no reinjection
, =1
.... ,\
' S1
1
a
\\
\ \ rein.j ection
0,5
\
",-:- -- --
\
-691 -
the saturated vapor pressure curve and of production of fluid at high en-
thalpy level.
7. CONCLUSION
The results presented in this paper constitute the first step of an ex-
perimental and numerical study about the phase change phenomena during the
exploitation of high energy hydrothermal reservoirs.
The work developed takes place among the problems put in evidence in
the literature survey i.e. the necessity of comparison between experimen-
tal and numerical results. In this aim an experimental apparatus has been
designed and first numerical and experimental results collected.
BIBLIOGRAPHY
- 693-
A THERMOGRAVIMETRIC LOOP FOR CONVERTING LOli ENTHALPY GEOTHERMAL
ENERGY INTO ELECTRICITY
G. TREBBI
Summa!,Y
-694-
1. INTRODUCTION
Under contract with the Commission of the European Economic Community,
ENEL has realized a pilot plant for use of geothermal (101-1 enthalpy) heat
2. OPERANTING PRINCIPLE
density is obtained by injection into the base of one of the two branches,
hydraulic turbine and enter the circuit again. The working fluid undergoes
cyCle.
Referring to the key numbers in Fig. 1 and 2 and vlorking fluid, from
its injection at the base of the two phase flow duct, expands and form
- 695-
established at the top of the two-phase flow duct. Because of the 10vl ratio
between the flow of the vlOrkin~ fluid and the cC),rrier fluid, such expansion
These two elements define the amount of heat (''1-8 supplied lJY the
source (eeothermal of solar) to the carrier fluid throW!:ll the heat cxd1an-
Hhen leaving the tHo-phase flol'l duct, the vlOrki~~ fluid is subject to
fluid reaches the condenser too; that's the third contrilmtion to the con-
denser capacity QC.
The studies on these systems have pointed out some typic2,1 chCLr8,cte-
ristics of the plant I'Thich define its dir.lensions aY'.cl its performo.nces. 'l'he
main characteristics are: T1' '1'2' G, Z/:;r,UI, \) and. ITorhnG i'luirl (see 'l'e.b.I
for nomenclature).
- 696-
The power of the system can be evaluated in many w~s which emphasize
various basic quantities.
Referring to the thermodynamic cycle:
Under the same conditions, if the plant height increases, its effi-
ciency decreases because of the increased dissipations. '1'0 limit height
and pressure drops, ZLIM and consequently the cycle temperature T1 will be
controlled.
-697 -
available power to the turbine and upper and lo\'rer temperatures of the cy~
cleo
1-1hen even the height has been fixed, ther's still a degree of freedom
represented by the mass velocity.
The mass velocity of the two phase flox G has an incluence on diameter
DB only for the lowest values vrhile the behaviour of the efficiency curve
is quite flat through a very broad range. As a result of previous resear-
ches the most profitable G values fall betJ-leen 1500 + 3500 kg/m2s according
to the pol';er of the plant. A qualitative analysis about the behaviour of
the depending quantities of the plant at the var,yinti of the independent
parameters G, 'iI, T1' T2, Z and l.orking fluid define a profitable working
ranee to the various po1rlGrs.
The turbine is Kaplan type horizontal axis coupled lVith a d.c. genera.-
tor.
-698 -
The working of the three above mentioned exchange - units involves
two more loop: the loop of the geothermal water (for the boiler and for
the exchanger) and the loop of cooling water (for the condenser).
Fig. 3 shows the gravimetric t01rler "lith the separator on the top.
During the project the safety problems have had great consideration
e~en though freon can be classifed as least toxic.
-699 -
nearly all workers may be repeatedly exposed, day after day, vlithout adver-
se effects. TLV's refer to time - weighted concentrations for a 7 - or 8
hour '"lork day and 40 hour - vJOrk - vleek corrisponding to about 7 glm 3 in
normal conditions. Concentration until 20~,i by volume do not appear to pro-
duce injury even for duration of exposure of 2 hours.
5. OFF-DESIGN PERFORMANCES
Keeping the total head on the turbine constant and considering that
the minimum pressure of the cycle (upper branch of the two-phase duct) is
defined by the temperature of the cold Hater, the freon flow rate GG is
the only indipendent variable. Is GG changes, the same happens to GL and
Pp
The zero power gas flow rate is arond 105~ of the nominal value.
-700 -
In the next table are summarized same results:
l-1ells.
Consictering that the installation cost of a 320 I,;'.! oil f'ired pOl'Ter
units is about 0.4 !:TL/k~! and that the cost of the fuel (r-apitalized ;:),nr3
refered to the installed pOHer) is of abour 1.6 j,:L/k:l for a total cost or
doubtful in industrial area,s, eJthouch not to the extent thd o:ny "
P08S1iJl-
'
i) cycle reGeneration
It is possible to heat the freon before sendi.n::; it to thn boiler,
usine; vapour freon (in sup8rheated com1 ition) at the ec:it of the se-
paratore
A study has indicatcd that it is convenient to usc a reGener::ttor \d,th
efficiency of appro;:imately 0.9; the totnl ejwhani~e Guric',r-cs reduce 1)]
about 9,: and (referrinc to plante of ') ]1.:) tbe cod of. tile ener:~Jr
decreases 11':/ 3',~j
ii) rlirect intiect ion of freon into the Gravimetric loop
- 701-
The total cost of the exchangers decreases by about 20-25% referring
to plants of 5 ~rw. The interest of the direct injection becomes as
smaller as the size of the plant decreases.
It will be necessary to investigate the evaporation of freon from the
mixture freon-water, and study a suitable type of injector;
iii) direct exchange with the cold sink
The costs of the devices for the separation by gravity of freon and
water after mixing are very high, more than the cost of the condenser;
iv) direct expansion in a turbine
An economical study on the basis of very simplified hypothesis pointed
out the interest of this type of plant.
A more detailed analysis of the turbine costs and availability is ne-
cessary and general conclusion can't be drawn at the moment.
-702-
TABLE 1
Nomenclature
Symbol Meaning
Greek
6 variation of a quantity
Yl, or~anic efficiency
-704 -
Fig . 3
-70 5 -
LIFETIME OPTIMIZATION OF LOW ENTHALPY GEOTHERMAL DOUBLETS
Summary
Presently FLOPETROL
-706 -
1. INTRODUCTION
The use of geothermal heat, traditionnally limited to countries with
exceptionnally high temperature gradients is presently extended to regions
with normal gradients , where calories are directly used for local space
heating. In most cases the heat depleted water is reinjected back into
its original reservoir, in order to protect the supply of water, and get
rid of large quantities of water generally with important salinity and
residual heat.
The most usual device is that of the geothermal doublet : it consists
in a production well coupled with a second well through which all the pro-
duced water is recirculated into the original aquifer after extraction of
its thermal energy. This reinjection locally induces a gradually expanding
bulk of cold water around the recharging well ; after some time it will
reach the production well whose temperature will henceforward progressive-
ly decrease (breakthrough time).
This type of exploitation being the object of an increasing number
of applications, it is of great interest to be able to predict the break-
through time and the subsequent temperature decrease of produced water.
That is why the European Community Commission has asked B.R.G.M. a
general study of the influences of the physical characteristics of a geo-
thermal reservoir on the temperature of the water produced by a doublet
(Project G/C7, contract # 093-76 EGF) Final report; B.R.G.M. # 78 SGN
405 GTH - (Ret. 3).
On the other hand, the Delegation a la Recherche Scientifique et
Technique (D.G.R.S.T.) has sponsored a field experiment of a doublet at
small scale (well spacing of 13 m) performed by B.R.G.M. together with
tracer tests (by C.E.N.G.::) one-well hot water injection tests (by
BURGEAp:::: and C• loG:.:::::) and cyclic one-well hot water storage (by BRGM)
-707 -
The interpretation of these data (DGRST - contract # 77-07-1117) has
allowed BRGM (Ref. 2) to verify the results of the theoretical study
for ECC and the validity of the physical hypotheses on which it is based.
The object of the present communication is to give the main results
of theses two studies •
2. NUMERICAL SIMULATIONS
Mathematical models based on finite differences or integrated finite
differences have been used to study the influences of each phenomer.onon the
breakthrough curve in comparison to piston flow curve taken as reference.
The aquifer is assumed horizontal and of uniform thickness, and the
confining layers of Uniform thermal properties . The symmetry with respect
to the aquifer horizontal mid-plane allows the limitation of the model to
the upper half of the system.
The mesh of the main model (rig. 1) comprises five layers of nodes
inside the confining rocks, plus one layer for the aquifer, using small
curvilinear cells in the well vicinity and large parallelepipeds in the
distance.
The accuracy of the simulations has been controlled against analyti-
cal solutions in the most drastic case for numerical dispersion : progaga-
tion of an abrupt thermal front within the reservoir. The thickness of
each layer of nodes representing the confining rocks, has been optimized
in order to get a fine approximation in spite of the small number of nodes.
Details on the conception of the models and the control operations
are giV8n in (Ref. 3).
In order to ensure a general applicability of the study, the results
have been expressed in terms of dimensionless variables and parameters,
such as :
- dimensionless temperature :e R = (e- eo ) / (e.1 - e0 ) (1)
-708 -
3. INFLUENCE OF PHYSICAL PARAMETERS ON THE BREAKTHROUGH CURVE
Table I
Conduction Pe ~ ~ 10 1
Parameters A ~
10 10 10
-709 -
It should be noted that conductivity Aa appearing in the definition
of Peclet number Pe is an apparent conductivity that should account both
for conductive and dispersive transfer, (Ref. 5) :
(5)
Table II
-710 -
The injected cold water having a higher viscosity than the natural
reservoir water, its progression towards the production well is slowed
down • Type curves have been plotted for various values of the ratio of
kinematic viscosities: N = vo/Vi (fig.5) N values inferior to unity
(vi> vo) correspond to a geothermal doublet with cold water injection.
-711 -
This site had been previously instrumented for extensive tracer tests
experiments in the period 1973 to 1975 (Ref. 4) with 13 observation wells
The heat experiments have consisted in 3 one-well tests on well C,
3 doublet experiments between well C (injection) and P (production) and
a cyclic one-well hot water storage with 4 periods of 12 days each.
Pe = 10.3 A = 24
= t(days) e - 13.3
tR 3 33.5-13.3
5.1. Data
The characteristics of the reservoir are the following
h = 40 m
eo = 70°C
o = 15 %
YF = 10 6 cal. m- 3 • K-1
YE = 0.5 106 cal. m- 3 • K-1
YA = 0.57 1& cal . m- 3 K-1
AA = 0.6 cal m- 1 s-l K- 1
Aa = 0.6 + a 10 6 Ivl
-712 -
v
o
=1 m/year
D = 1 000 m
Q = 120 m3 /h
e.1 = 35°C
5.3. ~~~!_~~~~~~~~~_~~!~_~~E:~~~_e~~~~~~
This low value indicates high exchanges with the confining rocks
partly restauring the temperature of the aquifer.
5.4. ~~~!_~~~~~~~~~_~Y_~~!!~~~~~!l_~~!~~~_!~~_~~~~!~~
A
a
= 0.6 + 33.10 6
-713 -
- c - for a high (but not improbable) dispersivity value of 100m,
A = 80 cal m- l s-l K-l
a
and ,
Pe = 3.
The important differences resulting on the breakthrough curves can
be visualized on fig. 3 for Pe = 3, 10, ~
-714 -
6. DISCUSSION: OPTIMIZATION OF LOW ENTHALPY DOUBLETS
-715 -
T T.
T. -----------
Ti
.,...... 0'
oI (0'
10'
• ~D
~ - Breakthrough curve of a doublet
.,
VI
C 0 .'
e
;:;
.,"
0,'
0,'
0.'
.. '
0
0,2 O"lo 0," 0.l1li Q.. 0;1 Qlllqt I .. l1li • 7 . tlO
Dimensionless time tR
1.0
0. 9
!
';
0. 8
...
~'!
:! 0.7
;;
: 0 .6
!
0.5
O.~
0.3
0.2
0 .1
0
10" tempa reduit 10'
-717 -
I~ m 13m
G) 8
o P, ,
~ ~
0 P,
, 01
•• ,.,1 ••• '.'
P, C 0 R .
r':;
C e C)
•o ~ •••• , . II •• I
• ~ 0"
;
P. o.
•
~ 0 •0• GO
L
"
...., 0,'
L
GO
o 0 !o.
-<
C
...... ~o,J
~ ~~.....o, GO
..-< e""10 20 JO 40
C....
time (days)
BIBLIOGRAPHY
(5) - J.P. SAUTY, A.C. GRINGARTEN, P.A. LANDEL - The effect of thermal
dispersion on injection of hot water in aquifers - Second Invita-
tional Well Testing Symposium, Berkeley (Californie), pp. 122-131
Oct. 1978.
(6) - A.C. GRIN GARTEN - Reservoir lifetime and heat recovery factor in
geothermal aquifers used for urban heating. - Joint IASPAI/IASVEI
Symposium, Durham (U.K.), August 11-12, 1977.
-719 -
INFLUENCE OF STRATIFIED HETERoGENITIES
OF PERMEABILITY ON THE LIFE SPAN
OF A GEOTHERMAL DOUBLET
P. GOBLET
ABSTRACT
-720 -
1. INTRODUCTION
This study points out the effect on the value of these evaluations
of stratified heterogeneities of permeability, which have not been disco-
vered before exploitation.
pwg (a
P e
- w a1E.
Ss +~) t + pgq
-721-
2.2. Thermal problem
+ +
~ - A • grad e
A = A
o
+; luly
'e
-722 -
Finally, energy conservation equation writes as follows:
as
= Y at
- -+- -+-
div (\ grad S) - Ye div (US)
Certain parameters of the flow problem, i.e. the mass per unit
volume of water. its dynamic viscosity. the heat capacity of the water and
the medium are dependent upon the temperature. Thus. the problem of flow
and that of heat are coupled, which means that they have to be solved
simultaneously and interactively. In order to simplify the modelling. we
have disregarded the variations of the above parameters, as their influence
has already been partly examined by Landel and Sauty (2) •
• The variation of the mass per unit volume causes the phenomenon of
segregation by density. which hastens the arrival of the cold front at
the production well by a maximum of 10% in the case of conventional
working conditions .
• The increase in dynamic viscosity of the water, when the temperature
decreases. slows down the progress of the cold front. which causes a
delay factor of up to 1.45.
4. AoIMENSIoNAL FORMULATION
-123 -
This temperature increases from 0 to 1 during exploitation of a
doublet and the time after which it becomes non nil in pure convection
is the recovery time t B.
t =-.!.
° tB
- Peclet number: it is used to characterize the ratio between the
conductive and the convective heat flows. It is written:
Vth x distance
Pe = --.:.;-'---;::----
Dth
Vth is the "thermal velocity". i.e. the velocity of the thermal front
in pure convection:
A
Dth = Y
and half of the distance between the wells is taken as a characte-
ristic distance.
As U and A are not uniform. we take their values at half distance
between the wells as reference values.
-724-
that the head potential field should not be too different from the theore-
tical one in the areas where the temperature variations are the largest,
especially between the wells. Fig.1 may be used to compare the two fields.
c) ~~!~!~g_~~~_!~~~£_~~~~~~£!~~
The top and bottom boundaries of the mesh are impermeable and
adiabatic. Thus, they may be identified as symmetrical planes, and all
the layers described become equivalent to all systems which may be deducted
from them by plane symmetry in relation to one of these planes. This
operation may be repeated as often as desired. In short, our model must
be seen only as the minimal unit needed to describe a periodical sedimen-
tary sequence. In order to represent physical reality, the complete
sedimentary structure must be surrounded above and below by walls which
are impermeable, but heat-conducting. We have not accounted for the effect
of these walls, studied by Landel and Sauty (2). It must be kept in mind
that these walls supply heat to the aquifer and thus delay the temperature
loss at the production well.
-725-
of the device, either hasten the arrival in the production well of the
cold water, or, on the contrary, completely prevent it (causing ·pollution"
by cold water in the downstream area of the flow). In most cases, this
flow is not significant, and here we assume it to be nil, which makes the
device symmetrical in relation to the line of wells. Thus, we will only
consider half of the device.
- The pumping and injection wells penetrate the aquifer from top to
bottom. The aquifer is confined, i.e. the medium is saturated. As the
mass per unit volume and the kinematic viscosity of the water are indepen-
dent of time, we are dealing with a hypothetical Dupuit flow with a cons-
tant hydraulic potential on a vertical. Furthermore, the pumping and
injection flow-rate are assumed constant.
The method chosen for the solving of the steady-state flow and
the thermal equations is that of finite elements. The Galerkin formula-
tion is used. The domain is divided into eight-node isoparametric elements
and the approximation functions are of the first order for the potentials
as well as for the temperatures. Fig. 2 shows, in the plane, the mesh that
was used. The three-dimensional mesh is made up of five layers of similar
equidistant nodes.
-726-
The experimental calculation of the anisotropy ratio is impossible
on the basis of an experiment with radial symmetry. However, if one
accepts, as the non-radial tests seem to indicate, that this ratio is large
it is probably preferable to arbitrarily assign a large value to it
(e.g. 10) rather than suppose it to be equal to 1.
-727 -
Our experiments have shown that each layer must be broken dOvin
into a least two strata of finite elements.
IIIIII1 Kl !- - -
111I1 1 K3
- - -- - - - =-1<4 3 xK3
Model 1 Model 2
K1 and K3 are chosen in such a manner that the two models are equivalent
to the same homogeneous model.
Fig. 5 shows the two recovery curves as well as the curve of the
equivalent model.
-728 -
The greater the heterogeneity, the earlier the arrival of the
cold front; however, the ends of the curves meet. In the entire zone
which is important in exploitation, however, the cooling down is greatly
accelerated.
When the cooling off of the slow layers does not, in a significant
way, reach their median plane, their effect on the contiguous layers may
be assimilated to that of a confining bed and identified by a coefficient
of the same type as that introduced by Gringarten and Sauty (1) in the
case of a weak horizontal conduction:
-729 -
where A~ and Y~ characterize the slow parameters, Yr the fast layersJ h is
the thickness of the fast layer and 0 the distance between wells. We have
used two sets of thermal parameters, which lead to values of A of the ord~
of 1 and 10 respectively.
In Fig. 8, the same comparison is made for the second case. One
observes a difference of factor close to 3.5 between the heterogeneous and
the homogeneous models. This difference decreases slowly as the recharge
by the slow layers increases: around 2.7for a reduction of 10 %, 2.4 for
a reduction of 50 %. However, the scaled temperatures of the heterogeneous
model sti 11 remain higher, "Ihich indicates a lower real temperature.
In conclusion:
- the model for which A =1 behaves approximately as a homogenous model at
the end of the exploitation, but not at the stage with which we are concer-
ned, i.e. up to around 20 % reduction in temperature. It would probably
be possible to fit it approximatively on a homogeneous model with low
Peclet, but the recovery time would be shorter than what could be expected
from the apparent permeability of the medium. This anomaly could be
-730 -
\
\
\
\
\
\
\ ,
-----_....:',
computed "ield
analytical field
P IEZOI~ETRIC HEADS
BIBLIOGRAPHY
tr 10
heterogeneous model
homoqeneous nodel
.5
tr
z z
.5
-733 -
detected if measurements taken in the field are to be interpreted at a
later date, and taken as an indication of the existence of fast layers.
It is worth noting that the role of the model is, above all, in
this particular case, to predict behaviour, not interpret a finished expe-
riment. In the two examples at hand, which represent realistic limits for
an existing case (Landel 8. Sauty compute a value of A = 2.6 for a doub let in
the Dogger), this prediction is impossible.
5. CONCLUSION
Very permeable strata which have not been discovered could chang~
by several orders of magnitude, the data which are important for the exploi
tation, i.e. those which concern a heat loss in the recovered water of 10
or 20 %.
-734 -
EXPERIMENTAL STUDIES OF ROCK-WATER INTERACTIONS OCCURING DURING
THE CIRCULATION OF A GEOTHERMAL DOUBLET IN A CARBONATE ENVIRONMENT
Abstract
-735 -
Le probleme a resoudre au niveau du reservoir du doublet geothermi-
que est pose ainsi :
ll_l'eau froide reinjectee, agressive visa vis de la calcite, va se
mettre en equilibre chimique avec l'encaissant "froid", aux environs imme-
diats du point de reinjection
2) En circulant vers le point de production, la temperature de l'eau
va suivre le gradient thermique impose par le front de temperature, et pre-
cipitera la quantite de calcite en exces.
Comme nous le montrons, nous ne pouvons determiner les cinetiques de
reaction directement sur la calcaire du reservoir, materiau heterogene, qui
ne peut etre represente par un echantillon.
C'est pourquoi nous abordons les travaux experimentaux de deux
manieres :
du point de vue de cinetique chimique a l'aide du prototype a colon
ne sur des calcaires broyes,
- dans leur aspect physique (porosite, permeabilite), sur des carot-
tes de roche au moyen d'un prototype de percolation a pression.
-736 -
t'I':SIIRE ET
f.NME(: I ~Tlu:tII-: N"r
i-------~-i ·r ll
· r.C02
1 LJ:dl, -ar. t i y i t p Cal C' i tim
t~~UR~ ~t :
1
F.NRI;ClSTIUWI\NT 1
I
·!)F.BIT :
. n :t1l'1;l11I'l'lllm CF.I.I.II I.P.
1
.....
w CEI .I.ULf.
.....
I'RI;C I PIT.
,
l'll:nm I
CHI.UI.P.
'(Q]
IIF.<:IJI.IIT1WR
TF.""'F.ltIlTIIII~;
B - Rappel bibliographique
Les travaux de H. ROQUES, A. GIROU (4) en France et ceux de R.A BERNER
( 5) et L.N. PLUMMER ( 6) aux USA, ont permis d'etablir des equations semi
empiriques de la cinetique de dissolution de la calcite.
Or ce type d'equation permet de determiner une constante cinetique qui
est fonction du modele qui a permis de l'etablir. C'est pourquoi nous
avons tente de trouver une equation plus generale et en tout cas, de defi-
nir la constante propre a notre modele.
C - Les experiences
Lss parametres essentiels qui dirigent les reactions chimiques :
• La temperature: elle est de 35°C a MELUN. C'est Ie domaine des
basses temperatures.
• La pression elle est de 180 bars. Nous sommes dans Ie domaine
des basses pressions.
Nous negligerons son influence directe au niveau du reservoir puis-
que l'ecart par rapport a la solubilite qu'elle peut provoquer sera Ie m€-
me dans tous les points du reservoir, en considerant la pression constante
c'est vrai a grande echelle).
La pression partie lIe de gaz carboni que dissous est un parametre prepon-
derant, dans la dissolution de la calcite. C'est pourquoi nous effectuerons
des manipulations a differentes temperatures, pression partielle de CO 2 ,et
quantites de calcite reactante afin d'etablir des relations entre les
principaux parametres, et par la de connaitre la cinetique de dissolution
de la calcite. L'appareil comprend trois parties principales (fig. 1)
- Une colonne thermostatee contenant la calcite reactante.
- Les cellules de mesure des parametres chimiques de la solution.
- Les appareils annexes assurant : la circulation de l'eau, Ie chauffage
et la regulation thermique, l'enregistrement des divers parametres.
Les debits de chaque manipulation variaient entre 0,02 a 0,6 litre/heure
Le calcaire utilise est Ie calcaire oolithique du Dogger de la vallee
de l'Arman~on (Yonne).
-738-
EVOLUTION DES PARAMETRES CHIMIQUES
0,4 0 ~\ : T 20 °C
\ : M 5 , 6g
~ P.CO z
" \ \,
0,3 \ " _ 1 6,5 1
-1 \. " ,
\
.....
.....
" ..... .....
" "- ..... ..... 6 0,5
"-. ..... S
0,2 - 2 "-
......
......
--
-...... - -- -- -
-. - - -
- ------------ ,
Calcium------ -- ------ -- --------"55
0,1
0,1 -3
= M
ko s -L R.f
o
( Log -L.)dx
x U
Avec
debit
Ca concentration totale en Calcium
Ko constante cinet-1que de la calcite
s surface specifique de calcite reactante
M poids de calcite
L longueur du reacteur
K constante de solubilite de la calcite
(Ca H CO) x
Ux
+
(H ) x
-741-
III. - CONSEQUENCES DE LA DISSOLUTION SUR LA PERMEABILITE DU RESERVOIR
Ce travail a ete effectue a l'aide du prototype de Percolation a pres-
sion, decrit precedement ( 1 ).
Des diverses manipulations realisees sur differentes carottes du re-
servoir de MELUN, on peut conclure que la variation de permeabilite due a
la dissolution est sans aucune me sure avec la quantite de calcite dissoute,
ni avec la variation de la porosite mesuree.
exemple Calcite dissoute 1 % - variation de permeabilite x 100
variation moyenne de la porosite 1 %
Pour chaque echantillon, nous avons obtenu des resultats tres diffe-
rents, empechant toute relation entre ces trois parametres (quantite dis-
soute, permeabilite, porosite), mais nous avons constate Ie developpement
important des heterogeneites preexistantes, ou alors, la creation d'hete-
rogeite dans des echantillons homogenes.
-142 -
Nous pouvons coupler les variations du chimisme aux variations des
parametres physiques du transfert thermique. Les reactions geochimiques in-
duisent une variation de la masSe de calcite contenue dans un volume donne
donc une variation de la parosite moyenne dans ce volume. Les principaux
parametres physiques du transfert couple d'eau et de chaleur susceptibles
d'etre influences par ces reactions sont :
dAA
dn
- la conductivite thermique AA et sa variation --A-
n
A
dYA
- la capacite calorifique YA et
YA =~ _ Ys
YA
) dn
n
dK dn (modele de tubes
- la permeabilite K et 2
K n capillaires)
AA = n AF + (1 - n) AS = 2,64 W/m. oK
d AA AS dn dn
AA =
1 - AA n - 0,14
n
YA = n YF + (1 - n) S = 0,58 thermie/m3.oK
dn 0,13 dn
n n
dk ~
k n
-743 -
1
dk
kj
/
/
/
/
/
/
/ dAA
/ AA
0 ~~
/
--
,
- A
-
________________________________ 1 -Y- _____ dn
,dy"
L-~A-
o 10;:
n
Figure 3 : variations relatives de K, AA A avec une variation relative de
porosit~ Y
ta figure 3 montre clairement que la variation de porosit~
-744 -
thermique. On en deduit la vitesse relative de l'eau par rapport au front
thermique :
On en deduit la masse de calcite precipitee (en fonction de la taille
du front thermique) :
Si on admet que la precipitation est homogene et constante, egale a la
masse qui a ete dissoute, la masse deposee sur la longueur Lp sera :
M = masse de calcite
m avec U vitesse de pore
p
Vth vitesse thermique
t = duree de la precipitation
p
L = largeur de front thermique
p
CONCLUSION
-w-
BIBLIOGRAPHIE
-746 -
IN SITU DETERMINATION OF THE HYDROTHERMAL PROPERTIES
OF A DEEP FRACTURED MEDIUM
BY A SINGLE WELL TECHNIQUE
Summary
-747 -
1. INTRODUCTION
-- ------ --------,
- - --:1
I
,_f.t"
,ch,U.
, ,.~ s.
If
D Ii:
A : thermistor
B : pressure transducer
c: "
D:
E:
P : thermistor
G : flow meter
FIG.l - HAYET DE HONTAGNE
Exnerimental set up
-749 -
In the following, we show, for each experiment, arranged in
chronological order, the injection level identified in relation to the sur-
face, the total injected volume, the duration of the injection, the natu-
rally discharged volume, the mean temperature T of the injected water and
the ~ registered pressure P.
Durat Natural
N° of the Level
Inject
of dischar Dischar.V
Hydraulic - -
Inject T Inject P
volume fractured
experiment (m) injec volume Inject. V °c Bars
area
(liters (min) (liters) %
-750 -
interpret the hydrodynamic and thermal data in terms of "unique equivalent
fissure", as the lack of precise information concerning the geometry of the
fissuring makes it impossible to describe the system in detail.
Exoeriment 12 at 38 m
!O Jo 35 time (mn)
P (bars)
Experiment 4 at 47 m
40
-to
-to -15 lS
time (mn)
-751 -
P(bars
so
80 Experiment 13 at 72 m
70
60
SO
40
Jo
.to
5 iO time (mn)
-752-
4.2. Thermal results
with A*: equivalent thermal conductivity of the fissure (if the dispersion
is nil: A* = Af)
Y*: equivalent heat capacity of the fissure
Yf : heat capacity of the water
</Is: outgoing flow exchanged with the walls of the fissure
e(r,t): temperature in the fissure.
~ computed area
... measurements
observed flow
10
-------,
I
."
I
j
injection discharge
..... _------
2000 4000 6000 8000 time (sec)
A granite
y granite
-754 -
5. CONCLUSION
-755-
WELL TESTING OF LOW TO VERY LOW PERMEABILITY
FRACTURED MEDIA - A REVIEW
Abstract
A constant rate pumping method for which new solutions have been deri-
ved has been tested at Le Mayet de Montagne (France) experimental HoR field.
This method has been compared with a frequency based water test method, in
order to delineate the fields of application of each of them.
1. INTRODUCTION
-756 -
Finally (paragraph 4) an example of a test carried out on an artifi-
cial fracture at Le Mayet de Montagne HDR experimental field is presented.
This example shows that linear models such as those developed until now are
not sufficient to describe the behaviour of fractures under relatively high
pressure conditions. Research is currently being carried out to implement
a model taking into account changes of the geometry of the fracture during
the test. The state of stress is of course taken into consideration in thE
approach.
-757 -
Z=h eponte impermeable as / az:O
puits
fracture horizontale
T kZ~
l
Z=O--------~------~-----------------
kr
eponte impermeable a s / az =0
Fig. 1 - .'thode hmporeUe - .odUe de fracture horizontale
2.1.2. Horizontal fracture moaeL
The model is sketched on figure 1. The reservoir is limited by no-flow
upper and lower boundaries. It has an infinite horizontal extent and a
thickness h, radial and vertical permeabilities kr and kz respectively, and
is completely penetrated by the well approximated by a line source.
A single horizontal plane fracture of radius rf is centered at the
well at an altitude zf in the reservoir.
A single phase slightly compressible fluid flows from the reservoir
into the fracture at a constant rate uniform over the entire volume of the
fracture.
With these conditions, the analytical solution is written ( 6)
r2
t - 0 v2
- 4r
So 2Jo e :r v dv Jx (1 )
o
2 2
-n 71 r n7l zf cos n7lz
(---;:;-r-)
00
x ( 1 + 2 I: exp cos -h- ) dr
n=1 0
h
~ =- Tt
with. to - (2)
Sr 2 Sri
f
ro r/rf (3)
471 kr hs (4)
So Q
hO " -rfh~
~
kz
(5)
-758-
It can be shown that at the beginning of a constant discharge rate
pumping test, the flow is linear from the reservoir matrix to the fracture.
And therefore a log-log plot of drC'wdol'Jn - at pumping well versus producing
time yields a characteristic straight line with a slope 0.5 at early times.
At long times, the drawdown behavior is the same as that indicated by
the Theis solution plus a constant that depends on the measurement point.
IJ~'
~ip
• -PF'
P'to ·~
f
- K,
By matching drawdown data with the type-curves, one obtains two equa-
· tions:
Vkr kz rf (6)
(7)
with (7)
From the above three equations, one can get kr' kz and rf if values
-759 -
for ·the storage coefficient S and the aquifer thickness h are available or
can be estimated.
When the fracture thickness hf is not negligible, analytical solutions
are a bit more complicated ( 6) and require the introduction of a ne~ mat~
matical parameter
hfO
= !!f.
rf
V5i
kz
10' . . . - - - - - - - - , - - - - - - - - - - - - - - - - - - - - - - ,
10 '
ltD'
!
-;
1: 10"
i
~ 10 4
10. 10
10' 10' 10· ' 10' 10 ' 10' 10' I. '
Signal d'entf'lit
Q( )
I Signal de s-or it
p ( )
I
I
,
Matric~
..,I_-t-_Rp
rocheuse
I,
!
-i-·~r-·--=-====-'-r ---" -" r
I Rf I
SCHEMA du MODELE 3
FIGURE 4
-762 -
The input signal is the harmonic injection rate Q (t) = Qo sin 2nft ln
well at the fracture depth. the output signal is the pressure pet) measured
in the fracture at the well level.
We have seen that the modulus of the transfer function is given by
IH (f) I JP(f) I
IQ(f) I
Using the following nomenclature :
Po fluid mass per unit volume at equilibrium
and Z (9)
with d = Log EQ
rf
b) The k resonance frequency is approximately given by
- c 1
fk = 2rf (k + 4) (11 )
with k = 1. 2. 3 ..... n
-763 -
as long as rp is small compared to rf
c) The spectral signature of the system (r p ' rf. e2) can be derived from
the spectral signature of the system (r p ' rf. e 1 ) by the relation:
( 12)
a) !~~~~!f!~~~!~~_~f_~f :
Pratically. there is no need drawing the file of theoretical signatu-
res to get rf. which can be derived from the characteristic frequencies by
using formulas (10) or (11).
We have evaluated the accuracy interval over rf which can be expected
from the use of those characteristic frequencies.
llrf
From f1" we can expect 3% ~ --~ 4% ( 13)
rf
From f2' % ~ t.rf ~ 2 % (14 )
rf
From f •
0
• t.rf 8 % ( 15)
~ ~-- ~
rf
We notice that it will be better to choose f1 (or f2 when possible)
o
than f.
. "1""0IIC1.11\11T1
1If" .... na Q[ u. 'lI.ll'"
rAP =O. 165M I
AF=50.0 M . IiP3IU.JtIIIlJ"JtTS I RP:O.165M
RF:15.0 MI
i i
IIF I~nJilllI[,""nSSJll(
(117.lunfI rx ..,. nCCI.II
etCG.DImcun...u~1.~ E-1.0MM E1('AtK5$llQl; ...... 'f$3ll'
C-ICIlDIT'i" OJ n.utQ[.WO'ftfI
E=O.1MM
N!fN.lll:lIlUnr UIllFVI' .... 'OQQKCIrlJ 1Id!~ IIQ.LIUClL Q.ll\I.ltfL1!II.XIC, 1'O
I. "OII.\lI1.uC.I_
,." P •• 10"
,,' 10"
/ 7 \ ........... 7
./
,
,,' vi I
" --
I ~"cm ~
,,' \ / \ /
~~
,.'
\
.... ,,' \ I
,,' . I
.-, , . 10
" " " " " " "
~t~ ,
II n
"
FN:outI'ittIHZI
",'
flG~ll •
40
" .. . " ..
FJIE:~'KZ)
-764 -
For the above mentioned value of rp and c, it will be sufficient to
scan the i~terval (0 - 60 Hz) in order to get at least f1 when rf varies
fram 5 m to BOD m.
b) !9~~!!f!2~!!~~_~f_~
Formula (12) shows that the characteristic frequencies do not change
when e varies at given rf, rp'
Once rf is calculated, one will have to plot one analytical spectral
signature at known thickness e1
By measuring the vertical translation ~: between the experimental and
theoretical signatures as shown on figure ( 6), it is then possible to get
the actual thickness e2 of the real fracture.
Calculations show that the accuracy over e2 will be about 6 % i f we
assume an absolute error of mm when reading 6 (for logarithmic modulus of
4 cm as in case of figures).
Pratically, and to summarize this method we see that to determine rf
and e with an accuracy better than 4 % for rf and of about 6 % for e, it
is sufficient to measure the experimental spectral signature between 0 and
60 Hz (for rf varying from 5 to BOD m), and to draw ~ analytical spectral
signature at known e, once rf is calculated.
The experimental spectral signature will be directly calculated from
the measured values of the pressure Ip(f) I and the injection rate IQ(f) I,
at given f.
- The time based method can be applied to the vertical fracture case
but till now, this case has not been studied with the frequential method.
Here and now, it seems urgent to test the frequency method on experi-
mental data.
-11>1> -
The fracture)which is situated in the middle of the aquifer. has a
rectangular shape with
a length 2 xf
a width 2 yf
a negligible thickness
The discharge from the aquifer to the fracture is performed with a
constant uniform flux (per unit area of fracture) and at a constant rate.
The calculation method is the "source function method" derived by
A.C. Gringarten and H.J. Ramey (1973).
The source function associated to the rectangular fracture is obtained
as the product of 3 elementary source functions as shown below
/
The product is then simplified considering that the thickness zf of
the fracture is negligible compared to the thickness h of the aquifer.
3.3. Dimensionless notations
The following dimensionless variables are used
- coordinates
Xc xlxf
Yo y/yf
Zo z/h
- drawdown
So = 4TTTs/Q with T h IKx Ky = transmissivity
- Geometry factors
F = (xf/yf) IKy/ Kx (shape factor)
- time
-767 -
3.4. Orawdown at the center of the fracture
The drawdown at the center of the fracture is given by the following
expressions which are easy to calculate numerically
So = J
0
t OF /TiF
erf, 4T • erf ~,p-
'f
4F' (1+2 -1
41Tp2, )
exp (- - - )
h02
d, ( 16)
or
So = j tOF
0 erf
~
I 4T' erf
~
r
hn ( 00
i
axis is given by the following expression.
1 +1 J. -1
1T 2 OXi r XD xo ] 1
'4 F Xo 2 l,erf IT + erf IT . erf xoFIT d, (1 B)
o
The drawdown at an observation well situated along the oy axis is
given by the same expression after replacing F by 1/F.
Expression (1B) has been calculated numerically for the following va-
lues of the parameters
Xo 1, 1.1, 1.2, 1.3, 1.4, 1.5, 2, 3, 10
-768 -
The corresponding type curves have been drawn and are placed in ap-
pendix II.
They show that for F > 1 (observation wells situated along the ox
axis). the curves are very close to theis's solution (xO = m) as soon as
xo > 2.
3.6. Orawdown inside the fracture
The drawdown have been calculated inside the fracture along the length
of the fracture for various dimensionless time. The results which appear on
figure (B) shows that the drawdown is approximatively constant inside the
fracture. It is then possible to derive that the equivalent transmissivity
of the fractured is 'Jery high. The drawdown corresponding to a pumping
test performed in an infinite t:'clnsntissivity horizontal rectangular fracture
is then closely approximated by the formula that we have derived for a
uniform flux horizontal rectangular fracture.
' '~%~:3
' '·E'····'>
tOO". 1,2
OOf.- i Iau-
o 0.2 0,4 ~ 0" I,Q X/XI
3 4
tOf"O.' " I
'IS_~
00'10 I
3.7. Conclusion
A new scheme has been derived for pumping tesmin fractured aquifers.
This scheme is characterized by a single horizontal flat rectangular frac-
ture.
The drawdown has been computed at the center of the fracture and also
at an observation well situated along the principal directions. Types curves
have been drawn which makes it possible to perform pumping test analysis or
to compute the drawdown corresponding to a known rectangular fracture.
-769 -
NOTATIONS
:: I
sO' Oimensionless drawdown
XOl
yo Oimen·sionless coordinates Permeability (hydraulic conductivity)
zO Kz
Xfj
yf
Frae t ure I half length
half width
K IKx"'Ky =
T = Horizontal transmissivity
Horizontal permeability
S = Storage coefficient
t = Time
Ss = Specific storage coefficient
T = Oummy variable representing time
F = Fracture geometry factor
4.1. Introduction
The purpose of Le Mayet de Montagne (France) experimental field, part-
ly sponsored by CCE, is, among others, to make the realisation of hydraulic
fracturing tests possible, with a view to using this technique for HOR pro-
jects.
The rock is a rather homogeneous granite, the petrography and struc-
ture (specially fracturing) of which have been studied in detail.
A number of boreholes have been drilled in this granite. The test the
results of which are presented here was carried out in December 1979 in
borehole INAG 3.3. Prior to this test, this 185 m deep hole was hydrauli-
cally fractured.
The test was carried out on an artificial vertical fracture, supportoo
with fine silica sand, at 155.40 m depth::. This fracture was isolated from
the remainder of the hole by a double packer, and water was injected into
it at a constant flow-rate.
The pressure was measured by a pressure transducer and the pressure-
time curve was directly plotted on a paper recorder. After injection cea-
sed, the recovery was recorded too.
:: Actually, a number of such tests were carried out. Only one is repor-
ted here, as an example.
-770 -
4.2. Interpretation of the results
The injection was carried out during 60 minutes at a 12.90 l/min
(2.15.10- 4 m3/s) rate.
The log-log pressure-time curve (fig. 9 ) exhibits a 0.91 slope during
the40 first seconds. This is much more than the theoretical half unit slo-
pe for a fractured medium.
- '...
Point 1 t 265 s
s = 4.8 bar ~ 48 m
Point 2 ! t
s
= 120 s
= 3 bar ~ 30 m
-+- r;=x-;y = 0.713.10- 6 m2 /s
-772 -
The corrected curve is given fig. 11.
I'r-.ilolc.,
I II
-- -
== r- ~ ~~
~r~
..... p-
-
- I--
-
i=
V Ff tt=f
I
-
Tlf-
II -
-== I=- - 1=
I
t- -
It -
I--
I
- t-
I
-
flflltll{tl
"
~ t = 21 s
f s = 1.9 bar = 10 m
With the same hypotheses as previously, S = 0.23.10- 6, which seems
much more likely than the values obtained without correction of the capa-
city effect.
Of course this method, which is based upon the assumption that the
fracture has an infinite hydraulic conductivity, does not make it possible
to determine the characteristics of the fracture (except its extension, in
the case where measurements can be taken in observation wells).
Only the characteristics of the porous medium equivalent to the actua
fractured rock can be determined. If the frequency of fractures in the rock
is know, it is then possible to calculate an average hydraulic conductivity
for each of them.
Matching of the experimental and theoretical curves is no longer valid
after a certain interval of time. An examination of the p, logt curve
(fig.12) shows that after some time, the pressure begins to diminish, and
after a longer time, increasing resumes, for a constant injection rate.
The same phenomenon was observed for all the tests carried out in this bo-
rehole, on fractures situated at different depths. It should also be noted
that in all cases, the pressure was much less than the "closing pressure"
as determined by hydraulic fracturing. This is not the only point for
which experimental results depart from the theoretical behaviour of the
fracture.
-773 -
Fic;I.12 . lnjeclion test.p It curve In semi- log coordinates
-
(...1
• ('1101
.. .. II
- - -
- I-
'"~
r-
. Jr.
'.C'-..;.L, uo:,M
f- I-r
,
.
I--
.. ..
..
.. . 1I
f - - 1-1-
-
l -I- f - t- 1-t-TMIPI (.)
0
000
In the theory, the recovery curve is the same as the injection (or
pumping) one for early times after the test has ceased.
One can see (fig 13) that this is not the case at all the pressure
remains very high and it takes a very long time to vanish.
This behaviour suggests that the geometry of the fractures (particu-
larly their "effective extension") is not an invariant and varies with the
pressure. Furthermore, this variation is not perfectly reversible and the
experimental results which we got exhibit a king of hysteresis. This may be
due to the fact that the injection process modifies the state of stress in
the rock medium.
Fill.13. Recovery
-774 -
Research work is currently being conducted in order to build a theore-
tical model making it possible to simulate the behaviour observed during
the tests.
It is believed that such a model could be useful not only for hydrau-
lic fracturing, but also for some other stress-flow problems.
5. CONCLUSION
Two methods for determining the hydraulic characteristics of fractured
media have been considered in this paper. The time-based method, the field
of application of which has been widened as concerns the geometry of the
fracture, does not make it possible to determine the thickness of the frac-
ture (that is its hydraulic conductivity). Only the bulk hydraulic charac-
teristics of the rock and the extension of the fracture (if measurements
are made in observation wells) can be determined. These parameters are not
the most important ones for Hot dry rocks projects.
Another point is that, in the case of injection tests, even with mode-
rate pressures, the behaviour of the fracture quickly departs from the
theoretical one. It is believed that this is due to changes in the "effec-
tive" extension of the fracture during the test. Work is under progress
to check this point.
As for the frequency-based method, which seems very promising, it
still has to be tested in situ.
-775-
REFERENCES
FRAS G., JOUANNA P. - Contrat DGRST nO STA/P 38b. Rapport de synthese provi-
(3) soire. Universite des sciences et techniques du Languedoc.
Montpellier (juillet 1978)
GRINGARTEN A.C. and RAMEY H.J. (1973,) - The use of source and Green's func-
(4) tions in solving unsteady flow problems in reservoirs.
Soc. Pet. Eng. J. Oct 1973
RAGHAVAN R., URAIET A., THOMAS G.W., (1976) - Vertical fracture height ef-
(7)· fect on transient flow behavior paper SPE 6016 presented at the SPE
AIME 51 st Annual fall technical conference and exhibition : New-
Orleans. Oct. 1976.
THILLIER J.M., JOUANNA P. - Mise au point d'un capteur harmonique pour de-
(8) tection simultanee du nombre et de l'epaisseur des fissures dans un
sondage. Rapport d'avancement nO 2. CEE contrat nO 565 78 1 EDF. Con-
trat C 121. Univ. des sciences et techniquesdu Languedoc.
Montpellier (juin 1979)
-176-
APPENDIX 1
10·~~~~~ ____~~~~~____~~~~
10· L:-~-"--:~='"'--~"""'---'-~~~
10" 10'" 10'" 1 10\,. :,;; 10' 10'" 10'" 10'" 1 10'~.·.:y~ 10'
FlUE 1 ~ POMPAc:E AU CENTRE 0 lINE FISSURE RECTA""""'A IRE
F1QH1£ 2 - POI'IPACOE AU CENTRE D lINE FISS~ RECTANGlLAIRE
FACTEUR DE FDRrIE. 1.0
FACTE~ DE FOfV'IE_ 2.0
10·L:~~~~~--~~--~~----~~~~ 10·~~~-L~~~~~~~~------~--~
10'" 10'" 10" 1 10'~ '" :;,: 10' 10'" 10" 10'" 1 10'lJ,_1.f ;,! 10'
~ - POI'PA;E AU CENTRE 0 ur£ F I&SUAE AECTANQJL.AIRE ~ POMPAI:E AU CENTRE 0 lINE FISSURE RECTANC;Ll.AlRE
FACTEUR DE FOME- 5.0 FACT[UR DE FORME .. 10.
-777 -
.
10'
",'
". T- ....
10 •
'0-
10 ·
1~~r---------~~~~----~------~~-r------~~~
·oa 0
-778 -
w ' ~ ____ ~~~ ______ ~ ________ ~ ____ ~
10" 1l1iU1(' . ":~ N.J UJmIl £:I UC1 ntsuJIC IIIlttlolCl.A.A~ ' ••'f,:;-T-,' W'
-- .0. 1 . 500
,,' ~----~------~------~-----~
w'
",.
-779 -
APPENDIX II
fog· T -
FlGUIIEll
-780-
Flr.tJII(18
'0 ·
-781 -
RESULTS OF THE NEW GEOTHERMAL DOMESTIC HEATING SYSTEM AT CREIL
P. JAUD
Abstract
-782 -
LE CHAUFFAGE GEOTHERMIQUE DE CREIL
L'operation de chauffage geothermique de Creil est l'une des premieres
installations de chauffage geothermique qui ait ete realisee en France
(1976). Elle presente l'originalite d'utiliser, en liaison avec les echan-
geurs geothermiques, plusieurs pompes a chaleur qui permettent d'exploiter
au mieux l'eau chaude disponible a la sortie des forages. Pour cette raison
Electricite de France a entrepris d'effectuer des mesures, en collaboration
avec l'Office d'H.L.M. de Creil, en vue de determiner Ie bilan d'exploi-
tation reel de l'installation pendant toute une saison de chauffage.
Aspect geologique.
L'eau chaude geothermale utilisee a Creil provient de la nappe aqui-
fere du Dogger qui est la mieux connue de tout Ie sous-sol du bassin pari-
sien grace aux nombreux forages qui ont ete effectues pour la prospection
petroliere. Quatre forages, d'une profondeur de 1 700 metres, ont ete rea-
lises pour permettre d'extraire puis de reinjecter cette eau geothermale.
Leur implantation a ete calculee pour qu'il n'y ait pas de baisse de tem-
perature aux puits de production pendant une duree minimum de 50 ans (voir
figure 1).
-783 -
*
N
Aspect thermique.
Le systeme de chauffage geothermique de Greil a ete con~u pour ex-
traire le plus possible de chaleur geothermique. et par consequent pour
reinjecter l'eau geothermale a basse temperature. Pour cela il est fait
appel a 3 pompes a chaleur disposees en serie a contre-courant. dont la
puissance electrique totale est de 2 112 kW.
-784 -
La production de chaleur peut etre assuree par 3 sources energetiques
differentes :
5800lh/h
16,7 "W)
!lO'C
""==='- -- -====:J3;:=======~=~'200-2000
LOGE"I:NTS • AAO'ATEUIIS ·
~'·c ~.(
• [AU CHAuOC 5ANITAIJ-jE
-785 -
Le schema de l'installation est relativement complexe, mais il pre-
sente l'avantage de tirer parti de la presence des deux reseaux de chauf-
fage : Ie circuit des Cavees - planchers chauffants avec retours a basse
temperature - et Ie circuit de la Z.A.C. - radiateurs avec des retours a
moyenne temperature. En faisant passer l'eau du circuit "planchers chauf-
fants"dans les evaporateurs et l'eau du circuit "radiateurs" dans les
condenseurs des pompes a chaleur, on opere un transfert de calories du
circuit basse temperature vers Ie circuit moyenne temperature, et il de-
vient alors possible d'utiliser dans les echangeurs l'eau sous-refroidie,
afin d'epuiser plus completement la chaleur contenue dans l'eau geother-
male.
-786 -
Cependant 1es resultats obtenus ont montre qu'un certain nombre d'hy-
potheses n'ont pas ete confirmees.
-787 -
Lorsque l'on fait une analyse plus fine des resultats en faisant
appel a un
programme de calcul sur ordinateur avec l'hypothese d'un debit
geothermique constant de 170 m3/h et un fonctionnement hors pannes, toutes
choses etant egales par ailleurs, on trouve les repartitions donnees par
le tableau ci-dessous.
Q~2~h~!!!~_!Y~£_~2!2~!_!_£h!!~~! Equivalent
Production de chaleur (en Gcal) : en T.E.P.**
2 237
- chaudieres d'appoint ••••••• 17 200 (35,6 %)
- echangeurs + pompes a
chaleur •••••••••••••••••••• 31 200 (64,4 %)
Total •••••••••••••••••••••• 48 400 Gcal
Consommation d'electricite (en MWh)
- pompes a chaleur ••••••••••• 4 045
- chaufferie 1 744
- pompes des forages ••••••••• 844
Total •••••••••••••••••••••• 6 633 MWh 1 459
"3""m"
fh!~!!~!!~_~!!~!~!2~~~!!~·
Production de chaleur •••••••• 48 400 Gcal 6 288
Consommation d'electricite
en chaufferie •••••••••••••••• 1 620 MWh 356
bm
§!~!!~!2~_~:~~_!2~£~!2~~~~~_!!~!_£2!2~!
!_£h!!~~!_!Y~£_~~! e£h!~g~~!_g~2~h~!!!g~!
2E~!!!!!!~!
Production de chaleur (en Gcal) :
- chaudiere d'appoint 29 450 (60,8 %) 3 826
- echangeurs seuls optimises • 18 950 (39,2 %)
Total •••••••••••••••••••••• 48 400 Gcal
-788-
Cette analyse permet de confirmer l'interet energetique du chauffage
geothermique retenu, meme en ce qui concerne 1e bi1an en energie primaire
(2950 T.E.P. soit un gain de 44 % par rapport a un chauffage traditionne1)
10rsque l'on convertit 1a consommation e1ectrique des pompes a chaleur en
prenant l'equiva1ence de I MWh pour 0,22 T.E.P.
••••
3000 4000
-189-
IV - ASPECTS ECONOMIQUES.
Geothermie :
Equipements speciaux
-790 -
~~-~~~!-~~-~~~~!!~~~.
Pour la saison de chauffage 1977-1978 Ie compte des depenses s'etablit
comme suit :
v - CONCLUSIONS.
Les mesures effectuees par E.D.F. ont permis de mieux comprendre Ie
fonctionnement de l'installation de chauffage geothermique de Creil equipee
a pompes a chaleur, et les bilans energetiques montrent que cette instal-
lation permet des economies d'energie primaire sans entrainer l'augmen-
tation des charges de chauffage payees par les locataires.
Les economies totales d'energie primaire ont atteint presque 3000 TEP
pour la saison de chauffage 1977-1978, et il faut remarquer que dans ce
domaine, les pompes a chaleur jouent un role benefique puisqu'elles per-
-791-
mettent une economie annuelle supplementaire de 700 TEP, deduction faite
de la part correspondant a leur consommation d'electricite.
-792-
OPTIMISATION D'UN CHAUFFAGE GEOTHERMIQUE EN FONCTION
DES DIVERSES CONDITIONS CLIMATIQUES.
D. LAMETHE-PARNEIX
Abstract :
-793 -
I. INTRODUCTION
En general, la decision de realiser une installation de chauffage a
partir de la geothermie necessite la connaissance de nombreuses donnees
et certains parametres, tels que les caracteristiques de la ressource
geothermique, ne sont pas fixes au moment du choix. II est donc tres
utile de connaitre la sensibilite de la rentabilite d'une operation geo-
thermique a l'ensemble de ces parametres. Un tel probleme a necessite Ie
recours a des programmes de calculs automatiques.
Ces etudes menees par EDF et OET ont ete reparties entre nos orga-
nismes de la fa~on suivante : EDF a plus particulierement etudie l'in-
fluence des parametres au niveau des forages (debit, temperature de l'eau
geothermique) et du reseau alors qu'OET s'interessait a l'influence du
climat et des types de logement. Au niveau de la chaufferie centrale, des
etudes semblables ont ete menees avec deux schemas d'insertion des pompes
a chaleur differents.
Pour toutes les courbes et tous les abaques etablis par EDF, les
memes conditions ont ete conservees pour les caracteristiques des loge-
ments et Ie schema de raccordement du reseau de chauffage a la geothermie.
2. RAPPEL DU PROGRAMME
Le programme de calcul GEOPAC EDF-INFRATEL traite les donnees de
l'operation geothermique, et effectue les calculs thermiques et energe-
tiques relatifs a divers systemes de chauffage faisant appel a la geo-
thermie.
Tous les resultats presentes sur les courbes ont ete obtenus avec
les hypotheses suivantes :
-796 -
CHAUFFERIE
LECTURE DES DONN~ES
¢::: ,--C,;;.;E;;.,;. NT.;. ;,;R;.;. ;,;AL;;,;;;E.. . .,J
DONN~ES
UTILI SATION ~CONOMIQUES
I ~ NOMBRE
D'UTILISATEURS
CALCUL DU SIMPLE
~CHANGE
DIFF~RENTS SCH~MAS
QUI
BILANS ~NERG~TIQUES
BILANS ~CONOMIQUES
LOQemenls
Biposse
, . ' .•5W/",)·C
150000
lIOGU
lSClOGCI
101'"
'00000
uaoo
15011'
JIM'
10GlOO
50000
1110.
",,-~-~~~-:.7.,..::-~-~-~-~-:;,,=,,:-,~_\l"
'01100 "'--,::'
..
FIGURE 3
EVOLUTION DES BESOINS ANNUElS EN FOIICTION DU IlOmRE DE LDGEI'EHTS
-798 -
Les parametres les plus influents pour une operation de geothermie sont
les conditions geothermiques (debit~temperature) comme on le voit sur
les figures suivantes (figures 4 et 5).
C"".4. I.U'I,,,,i. r,jllL e
41. ,.UI"
..
c ....... " ..
"
511 U"C
)SlXIl','''''fl.lt
~12
IO'C )000 I.,"'''fltl
1'O/50·C
11
"
"
lOG ... '~~------~,,------~.~.-------7.-------7.,,~. ..
f~ .. i'ohl'l ( "e l
FIGURE 4
FItuP.E 5
IIIFLUENCE DES COrlDITIOIiS GEOTHEPJ1I~UES <TE!:PERATURE - DEBIT>
HlFLUErICE DE LA TEI':PEr.ATURE GEOTHERt:I~UE
SUR L'EFFICACITE GEOTHEP~ICUE
SUR LE COUT DE LA THERIHE UTILE
POUR DIFFERENTS DEBITS GEOTHERf':I~UES
~~~a
.. e
!tlO"'~,,,
"
.. ~f-----"
.,
>.
" @]
FIGURE 7
FIGURE 6
INFLUENCE DE LA THl'ERATURE DE SORTIE CES ELEMENTS DE CHAUFFE
H/FLUENCE DES S'!STEfES IlHERIEURS
SUR LA PART D'EfIERCIE GEOTHERr':IOUE
SUR L' EFF I CACITE GEOTHERfi ~UE
DA.~S LA FOURIHTURE DE CHALEUR
- 800-
(..til • • l. lhH""I. "th •
•• e".vt,,-,••clu.li,it ."' ''~
"
"
11
"
II
10
FIGURE 8
INFLUEllCE DU COUT DE L'EtIERSIE D'r./'?OINT
SUR LE COUT DE LA THERI11E UTILE DE CHAUFFAGE
4. PRESENTATION D'ABAQUES.
Les nombreux calculs que nous avons effectues a l'aide du programme
informatique GEOPAC - EDF - INFRATEL nous ont permis de tracer des abaques
qui permettent d'estimer l'interet d'un projet de chauffage geothermique
lorsque les parametres essentiels sont connus.
Des exemples de ces abaques sont representes sur les figures 9 et 10.
Sur la figure 9 est indique Ie cout de la thermie utile de chauffage geo-
thermique avec pompe a chaleur pour une operation definie par la tempera-
ture de l'eau geothermique, Ie debit etant de 200 m3/h, et les logements
equipes de radiateurs 90/70 0 C.
De tels abaques ont ete traces pour trois debits geothermiques 100,
200 et 300 m3/h et pour trois types de systemes interieurs (radiateurs
90/70 0 C, ou 70/50 0 C et planchers chauffants 55/40 0 C).
- 801-
t::oiit •• 10 th.rmi, "tile
• c ..... ,fa,••ctuaU .....n .....
..,-
15
-"/,.c;._ _ _ _ _ ssoe
r"/,..c...______ ,•. C
10
7-7'------- ,0'C
-----to·c
7L-_~_~
20 ]0 40 so 50 t. 90 ,.
_ _L-_-L_~_ _L-_-L_~L-_~_-L_-.
100 110 no NIT
1/ : IO'BRE DE LilGa"a/TS
FIGURE 9
3000
1000
1000 \
\
\.
2000
'00L-__- i____4-__- i____4-__- L____4-__- L____L -__- L____4-__•
20 ]0 U 50 U 70 to go laO 110 120 Nif
tl tUilRE DE LOGEMEIITS
FIGURE 10
- 802-
STUDY AND OPTIMIZATION OF A HIGH EFFICIENCY CONVECTOR
Abstract
-~-
I. INTRODUCTION
The present work concerns the development of a convector for the heat-
ing of dwellings, adapted to the use of tepid water. The feasability
of the use of tepid geothermal water without heat pump (less than
55°C) depends on the return temperature of the heating water. Return
temperatures below 30 °c are very attractive. In a previous work
SOCIETE BERTIN & Cie of Plaisir (FRANCE) showed the feasability of co~
vectors using natural draft and adapted to tepid water. A mock-up of
convector was built. Its thermal power was more than 1.5 kWth with
water at 50°C. This mock-up had the same area on the ground as a con-
ventional convector of the same power with water at 80 °C, but used a
high duct (about the height of the room) in order to increase the pres
sure head.
The present work concerns the optimization of the exchanger area which
is made of a new type finned tubes, and the building of a prototype.
The technOlogy used in the building of the prototype was selected to
ensure low cost in the context of mass production. The latter remark
is very important for the applicability of the results of the research
-804 -
III. ~ESCRIPTION OF THE EXPERIMENTS
The exchanger surface tested in the work presented here are dimension-
ned with different spacings between fins and different heights of fins.
The other parameters of the surface like the kind of metal for fin or
tube, the diameter of tubes, the length and the thickness of the fins
have an influence either very weak or well known on the heat transfer
number and the pressure loss. As compared to conventional finned tubes
installed in high temperature convectors, the spacing between fins is
much larger and the height of the fins is much smaller.
-~-
IV. CHARACTERISTICS OF THE PROTOTYPE
In the final step of this work, we have built a prototype and tested
it in a quiet room to determine its thermal characteristics. The pro-
totype has also been tested by a French technical center
(CENTRE TECHNIQUE DES INDUSTRIES AERAULlQUES ET TECHNIQUES - CETIAT)
which confirmed our results. The thermal characteristics are given in
fiqures II, III, IV.
The width of the prototype is 0.67 m and its depth is 0.187 m. These
dimensions are common for a conventional convector with thermal power
of 750 Watts with water temperature at 80°C.
The tests conditions are that of a geothermal water with large inlet/
outlet temperature difference. In fiqure II, the variation of the ther-
mal power is plotted on figure II against inlet temperature for a va-
riety of flow rates between 15 and 35 l/hr. The variation ~f the thermal
power against flow rate for a variety of water inlet temperatures is
between 40 and 70°C. Finally water outlet temperatures are plotted
against thermal power with water inlet temperature as a parameter on
fiqure IV. All these characteristics are given for the same conditions
of temperature: 19°C measured at 0.75 m above the floor. The nominal
operating conditions are :
thermal power : 700 Watts
water flow rate : 25 l/hr
water inlet temperature : 55°C
water outlet temperature : 30°C
-806 -
the water outlet temperature versus the height of the duct for the fo110
wing conditions :
water flow rate : 48 l/hr
water inlet temperature 50°C
air temperature : 19°C
It can be pointed out that the influence of the height is reduced for the
conventional convector above about 0.7 m and even a very high duct can-
not change a conventional convector into a convector afapted to geother-
mal heating. The outlet temperature would not be low enough.
The characteristics on the air side of the new convector bring large ad
vantages with respect to home comfort and control of operation. The im-
portance of the air flow rate through the convector corresponds to about
5 volumes per hour for a room. It induces a fair homogenization of the
air temperature. A comparison between gradients in a room heated with
conventional convector and with BERTIN convector shows that the mean
temperature of the occupied zone can be reduced at least by 1°C. This
value corresponds to a reduction of energy consumption of about 7 %.
6. CONCLUSION
The results of the present work consist in the optimization and the
construction of a prototype of convector which allows to supply the
whole heating in dwellings with geothermal water at 55°C with competi-
tive prices which can help the development of direct geothermal heating.
The characteristics of this convector are such that the water outlet
temperature can be constantly below 30 °C in any climatic conditions.
This is very important for the profitability of the geothermal heating.
-~-
heating power 35
W (watts) 30
11
25
mo
9)0 20
SOO
700
600
Fig 1_ High efficiency COf'Pt'eCtor
500
outlet water temperature
400
Ts is obtained from :
300 Ts =Te - P / 1,16 Q
Ts
45 room temperature 19°C
40
7
roo
500
40
400
300
15 Heating power W (watts)
2001+-_ _ _ _water flow rate (1/ hr )
---=...:..:...----.--~-r-
400 9JO 600 700 800 9JO 1000
10 20 lJ 40
Fig ]I[ _ High efficiency convector Fig 12" _ High efficiency convector
Heating power versus water rate flow Outlet water temperature Ts versus heating
for various inlet water temperatures power W for various inld water tempera-
tures Te.
-808 -
heating power outlet temperature
W(watts) Ts
1000 40
Finimetal 3 HS 100
500 ~
inlet water temperature Te =50°C
30
inlet water temperature Te = sooC
water rate flOl.Y Q =~ 1/ hr water rate flow Q = ~ I/hr
Or-__~~'g~~~~__d_raft
__c~hi_m_~~(_m~m__
) 20 height of draft chimney( mm)
500 1000 1500 a 500 1000 1500
Fig Y _ Comparison with classical Fig i [ _ Comparison with conventional
convector. Heating power versus convector. Outlet water temperature
height of draft chimney. versus height of draft chimney.
water evolution of
temperature water rate flow
55 (OC ) % 100
30 40
30
25
20
20
10
I temperature(OC)
15
.5 a 5 20
Fig 'lZI[ _ Caracteristics of high efficiency convector
versus climatic condition.
-809 -
TECHNICAL AND ECONOMIC FEASIBILITY
IN THE E.E.C.
J.M. LEJEUNE*
DEPARTEMENT GEOTHERMIE
Abstract
-810 -
RESUME
-811-
INTRODUCTION
I - De la ressource a l'exploitation
-812 -
Une etude de faisabilite et d'avant-projet type integrera
toutes les donnees et hypotheses concernant la ressource et l'utilisation
permettant de definir successivement un schema technique general, un
programme de forage, des installations de surface, des reseaux de
distribution ..• L'analyse des bilans economiques et energetiques traduit
la viabilite du projet dont la realisation totale est soumise a un
planning s'etalant generalement sur plusieurs annees.
Couverture des
Certification et etude des
differentes ressources exploi tables .. besoins par 1.
geothermie
(=F solutions!
Dcmnees econOllli'l'ulOs
et financierss
V
- - - - . . Bilans economiques
et financiers
Bilans energetiques
(gain en Imergie prima ire)
- 813-
II - Utilisations de la geothernne basse energie
-814-
Le chauffage de locaux ou d'equipements specifiques (ex: piscines)
ainsi que leur climatisation posent des problemes particuliers a analyser
cas pour cas, mais peuvent etre tres favorables compte tenu de niveaux de
temperature parfois tres faibles.
000
-banene dOkhilllge Couche d. m6l1 - - - - - - -
.COO-.. gfolht.!tmique/aU'. a..tchCl 0:>0>
&$.5l,Ullnl If: pr«houH.ge:
de l'iLlI eJeltntuJ
AIR ~ e
~ ~
II
ern> - 10'CI
:>
c ,I".
AJR~
t;.=====E:===~,-=_=~~==:;.</ __ __h_"_
Sc n _,a._typ~_e d'une touraille de ..khage
JL/ UTILISATION INDUSTRIELLE
SO'c 15'C ./ Rechauffage d'air de sikhage par
geothermie (appoint gaz)
- 815-
III - Bilan des operations existantes
-816 -
Etude complementaire modelisation des coOts de production
Nota Les programmes ont ete con~us pour etre utilisables sur des or-
dinateurs de bureau de petite taille.
- 817-
SlRUCnJRE PROGRA.'NES ECOGET ECOGES
lMODELISATION - F'AISABILITE ECONOIIIQUE DES PROJETS GEOTHERIIIE)
ECOGET
MODULE RESSOURCES
ECOGES
MODULE COUTS D' INVEST] SSEllENTS
Forages, reseau geothermal, 'changeurs,
pompes.
COUT GEOTHERIIIE
\_----~
-818-
CD
(enM' cOlh D'UN DOUBLET
valeur 1/ I /80 CF
A I
25 (on ,,~ COUl 0 UN FORAGE DROIT
valeur 1/1/80 HT
12. 5
20
10
15
7.5
~
'Q
10
5.
/' '.'
,," 1.7 _ /
" ....... "
... :t...,....--"
2.5
500 1000 1500 2 000 2500 3000 500 750 1000 1500 2000 2500 3000 PF(m)
SESSION VII - Geophysics II - Electrics, Magnetics and
electro-magnetics
-821-
APPLICATION OF D.C. DIPOLAR METHODS IN THE UPPER RHINE GRABEN
R. BAUDU, J. BERNARD, J.M. GEORGEL, P. GRIVEAU, R. RUGO
Departement Geophysique,
Bureau de Recherches Geologiques et Minieres
Abstract
The aim of this project was to improve both the technology and the
methodology of the electrical D.C. soundings when used to prospect geother-
mal reservoirs.
- 823-
The aim of this project was to improve both the technology and the
methodology of the electrical D.C. sounding used to prospect geothermal
reservoirs. The program consisted of three main steps :
1. The study of the response and performance of the various electrode ar-
rays.
2. The adaptation of a computerised measuring eqUipment.
3. The experimentation on a well known geothermal anomaly of the Rhine
Graben.
We took as criteria
s1
the Schlumberger (p apparent resistiVities :
Pee' = 1 - ~ where
ps = slope of Schlumberger curve
Ps a
a " 2 + tg e tg e' (fig. 3)
(iii) - the intensity of received signal, not for a given spacing bet-
ween the dipole centers, but for a given depth of investigation.
- 824-
However, the dipolar nature of the sourc.e field allways results
in low level measurements at large spacings. This can be improved through
the use of a large AB, (a bipole) A B
,,
since the apparent resistivity
remains of the equatorial (i.e. ,
,
,,
Schlumberger) type, when the ,Ip
distance AP is taken as the 1'1
~
N
effective array spacing.
This process allows the operator to see the signal after a cer-
tain number of stackings and to decide wether the signal is too much alte-
red by the noise (in this case be eliminates this signal) or is good enough
to be kept in a memory. The operator stops the process when he estimates
that the stacked signal is fair.
3. Experimental results
-m-
IAn. r" I / J A Mt lN ~B /
9'
Fig 1 DIPOLE'-DIPOLE ARRAYS USE LESS WIRE THAN THE SCHLUMBERGF:R ARRAY I\: azimulhal EQUATORIAL
.90' ,
1.0
2.19 9 . l g 9 '
(3::
t • '99.199'
equatorial rad ial
~
Y-- azimuthal
, .. 1.0 .. 8 ""I...
,, perpendicular
2.0
_+ parallel
,
, ,,
, I
" _
00 ~
" -. - polar o -~ .. t ' O ? i 9 O· ...
N 9
....., A B
R 7 p
... E .',0
1711///1/1// P..
'. 9 f ..·
I • -
P I
/3' - ;;\-- -
.. ~; Ee.~
I
, r III ~1. ~
I
I
I
I
-90'1""" ' -- 1.0 \J
N.B . ASY"'.ptotlt position "'" e
A
~\~-- In",.stlguion depth
B I .1
[Hac t!ve array si z.. " 8
F; II 4 HAP or THE: LttPTH PARAHLTER 8
fig 3 GENERAL DIPOLE-DIPOLE ARRAY
Number of stackings
6)( 4
J 20 ~V
....co
'-0
o 5 Km
~
~ I so tttnperiturf! c.une cOIfIPuted .t the Buntnndsteln top (. C)
_ ''20 - -
(, F'.ult
w
l ! t t E
o "00 1000 4500 5000 7500
,
A , >4 N
~.
,-=--
,, ,, COLLINEAR DIPOLE DIPOLE PROFILE
''''r. . Gunstett Surbour9 )
Fig 8
10
+
~
",CJ> ~~
0
Ol
0
t"
SoPOLAIRE EW
+
_ _ _ _ _-.!..B~ "
~+
"
"
.,..
'\
,!f>~ ~~
~.
"'~ "'~
(') N 0
•
p-
Ol
m
N
-.j
(')
t"
t" t"
SoPOLAIRE NS
Fig 9 : E-W and N-$ polar dipole dipole soundings with the proposed interpretations
x experimental data , not taken into account during the adjustment step
- 831 -
t..
.
>
'0
~
I
\
I
~
~
1 ~
-= .: .
~ .I
>
;; '"~ t
o
i
_~
0 ~
z !
:. i
:;;
j ::
'"
§
I
I
- 832-
DIPOLE-DIPOLE Sl'UDY OF THE TRAVALE GEOTHERMAL FIELD
Sumnary
The present paper deals with the application of the dipole sounding
rrethod in the Travale geothermal area ('fuscany) and is a part of a wider
research addressed to a systematic study of the exploring power of the rre-
thod. It constitutes a first attempt for the realization of an efficacious
prospecting technique in areas of geothernal interest. For this reason the
choice of the lalown geothennal area of Travale appears largely justified
owing to the nurrerous infonnations at present available about geological,
hydrogeological, geophysical and geothennal aspects. We give at first a
brief outline of the adopted field technique and of the method used for
data acquisition. '!hen, after showing the obtained field dipole apparent
resistivity diagrams, the method of transfonnation into Schlt.mi>erger equi-
valent curves is discussed together with its implications as far as quan-
titative interpretation is concerned. 'IWo preliminary geological sections
are finally presented as a result of a combined geoelectrical, geological
and geothermal interpretation.
- 833-
1. INTRODUCTION
3. FIELD RESULTS
We carried out sixteen deep dipole soundings (fig.l). In particular
soundings TRl and TR6 were sited so as to be directly calibrated by local
wellS, the stratigraphic columns of which were already known. Fig.3 shows
the obtained apparent resistivity diagrams against the spacing. These cur-
ves appear strongly influenced by lateral effects, among which the most
conspicuous are thought to contain informations about significant geologi-
cal structures in the underground.
- 835-
To obtain the resistivity model of the underground from the analysis
of the apparent resistivity diagrams, we used the mathod of transfomation
into Schlumberger curves (3), that pernri.ts the adoption of any one of the
well known quantitative interpretation systems at present available for
Schlumberger diagrams. The mathod originates from the relation (2), (3),
(5):
~~(r) = ~~(r) - (r/2)d~~(r)/dr
connecting the axial dipole apparent resistivity Q~(r) with the Schlumber-
ger one ~!(r). Considered as a linear differential equation, the above for-
mIa, after integration, gives:
~! ~ -2r2J&~ ,,,3]
(r) (r) dr
5. ROCK RESISTIVITIES
On the basis of the obtained results and taking into consideration al-
so the results of previous ~oelectrical conventional surveys (7), (8), we
were able to compile the following stratigraphic column of the resistivi-
ties, schematically reduced to three electrostratigraphic units.
On account of the lithology and of the mean temperature, the resisti-
vities of the fomations of the impermeable cover vary normally in the ran-
~ 10-40 !In for the prevailingly clayey fomations, and reach 200-250 Om
for the prevailingly marly-calcareous or calcarenitic ones. These resisti-
vities decrease notably, even of an order of magnitude, in correspondence
with the geothemal areas interested by surficial manifestations and by hi-
gh temperatures. The resistivity of the carbonate formations of the reser-
voir is about 1000 Om for cold surficial carbonate rocks, in the range 20-
100 Om for permeable hot rocks characterized by intense circulation of hi-
ghly saline geothermal fluids, and appears not easily evaluable, but very
likely in the range 100-1000 Om for deep carbonate rocks with rather high
temperature, low fluid content and low permeability. The metamorphic com-
plex, ~nerally characterized by low permeabilities and low fluid content,
has average resistivities decidedly higher than those of the overlying com-
plexes, so that it is attributed the role of resistant basement (greater
than 1000 Om).
- 837-
6. EXPLORJNG POWER OF THE MEI'HOD
The quantitative analysis of the ~tric soundings 'IRl and 'IR6 put
in light the good capacity of the dipolar method of evidentiating the for-
mations interested by the presence of circulating endogenous fluids, and
of differentiating them from the over and underlying fomations, such as
the more conductive ones of the covering impermeable complex and the nota-
bly less conductive ones of the regional basement. Of particular interest
was also the tentative interpretation of the lateral effect of the soun-
ding curve 'IRl, whose right-hand final portion shows a slope toward high
apparent resistivity values greater than the ~ permissible for pla-
ne parallel layers (fig.3). The interpretation of this effect in terms of
a sharp temperature contrast between a relatively "cold" zone under the
center of sounding 'IRl and a "hot" zone under sounding 'IR6 (fig.5), where
no geological lateral contact exists, may be retained, in our opinion, as
one of the most significative results of the discriminating power of the
method, at least in the particular geological context around the centres
of the two soundings.
Finally, the global analysis of all the examined soundings showed the
good capacity of the method of distinguishing, in the general geological
situation of the Travale geothemal field, the various structural features
such as the cover, substratum, uplifted structures and tectonic depressi-
ons.
REFERENCES
(4) LODDO M. and PATELLA D., 1977, Spectral Analysis of Volt~e Measure-
ments for the Interpretation of Dipole Geoelectr1cal Soundin-
gs in Low Resistivity Geothemal Areas, Geotherml.cs, 6, 227-
236.
(5) ALPIN 1.M., 1950, The Theory of Dipole Soundings, Gostoptekhizdat, Mo-
scow.
(6) PATELLA.n., ROSSI A. and TRAMACERE A., 1979, First Results of the Ap-
-838 -
lication of the Di ole Electrical Soundi Method in the Geo-
thermal Area of Travale-Radicondoli Tuscany , Geothermics
(in press).
(7) ALBO M., CAMELI G.M. and CATALDI R., 1973, Does TellJlerature Affect E-
lectric Resistivity of Rocks in the Larderello Region ?, Geo-
thernrl.cs, 2.
(8) ALSO M., CAMELI G.M. and CATALDI R., 1974, Studio della conducibilita
elettrica delle rocce nell'area di Larderello, Publ. ENEL.
0 1) D2) ~~~~~\ 3)
., .. .. . .. , .....
, ~~~ 84) B!its) [[[illS)
7)- 8 ....
2 4 6, ekm
: : : : : : . : : : : 51£ NA:
• 0-
. .'
.r----\
................
F--"t;p-;. . .
. . • • . • •
...
"
.....
", ....
o •
.. '
•.....
"
"
.. '
~,
.
"
~:
"
, '
"
I\~
"
TAf TA2 TA3 TA4
"
TAe TAB
.. ..... ,.
,
... m
•
I\~
m
• .... ... .... m
... ... • m
n~
. ..
.. . .. :.
~
' "
'" .... m
• ... WI' . ... ...
Fig,3 - Dipole apparent resistivity diagrams,
- 840-
..
TR1 ? TR2 TR3 .. ~ TR4
.,'
m
,0" tis' tis' R tifo tii' tis' ,.' Wi' A ia"
TR7 TR8
. 't:
..
., .. a::
10
tis' R tifo
,
I. fri' tis' tifo R tis' tifo A
TR10 TR11
TR9
, ..
.1:: ...
.: .t
- 841 -
TRl TR6 TR2 TR5
I I FtI ~r:. ... 1
A5 oom
F,
r--//I\ /I~~
o "
,
, . /' ''... !] , ,
-5 00
-'-~
......' 1,\
\. \,\
,
,
, "
,.,&
I \. 1-/
+ -'< -1000
__ , -....0:::
- 10 00
/' I
I
I
I \ ' , / .'.- ..: '\ _ __ _I
' , 10 0 .........
I <"
J'
~
' / I / - ( ;:'\
1500 " -1500
'" ~
I
~\ "', . . . ,
__ ... \~o
0'
- 20'00- ~ "'\1'
00
I
I, 2Km
I -2000
'~O(f'
500 m
,§
m
0 0
2[]
~
4m
- 500 - 500
3j
- 1000 -1000
sd'-r
- 1500 -1500
- 200
A.
-2000
- 842-
DISTRIBUTION OF ELECTRICAL CONDUCTIVITY IN CRUST AND UPPER MANTLE
IN THE AREA OF THE GEOTHERMAL ANOMALY OF TUSCANY/ITALY
Sunmary
Magnetotelluric measurements have been carried out in the geothermal
area of Tuscany in order to determine the distribution of the electrical
resistivity within the crust. Herefrom the geological structure and the
distribution of temperature will be inferred.
The here presented results indicate clearly a rather high total con-
ductance of the surface layers up to 1000 S which can vary considerably
at some places. A geological interpretation points to a very high con-
ducting layer in depths down to 10 km overlain by a higher resistive
material. This deep, highly conducting layer and the transition to the
underlain high resistive basement may correlate with the deep reflection
horizons detected by reflection and refraction seismics.
1. I ntroducti on
The aim of this magnetotelluric project is to determine the distri-
bution of the electrical conductivity of the crust in Tuscany by the
magnetotelluric method. The measurements of the time-varying electric
and magnetic fields have been performed in October 1978 and in the
main field campaign in autumn 1979. In figure 1 the position of all
measuring sites are represented wherefrom a part could already be pro-
cessed. One part of the measurements were made along a profile which
coincides with the refrac~ion seismic profile close to Larderello
(GIESE et al., 1980), the other part of the measurements were made
at selected sites (e.g. Monte Amiata: MTA). The geomagnetic activity
during the main field campaign was moderate, recording times of 1 to
2 weeks at each site were necessary in order to obtain sufficient
material for interpretation.In figure 2 some recording samples are
displayed in order to show the quality of the obtained data. The two
components of the electric field and the three components of the magnetic
field were recorded digitally on tape.
- 843-
MAGNETOTELLURIC· and 0 , GDS .1 ••
GEOMAGNETIC DEPTH· 0: MT-meosudng Ilh, 8. (ISs) - 3000,
SOUNDING . : MT-m.e.lurlng lit., 151 - 6h
1.2
+
10° 12 .Roma 13" 11.°
-844 -
0 81
Z8 1~__~~~~--~__--~__~--__~~~
MT.MURLO (MUR) 14 UT 15
20 UT 21
MANORIACE (MAN) 19/10/1979
1rnV/lcrn
NS 6 -"'"-'''''
Ew 61
H 61nT
0 61
Z6 1 .~ ________ ~__~________~__----__~__~~------~____~
- 845-
0<===-_ _
50 km
- 846-
3. Apparent resistivities and resistivity models
The apparent resistivies as function of period are represented in
figure 4 and 5. Some of them are of rather high quality and permit
precise interpretation. Nevertheless. all of them start at low resistivity
at low periods , increasing towards longer periods. Therefore. the same
type of resistivity models can be expected.
The true resistivity as function of depth have been obtained by an
inversion algorithm (SCHMUCKER. 1974). which also permits to transform
the measuring error into the corresponding variability of the model
parameter. The resistivity models are given in figure 4 and 5. The va-
riability of the model parameters (resistivities and depths) permits
to discern dominant and non dominant features of a model: In MUR e.g.
(see fig. 4) the existence of a conducting layer in the deeper crust
is a dominant feature, whilst the depth of it (varying between 20 and
35 km) is obviously not dominant.
All resistivity models are summarized as bloc diagrams in figure 6.
The most prominent aspect of all these models is the very thick and
very conductive surface layer. The period range from 10 s upwards does
not permit to resolve a fine structure within this thick conductive
surface layer. Nevertheless. two quantities are determined quantita-
tively: The total conductance (= the total thickness of the conductive
layer divided by the resistivities of the layers) and the depth of the
transition from the high conductive layers to the high resistive base-
ment. Therefore, the thick conductive layer at each site may be split
up in a number of layers of higher and lower resistivity arbitrarily
as long as the above mentioned two quantities are not changed. One
possible equivalent model could be built up by a low resistive layer at
the surface proper. a more resistive layer. in an intermediate depth
and a very conductive layer at the base of the thick conductive layers
in figure 6. In order to "measure"such a resistive layer, much higher
frequencies must be measured. Nevertheless, to determine the depth and
the existence of the deep and conductive layer, periods up to several
hundreds of seconds must be measured.
-847 -
103 100 101 10 2 flm 10 3 ~-
0,1
t~a
MUR MUR
r---.J
II I
III
111-Z·:1OS
10
-g.,
.J::.
I I-z·
'0
,00 2000s
10 T- ,0' '00 I 1 I
km I I I
103
102
.!lm
10
~c- VIL
-g.,
.J::.
101 '0
'00
km I
T_ ,0'
103
'I 'I
ha
MOL
100 10' 10 2 flm '0 3 ~-
1
102 -
-g.,
.J::.
MOL
•
.!lm
'0
/
'0
10' - km
,I ,I
,00'0" lOL '03 S T- ,0'
-850-
~--~~--~------~~ o
km
10
10
- 851 -
The well conducting layer below MUR at a depth range between 20 and
35 km could have the same explanation as the low velocity layer deter-
mi ned by refracti on sei smi cs, i. e. by temperatures close to the soli dus
temperature of a more sialic material.
It must also be mentioned that in a previous magnetotelluric study
by DUPIS et al . (1974) in the Larderello area a high conductive layer
in 10 km depth has been found, close to the measuring sites of the present
study .
Taking into account the known geology and the present new results
inferred from magnetote 11 uri c and sei smi c studi es an ideal i zed geo 1ogi c
cross section of the crust of Tuscany may be constructed as it is shown
in figure 7 (GIESE et al., 1980).
l~~"Tf~'~~
• nOChlhOnO"!6 nappe
... 'C
Irfl~II!1:!;.:.R~?!fl~~~!tt~*i.iiii~~*Iii;}§.~*-*-;4~\
~: A schematic and idealized cross-section through the upper
crust of Tuscany. The upper half summarizes the dominant
features of the known geology, the lower half summarizes
the magnetotelluric and seismic results in terms of a
possible geological interpretation. The essential and new
aspect of this model is the deep, intercalated layer upon
the granitic/gneissic basement and below the phyllitic/
quartzitic allochthonous nappe. This deep triassic/palaeo-
zoic layer/zone should contain according to the high elec-
trical conductivity a high amount of hot, saline water.
- 852-
6. Conclusions
Most promissing for new geothermal energy in Tuscany seems to be the
high conductiv~ layer or zone in great depth below the higher resistive
"intermediate basement", which most probably is caused by water bea-
ring material at high temperatures. In order to be able to exploit
such deep buried layers it is necessary to know the morphology of it
more precisely. Such precise information can be given without doubt
by the seismic methods. Nevertheless the magnetotelluric method could
yield in a first, inexpensive stage more general information of the
morphology of this conducting deep layer followed up by the more expen-
sive seismic studies in the second stage for more detailed information
in areas of interest.
In order to obtain the full information from the surface down to the
transition to the high resistive basement periods from 0.001 s up to
several hundreds of seconds should be measured.
Acknowledgements
We want to thank Prof. C. Morelli, Trieste, for his help to organize
the field measurements. We also thank Prof. A. Norinelli and Dr. S.·
Spitz for the cooporation during the fieldwork. The field measurements
have only be possible with the intensive help and cooporation with the
"Magnetotelluric working group" (Dr. M. Beblo, Dr. A Berktold, P.
Neurieder and P. Wolfgram) from the Institut fUr Allgemeine und Ange-
wandte Geophysik der Ludwig-Maximilians-Universitat MUnchen. We feel
to be very indebted for their cooporation.
References:
DUPIS, A., V. ILICETO and A. NORINELLI: First magnetotelluric measure-
ments on Larderello site.- Boll. Geof. Teor. Appl., 16,137 -152,
1974. -
GIESE, P., P. WIGGER, C. MORELLI and R. NICOLICH: Seismic studies for
the determination of the crustal structure in the area of the geo-
thermal anomaly in Tuscany.- This volume, 1980.
BATINI, F.,P.D.BURGASSI. G. M. CAMELI, R. NICOLICH and P. SQUARCI:
Contribution to the study of deep 1ithospheric profiles: "deep"
reflecting horizons in Larderello-Travale geothermal field. -
Mem. Soc. Geol. I tal. , 1978.
SCHMUCKER, U.: Erdmagnetische Tiefensondierung mit langperiodischen
Variationen.- Prot. Koll. Erdmagn. Tiefensondierung. Grafrath,
313 - 343, 1974.
- 853-
THREE-DIMENSIONAL RESISTIVITY MODELLING BY THE INTEGRAL EQUATION
METHOD
ABSTRACT
- 854-
INTRODUCTION
Integral equation have been used for solving the 2-D and 3-D
problem of an inhomogeneity buried in an homogeneous material
(Dieter et al., 1969 ; Hohman, 1971, 1975 ; Parry and ward, 1971).
More recently, Snyder (1976) published an interesting solution for
several 2-D inhomogeneities embedded in a homogeneous medium. This
method can be modified for a two-layered medium. Lee (1975) gave
an integro-differential equation for a cylindrical and a spherical
inhomogeneity buried in a two-layered medium. Weidelt (1975) published
an integral formulation for the electromagnetic induction in a 3-D
structure and gave an algorithm for the calculation of Green's function
in a layered medium.
- 855-
Our purpose in this paper is to solve the problem for prismatic
inhomogeneities embedded in a medium composed of an arbitrary number
of homogeneous horizontal layers (Fig. 1). Such a choice is a realistic
approximation for an arbitrarily-shaped inhomogeneity which can always be
broken down into several component prismatic bodies. Moreover the bounding
surface of a prismatic body can be easily decomposed into rectangular
cells.
THEORY
All the mathematical notation used in this section are defined in the
list of sl'l!lbols.
- 856-
div(o
...
grad,) = 0, (3)
n n
except at the sources. The difference,
(4)
...
diV(On grad. 'a) + div(oa grad ,)
... o. (5 )
(6)
(7)
-8S7 -
Using Green's theorem (Morse and Feshbach, 1953, p. 803), the
volume integral of the left hand side can be transformed in a surface
integral which vanishes since ~a and Gn behave like 1/r at infinity.
Therefore we obtain
(8)
(10)
- 858-
-+-
- When P is outside 5, 0a gradG n (P,PO) is a continuous functior
of P inside the inhomogeneity, is null outside since 0a is null, and is
discontinuous through the surface 5. Using distribution derivation
properties (Arsac, 1961, p. 122), we write
Since a
a is constant inside the inhomogeneity and zero outside,
"l"t:is easy to show that div(Oa gr!d Gn ) is equal to zero over D.
So we get
(12)
If5
-+-
- ~ is inside 5, 0a grad Gn is not a continuous function of
POinside 5 but has a pole at P in the vicinity of which Gn behaves
like 1/(4no nPPo). In order to remove this singularity we define
-+- -+-
of:
t = aa grad Gn - X grad(l/PPo) (13)
a
with X= ~ 4n
a
n
-+-
Then f is a continuous vector inside and outside 5 but
discontinuous at the surface 5. We get
- 859-
Since aa and X are constant inside D,
div 1= a 6 G - X6 (_1_)
a n PP o
a
Iff $ div f dv fff - / Il(P,po) $ dv -
D n
D D
(15)
Using (10) I (13) and (14) the new equation for ~a (PI is
(17)
G ~-L
n ern 4Tf PPo
- 860-
The integration over the disk is then
fJ [
A
:a:l+
n--
o~~dS
nan
aG
[:: I ~ (p ) 4~ I~ aan (p,~o) dS (lS)
II ~ (__
A n
1_) dS is the solid angle from which the disk is seen from P •
PPo
~ a (P ) = II [Oa
°
J+ °nan
aGn ~ dS +12 (19)
s_{pn} -
Therefore three diffex:ent expressions of <P-a (1)) (U2), (7) I (9»)
are obtained depending on the position of P relative to the inhomogeneity.
Since ° a (P) is discontinuous accross S we use the convention that
1
0a{P) = 2 (oh - on) when P belongs to S. Then the three expressions
«12),(17),(19» reduce to the following equation, where ~a(P) is
replaced by ~(P) - ~n(P)
cP (P ) = ~n (P ) + ff
S
cP (~) a r:
r.L n
a ] +
-
:G~ dS (20)
- 861-
The general expression of the integral equation,applicable for
any position of the inhomogeneity, is
aa (P ) + an (P )
~(P ) =~ (P ) + IIS ~(P) [a~ 1+ (22)
an(p ) n an -
NUMERICAL RESULTS
E
I (38)
- 863-
The maximum value of t:. is about 40 percent and the position of this maximum i
shifted relative to the center of the horizontal projection of the
inhomogeneity.
-864 -
The decrease of 6Max according to the overburden conductance shows the scree-
ning effect due to this conductance, which may be explained by a
channeling of lines of current lessening the depth of penetration.
The models considered are shown in Fig. 8. The upper surface of the inho-
mogeneity appears at the ground surface. The resistivity Pn of the inhomogeneity
is about 30 n m, its thickness e is about 30 m,and its horizontal dimensions
are 100 x 500 m. Schlumberger soundin~were centered on the inhomogeneity
perpendicularly to its largest dimension.
- 865-
In the numerical model study, it was not possible to locate the
current electrodes at the inhomogeneity surface, because the very sharp
potential~ariation near the electrodes is not compatible with the
approximation of function $n by a a constant on each surface element
t. j • So we have determined the apparent resistivity only for a line
length AB/2 > 50 m.
- 866-
Such a model cannot easily be simulated in electrolytic tank,
whereas numerical simulation is straighforward. In figure 11, it can
be seen that the curve nO 2 (numerical model) fits quite well the
curve nO 1 (measured on the ground). Both curves nO 1 and 2 show a
flatening for large AB/2 toward Pa = 150 nm whereas the resistivity
of the assumed substratum is 420 nm ; this resistivity departure confirm
the importance of the perturbation of a syncline, even when its dimension
are small compared to AB/2, and when the resistivity contrast between
the syncline and the first layer is not important (in the present
case, the ratio of conductivities is about 1/8).
CONCLUSION
ACKNOWLEDGMENTS
- 867-
REFERENCES
LEE, T., 1975 - An integral equation and its solution for some two
and three-dimensional problems in resistivity and induced
polarization. Geophys. J.R. astr. Soc., v. 42, p. 81-95.
SNYDER, D.O., 1976, A method for modeling the resistivity and IP response
of two-dimensional bodies. Geophysics, v. 41, p. 997-1015.
- 868-
LIST OF SYMBOLS
Model hI PI e ZI X X Y Y
nC: (m) (n m) (rt~) (km) (km) (k~) (kM) (~) (iJ\) llMax(%)
1 100 100 1 2 8 6 4 -4 -4 15
2 100 100 1 2 4 6 4 -4 -4 43
3 400 100 2 8 6 4 -4 -4 7
4 100 100 100 2 4 6 4 -4 -4 30
5 100 100 1 2 4 5 5 -5 -3 41
6 0 0 1 2 8 6 4 -4 -4 23
7 1000 100 1 2 8 6 4 -4 -4 4
8 100 100 1 5 8 6 4 -4 -4 15
g 100 100 1 2 8 7,9 2 -6,4 -5,6 14
10 100 100 1 2 8 8 2 -12 -12 14
- 869-
x
.: .: .: _-.: =;A - - - - - - - ? B
,,~:/d
" ,
level
,
/',
,"
O,~-------L--------~------------------~
YA y.
__ y
a;
h,
(1"2
h2
crp
hn
Un ... '
x
Fig. 1 - Conductivity model
considered in this paper
I
, I
,
x , ,,
- - 1 0 k t n - -_
-F ig. 2 - Parameters of the
conductive inhomogeneity buried
in a two-l~er medium
- 870-
ex>
-.l
Fig. 4 - Apparent resistivity contour map for a Fig. 5 - Apparent resistivity contour map for the
two-l~er model: hl = 100 mj Pl = 100Q.m same model as fig. 4, with an inhomogeneity at a
(without inhomogene1ty) depth of 4 km.
Ai A_ 2Km _s
e
w
81
,«'(<<4 / / A x
00
........
I
Fig. 6 - Apparent resistivity contour map with inhemegeneity (left part) and without inhomogeneity (right part)
a - dipole parallel to Oy model nO 5 b - dipole parallel to Ox model nO 2
Am.& !
~ e·
A·
P. • j
":8 \ i
20-\
Ir,
\'\ 6 -
15- " '" B P, ~ ~~
P,
I / Ph ~ I o
00· '" J
'~~--~_ e .7 ~ 1;1'.
s- ~~
.d ~
I -------------1 , :;-:::;
o • 00
o
o
I CoCondo."n ••
. mho.
- AB
Fig. 7 - Variation of maximum effect as a function of the Ph - 2-
e 10e
conductance hl/PI of the sedimentary cover for an inhomo- P.. r-
geneity buried a~ a depth of 8 km. The position of the j
o current electrodes with respect to the inhomogeneity is r, " r3
~ shown in the upper right corner. / / 11
o
V ~
/ .d
~ I /Ph -';
P, ~
r- , ~
/'
r
[7.' 1
I AB
Ph 2"'
10e -
B
,,_3
3 V
2
._12
100 1/ lP
10'
'''lllll llltl11IJTIIIIj.f
30
..-V 301 ~~
-- 111111 I III 111111
"".....
...
10' II! III I ~
10
10 100 1000 '"
2 10 100 1000 2
~~, .....-li....
[!J 2 3
l!J .. ~ 'OnoM r. lO"".m 15M
~
~lOM lW~30M li!Q!!iOJ
r• 2OOA;II, '" r-200Aam
170 Om
W 25m ,,0.
1700"",
------170 . . no",
.+. ,.- f- .. 1100
420 O m
Om
11&",
85m ~
420 Om
Fig. 10 - Comparison of the apparent reBistivity of Fig. 11 - Comparison between experimental measurements
laboratory I numerical and 3-l..,-ered models and numerical model response
MAGNETIC SURVEY IN THE TRAVALE GEOTHERMAL FIELD, ITALY
Summary
The detailed magnetic survey has covered the area between the villages of
Boccheggiano, Travale and Radicondoli, where supposedly is confined the
most active geothermal field. Along with the main structural setting of
the region, the magnetic anomalies (ranging in the order of 100 gammas)
seem to depict the major trend of the westerly edge of the Era-graben and
the associated fault system across it, where the present geothermal
activity has taken place. Changes in the rock magnetic properties for the
crustal layer affected by thermal processes suggest an explanation for the
local anomaly closely related to the geothermal area.
- 875-
1. INTRODUCTION
-876 -
The Travale geothermal field is located in an area of the Apennines in
Tuscany where similar rocks (massive to pelagic limestones, arenaceous to
clayley little consolidated sediments, etc.) are responsible for major
accumulation and preservation of the geothermal reservoirs within the
sedimentary layer and for the cristalline basement. With increasing
temperatures, remanent magnetizations and susceptilities tend to vanish in
the whole rock body, thus accordingly affecting the geomagnet i c field
distribution on land .
36~. s. 5 371.0.1
5.·1 5 • •J
.5 .5
It · C) n'e)
- 877-
The present work has been devoted to verify the dependence of surface
magnetic intensities on thermally anomalous areas where otherwise the
geomagnetic field would have been quite "normal" over sensible extensions.
The Travale region shows adjacent portions of very active geothermality
and its almost complete absence, and thus most convenient conditions for a
field survey on rock formations of faible remanences for the mentioned
point of view.
3. MAGNETIC SURVEY
Following the scheme already used for the detailed profile across the
region (Napoleone, 1974), the vertical and horizontal components of the
geothermal field were measured by means of field variometers in station
points at 300 m distance over an area extending about 7 x 15 km. Each
reported value has an accuracy within! 3 gammas and represents the
average of a central station and 6 surrounding stations at a distance of
few tens of meters, in order to check the local normality; in fact the
area is intensely affected in places by industrial manufacts around the
exploited geothermal wells especially metallic pipelines not easily
visible.
-878-
The survey started in early spring of 1978, by successively defining the
area through several North-South and cross profiles, but the major portion
was accomplished during winter 1978-1979, due to better vegetation condi-
tions in the mediterranean "maquis" typically covering most of the area.
Almost 1 000 of multiple points were surveyed and corrected to the absolute
values of the nearby reference base line of the Italian 1st order magnetic
network at Mt. Fili. For both components the station absolute values are
referred to the 1979.0. The correction for the daily variations has been
made by means of local recordings and reference values from the National
Geomagnetic Observatory at Aquila; the normal geomagnetic field for
Northern Apennines was computed by Battelli (1979) of the National Geomagnet-
ic Observatory.
The amplitudes of the magnetic anomaly range in the order of 100 gammas,
except the area of Mt, Gabbro, where gabbros and serpentines outcropping
in the allochthonous ophiolityic body produce field alterations of several
hundreds of gammas; such anomalies display a very localized distribution
and sharp alternance of steep gradients, and therefore have been filtered
out by both procedures hand-smoothing the contour lines and computerizing
all recordings for numerical analysis.
4. MAGNETIC ANOMALIES
All corrected values, averaging the 7 measurements sites for each station,
have been plotted on the maps shown in figures 2 and 3, with a contour
line equidistance of 10 gammas. Two major features can be noted : a
general trend NW-SE of the central positive anomaly that possibly extends
outside the surveyed area, and a localized negative anomaly just above the
active geothermal field.
Minor anomalies but exceeding the measurements noise, may infer either
local displacements of tectonic structures or superimposed thermal activity
that affects rock magnetic properties of the epi-superficial sedimentary
layer and therefore anomalously contributing to the measured values of the
geomagnetiC field (given the inducing field has natural remanences in
addition)
- 879-
TRAYALE GEOTHERMAL fiELD HD"lUI)I"l TRAYALE GEOTHERMAL fiELD '(lTlU,
UU(lH IIUU'" uutlle IITUSIJ't
USllll1lll U'
I
00
gg
Figure 2 - Residual map for the horizontal Figure 3 - Residual map for the vertical
components of the earth's magnetic field. component, as in figure 2.
Contour intervals are 10 gammas.
An example of tridimensional plot is given in figure 4, which synthetizes
the previous maps and makes more evident the mentioned weak anomalies.
Just North of the major negative, in the little magnetic depression a new
productive well has independently been located and drilled a few months
ago.
- 881-
1 -~;:::--
.
f- •
o "
w 3
~
L
W "
_J '-
<: 0
:;:,
<:
•
((
-882-
REFERENCES
Battelli, O. ,1979. The normal geomagnetic field in Northern Apennines. Int. Tech.
Rep., National Geomagnetic Observatory. Unpublished.
Dunlop,D. ,1972. Magnetic mineralogy of unheated and heated red sediments by
cohercivity spectrum analysis. Geoph. J. R. astr. Soc.,27,37-55.
ENEL, 1963. Magnetic survey of the active geothermal field in Tuscany. Int. Tech.
Rep.,R.D. Div., pp. 54.
Heller, F ., Market, H. , Schmidbauer, 1979. Partial self-reversal of natural rema!!
ent magnetization of an historical lava flow of Mt. Etna (Sicily). J •Geophys. ,
45,235-257.
lliceto, V. , Napoleone,G., Pinna, E., 1973. Geomagnetic changes during the
cooling process of the Mt. Etna 1971 lava flow. Riv. Geof. Ital., 22, 89-94-
Nagata, T . , Uyeda, S. , Akimoto, S., 1952. Self-reversal of thermoremanent
magnetism of igneous rocks. J •Geomag.Geoelect. , 4,22-38.
Napoleone, G., 1974. Detailed magnetic profile across the Travale geothermal
area. Int. Tech. Rep., Geol. Dpt., Univ. Florence. Unpublished.
Napoleone, G., Poggiali, G. , Ripepe ,M. ,Savino, D. ,1979. Geomagnetic changes
produced by the cooling of the Mt. Etna lava flow in last august. Eur. Geoph.
Soc., 7th Annual Meet., Vienna 11-14 sept. 1979.
Napoleone , G. , Premoli Silva, I. , Heller, F . ,Cheli, P. ,Corezzi, S. ,Fisher, A. G. ,
1980. Eocene magnetic stratigraphy at Gubbio, Italy. Geol. Soc. Amer. Bull. ,
in the press.
Neel ,L. ,1949. Theorie du traThage magnetique des ferromagnetiques aux grains
fins avec applications aux terres cuites. Ann. Geophys. ,5, 99-136.
Roggenthen, W. M., and Napoleone, G., 1977. Upper Cretaceous-Paleocene magnetic
stratigraphy at Gubbio, Italy. I V . Upper Maastrichtian-Paleocene magnetic
stratigraphy. Geol. Soc .Amer. Bull., 88, 383-388.
Thellier,E. ,and Thellier,O., 1959. Sur l'intensite du champ magnetique terrestre
dans Ie passe historique et geologique. Ann. Geophys., 15, 285-376.
- 883-
WE OF THE DIFFERENTIAL MAGNETIC SOUNDING PUR STUDYING THE
GEOTHERMAL ParmIAL RESOURCES OF THE RHlNEGRABEH
J. MOSlIER - K. BABOUR
Laboratoire de Geomagnetisme
- 884-
COJlSideriDg that the hotest layers in the orust are likBly to
be also the ... t oonductive, it is pe88ibla to sustitute to the direct pros-
peotion of geothel"Ml 8J'01118l:1es sa indireot prospection of ~ conductive
areas whicll are like17 to Ie resenoire. In this aia several geophysical 118-
- 885-
F1Dally, a third one is located at a few ldlometers in a
"no1'll8.l" area where only weak telluric currents flov •
Let 11 be the intensity of the currents flowing between M and
Mi, let 12 be that of the cmrrents flniIlg under M'. The I18.gnetometer M measures
a field BM 8uob as :
(2)
and the last two one are the components of ll. BT.lt can be seen that
- 886-
a) - The two anomalous fields H aDd H are linearly ptllarised
1 a
according to the same direction (roughly ,o0/_gIletic North).
D) - Hl and ~ have the same law of temporal variation.
c) - The amplitude of A lis and A H.r are roughly the S8IIIe which
implies l1. # ~ ·
The depth of M' in the previous experiment was 650 m. The
electrical logs performed by the M.D.P.A. Society allow one to calculate the
total conductance of the layers comprised between 0 and 650 a. Froa the same
kind of data obtained by oil companies in this area, it has been possible to
calculate the conductance of the layers comprised between 650 a aDd the base-
ment (the depth of which is supposed to be at 2500 a in the area of Mulhouse).
The results of these computations show that theae two coDduc-
tances are of the same order of magnitude. (D. de FERRlERE 1978). It can .e
ded1lCed that the total intensity (11 + ~ ) of the currents floviJlg in the
sedimentary filling of the graben is sufficient to explain the whole observed
anGIIalous field. Besides, for further interpretations, i t will be possible t.
oonsider the source as a sheet of pseudo direct currents flowing between the
surface and the resistive basement.
The results of the above described experiaent have iamediately
been used to organize another one I aore directly liDked. to the geothel'lQ' in
the RhilIegraben. Iudeed, i f we consider that the whole aJ1OIII8.lous field obser-
ved in this area is created by a sheet of pseudo-direct currents fl.wiJlc in
the sediments, then the areas baving a large geothel'llal flux aDd supposed to
be hi&:hlY conductive,lRUSt locally disturb the current distribution aDd oonse-
quently the anomalous field distribution.
Ve tried to point out this pheDGllleD8ll in a region vell-kllon.
for its geothermal activities, the area of Soult... sous-Forlts. Te this pur-
pose we per.foraed a differential geoaagnetic sounding including 9 stations
settled along an E.S.E., V.N.V. (magnetic) profile, creasing the geothermal
&DOIII8.ly (Figure 4). The l18e.nreaenta have been foUDd to be difficult, fer
this area is fer from the maxiJauJI of the regional 8.DOII8.ly aDd cODSequentl1'
the '98riatioDS of the 9JlOIIalous field aodulus are 'Y81'Y weak. Besides, the
natural III8.grl8tic activity was weak aDd the iDdustrial interference noise
vas large.
-887 -
levertheless the obtained resUlts are clear enough, as oan be
inferred from figure 5. The upper curve represents the repartition of the
anomalous field alODg the profile. It olear~ shows the regional anom~
whioh inoreases regular~ from the station of REICHSOFFEN (REI) to the
station of HA'l"l'EN (HAT), then further on towards SELTZ. SUperposed on it,
there is a looal anom~ that can be pointed out uBiDg small station-spa-
cing, between PEeBELBROIN (PEe) and KUHLENDORF (KUL). This local anom~
main~ consists in an increase of the regional anomaly near Kutzenhsusen,
followed b,y a decrease of the anomalous field near Soultz. It is not easy
to give an interpretation of this anom~ for one must not forget that the
anomalous field depends on all the currents that flow in the sedimentary
l~ers between the surface and the basement. The geological section aloDg
the profile (Figure 5 at the bottom) clear~ shows, between Pechelbronn and
Soultz, an uplift of the basement which perhaps can account for the decrease
of the anomalous field at station SOU. The increase of H east of Soultz
a
would be due main~ to the increase of the thickness of the sedimentary
l~ers while the maximum observed to the west could be due to an increase
- 888-
12-
- 889-
....
ID
-" C\j
0 0
~ C\j
0 W
0- Z
::>
0-
0 -,
CD
.., ~
~ 5!! 5!!
- 890-
I
§.
,
I
I
/
I
, ,'/........--
"
- 891-
Amplltudt dt /0 compo,ontt 03, du chomp ono'IIIol
normoli st par rapport 0 (03 ')R M W
1,0
----
jT\
0,5
ESE
+ ~«U<UU/~ffi' .. ,~////,//
. . ~« ~U/_S
2000 +
+
Abstract
With a combination of instruments to measure electromag-
netic field variations in the period range from 6s to 24h,
and appropriate station coverage of the region comprising
Rheingraben, Schwarzwald and Swabian Alb, it should be pos-
sible to obtain data for improved conductivity models of the
Rheingraben and Urach geothermal anomalies. Our surveys in
1977/78 were designed with this objective, and we have some
results for selected effects to present. The perturbation
of magnetic field near the Rheingraben indicates the presence
of anomalous currents. The effect is frequency dependent,
the higher frequency results reflecting the graben sedimentary
structure and the lower frequencies a possibly deeper seated
anomaly. Magnetotelluric soundings using pulsations also
show this contrast in shallow structure between the graben
and Schwarzwald. At longer periods a deeper structure in the
graben is indicated. Near the Urach volcanic area there is
also a strong magnetic field perturbation.
- 893-
1. Introduction
As part of a study of the application of magnetotelluric
(MT) and magnetovariational (MV) techniques to geothermal
areas two east-west profiles were made across the Schwarzwald
in 1977 and 1978. The westernmost station was in the Rhein-
graben and the easternmost in the Swabian Alb, as shown in
Fig. 1. Instrumentation included induction coil magneto-
meters with response suitable to measure magnetic pulsations,
fluxgate and Askania magnetometers for bay-type variations
and telluric apparatus, using Ag-AgCI electrodes, suitable
for the entire frequency range. The regu:ar period ranges
of the apparatus, 8 to 1000 s for pulsations and 5 to 200 min
for bays, has been somewhat extended by analysis of long seg-
ments of pulsations or bay recordings during times of pro-
longed magnetic activity. The equipment used at each of the
stations to be discussed is listed in Table I with information
about the period ranges used in the analysis.
The analysis has been divided roughly along two lines, the
distortion of magnetic field variations as a function of fre-
quency at the survey stations relative to some reference, and
the correlation of electric with magnetic variations to esti-
mate impedance as a function of frequency and obtain a first
estimate of local conductivity distribution with depth. As
the magnetiC variation results have been discussed earlier
(1,2) and are further described in the project final report,
we discuss here the magnetotelluric results. Some of these
have not been previously reported.
2. Review of magnetovariational results
It is, however, appropriate here to repeat the essential
findings of the correlation of anomalous magnetic field varia-
tions. A reference station was selected (ENZ or HTZ) and the
difference fields compared with the reference. The analysis
followed the method to be outlined below for magnetotelluric
analysis (see (2 )). The results parallel closely what is ob-
served for individual bays: the EW-component (D) is strongly
-~-
enhanced in and near the Rheingraben, and the vertical compo-
nent (Z) is enhanced at the edge. The anomalous D variations
3. Magnetotelluric soundings
- 895-
for SIG analysis because several longer effects where polar
substorm variations lasted for some hours could be used. To
span the gap at higher frequencies the SIG pulsation effects
were selected when long period electric variations were also
observed; the results merge satisfactorily with the highest
frequency bay results. For ENZ, NBU, TIE and HTZ the pulsation
data were first filtered and decimated so that long period
effects could be selected. At the northern stations this pro-
vides lower frequency results with consequent greater depth
of penetration. However, the patch between bay and pulsation
frequencies is not entirely satisfactory at HTZ, probably be-
cause of insufficient signal at the higher bay frequencies.
,..
WN+
+
9"
I
I
"-
I
I "- ,,
.
I
\
I STU'lGAAT
/
I
/ E~Z /
/
"
NBUo ~T
/
/
/
/
/ ";EUFFEN
/ oENT / x .,KOH
/
/
,," WT~~ XBOE ~
<{ / /" OHNx ,.. \.
~ 0 I / KBGx
'" TIE I /
,
Q; /
I
<( /
~ /
:t: I
/
0
I
'" LFG,
I
x
I
MULHOUs£'
.
-896 -
At the station south of Stuttgart the results are a compo-
site of pulsations at ENT and bays at ALT. The two locations
are 6 km apart. The lower frequency portion can be used for
either location, but this is probably not true for higher fre-
quencies.
3.1 Analysis
Following selection of data segments the effects are fil-
tered, decimated and transformed to frequency spectrum esti-
mates as described by Schmucker (2). One variation from that
method is the premultiplication of pulsation data by a cosine
waveform. A linear relationship between electric field compo-
nent and horizontal magnetic components is assumed, and the
transfer function relation
H
(1)
D
- 897-
'valent depth of penetration, and p*(i*) provides a first in-
terpretation of impedance. All the parameters are functions
of frequency, and for Z=U+iV,
P~
I = _2_ u2 h = (V-U) ,
]JOW lloW
u 2+v 2 2
*' 1
PI! = 2110W
U-V
(-V-) , T = 2 2
(2)
U +V
z
-It-
..:L
lloW
Figure 2 shows the results for northern and southern profile
stations. The upper panels of 2a and 2b refer to ZNE and the
lower to ZEN. Errors are shown schematically on the diagrams
by lines bounding the curves. In NS/D polarization the resis-
tivity indicated for Schwarzwald stations is 100-200 times
larger than for SIG or RMW. This is emphasised in Fig.3 where
the curves are superimposed and errors omitted. Falling bet-
ween the two extremes are the southern Rheingraben station
MHS and ENT/ALT near Stuttgart. With the exception again of
these two, where data are inadequate, the curves show a devia-
tion toward lower resistivity at a depth of 30 to 50 km,
followed by an increase. Most stations show a decrease again
at greater depth, and, except at SIG, an increase at shallo-
wer depth. A further superposition of curves for the two
polarizations would reveal that at HTZ and ENZ the impedance
is practically isotropic. BAD and NBU show, to varying de-
grees, the influence of lateral variation of conductivity,
and the electric field in the Rheingraben is too strongly
distorted to use the EW/H polarization to calculate p* (z*) •
The depth of penetration is very different for these areas.
At ENZ, BAD and NBU/Schwarzwald where only pulsations were
used, the depth of penetration is greater than at the two
Rheingraben stations where bays were also included. At HTZ
tne full frequency range is available, and the maximum depth
is attained. On the other hand for the Schwarzwald stations
there is no information for the upper 20 to 40 km.
- 898-
logplllml
,-,
------.
o 10
· 1
2
I Z'
I~I 3
-, ~-----------------------------,.---~------------------------~---.
logp'Il1m2
-'
0 ·· 20.
200.
• 2000.
, 0
,. .,
O.
3
L-__ ~ __ ~ __ ~~ __ ~ __ ~=-~ ____ __ __- L__
~ ~ ~ ____ ~ __ ~ __ ~ __- L_ _ _ 3
~-'
-~ I
log p'lllml
o
• 20 •
.2000.
• 200s
2
logZ'
Ikml J L-__~__~__~____~__~__~__~__~ru ___~__~___
log p'IOml
-,
o • 0
b.
- 899-
o ~=---~---------------,,-----------~
NS/D
1
•I
10g:2:
[km] {.
2 logp[~mJ3
5
..
o ~~~----------------------------------~
EW/H
. . .. ..,.,
..
l09Z !
[kmJ
o 2 • 3 5
l og p[~m J
~
-900 -
p*( Qm)
11 10 100
150 -f120
~~ 1'5
f
~
~
t" ~ 9
B+-f=+s
+
10
f
pulsations -+-'
z* ~
+2.5
+
(km)
bays --¥o
~5
+ longer effec ts 100
--f2.75
100
TO.25cph
.15
Station SIG
(Rheingraben)
1000 L-----------------------------------~
-901 -
unlikely the curve can be extended to lower frequency/greater
depth.
Applying Swift's (4) criterion to orient the impedance for
two-dimensional structures, the SIG data should be displayed
in a coordinate system aligned with the graben, and RMW coor-
dinates should be rotated to N20 oW. Figure 4 shows p*(z*) for
SIG in the polarization with electric field component along
the graben and magnetic component perpendicular. In this dia-
gram errors are shown, and also frequency for some values
along the curve. The rotated values for SIG and RMW are used
in the next section where they are interpreted by a one-dimen-
sional model.
3.3 One-dimensional models
Under certain assumptions the induction equations for a
structure that consists of a number of uniform layers can be
linearised to a form that allows model parameters, layer con-
ductivity times (thickness) 2, to be estimated with allowance
made for errors. The method of Schmucker (5) has been used
to examine p* (z~) for SIG, RMW and HTZ from the point of view
of such models. The simplest realistic model is two layers
over halfspace. Increasing the number to three makes a small
improvement, and this model is shown in Fig.5. RMW and SIG
models are basically the same, although the layer above the
halfspace is relatively resistive for SIG and conductive for
RMW. The thin layer for RMW is unresolved when errors are con-
sidered. The model for HTZ is much more resistive. There is an
indication of a good conductor again, however, at a rather
large depth.
4. Concluding remarks
Some magnetotelluric depth sounding results for the Rhein-
graben and neighboring areas have been presented. There are
indications of a better conductor in the lower crust from the
-902-
log p [nm]-
O°r-~____~____~2r-______3.-____-,4
1
logZ
[km]
l
2
3 ~------~--------~--~--~~------~
-903 -
MAGNETOTELLURIC SURVEY OF THE LAKE LAACH (EIFEL)
VOLCANIC AREA
Abstract
In 1978 and 1979 a total of 9 geoelectric Schlumberger soundings
and 14 magnetotelluric soundings were carried out along a 10 km profile
across Lake Laach in the Eifel volcanic area.
It is the aim to detect the proposed existence of a magma chamber in a
depth of a few km which may cause a conductivity low.
The evaluation of the geoelectric sounding curves is completed and shows
a clear differentiated stratified subsoil in lateral direction down to
a depth of about 300 m.
The upper resistivities are about 300 om around Lake Laach and only 50 om
below the lake. These values will be fed into the later model calculations
for MT evaluation.
The interpretation of the Pa-curves from the MT-recordings is fairly
complicated due to the high noise level allover the area and due to short
distance lateral inhomogeneities.
We observed strong differences between the two corresponding orthogonal
P -curves. The final interpretation of the shown pseudo-cross-section
nteds three dimensional model calculations which is under investigation.
-~-
1. INTRODUCTION
In 1978 and 1979 a total of 9 geoelectric Schlumberger soundings with
electrode distances up to 1.6 km and 14 magnetotelluric soundings in the
period range between 0.2 sec (4 sec respectively, depending on station)
and 1000 sec were carried out along a 10 km (NW-SE) profile across Lake
Laach in the Eifel volcanic area.
The results of the upper 5 m are not shown due to scale. In addition
to the thickness and resistivity values of all different layers the profile
of the Lake Laach and the position of 5 flat drillings for heat flow
measurements are shown.
-905 -
The resistivities below the lake are determined by 25-50 om which
means clay between 10 and 300 m depth. These results are in agreement
with the results of the drillings No.2 and 4.
For point 4' in the north we measured in high resistivity of about 1400 om
which could be identified as Basalt in drilling No.1. The layers with
about 280 om are Lava-ashes. For point No.2 the resistivities of 63 om
and 107 om were identified as dry clay and sandstone by drilling 5.
South of Lake Laach a sharp vertical structure between point 6 and 7 with
550 om resistivity and a thickness of more than 200 m rising up to the
surface at point 9 can be seen as pumice all around that area. From these
results an average resistivity of about 300 om can be assumed for the
upper 300 m.
REFERENCES
Klocker, N. und G. Musmann, Geoelektrische und geomagnetische
Messungen am Laacher See (Eifel), GAMMA~, Institut fUr Geophysik,
Technische Universitat Braunschweig, 1979.
o 2km
FIG . I
:\ ;\
Krun ... Ol.n
2, ,_:\
8ohrpunkl.V.Ut.kOPf ,: \ / \
SI 'I,' \ 'I II
lOO __-r-.,.,....;,r-r_-"'..!;..~....:;!..;:..7'7.;r""C/:;;f..;;;1"''ilsch.cr'.. Kl'M1ppc.h..rl
f,.
200
'00
N CD
-908-
--.
1": \l ~
IS
•
. ! ~ T
·S •
-"
impedance tensors for
different frequencies
rotation angles "'_<-1_ _ _ __
• -1.1
...~11...,"~ I
apparent specific resistivities Pa [P12 and P211
phases between Hand E [~12 and ~211
Fig. 4
.... t
NW MT'
Olt-------~-----L----~--
MYI1 MTI MTIO
_______________________ L_ _:>::>::Z
LLL_L_
~
~~~
MT4 __MT6____________L __ _ _ _ _ __ ~
SE
• \
.
............,~
,..,
\
...,
I':"
I' " \
/ I" \ , ....
.
/ I \ \
"-
I
/ I
I -:- I
\
" ",
-
I",
, I ,
,
I:
-----.
111 _ _ _ 1:0
\ I
,
I
\
...'./ /
I
-- --- .-- - :!!'---- -'"
,"r: .- . .
I
i .!oO__ -
.- --
10 ,-.~.
"
.-.-
-- - - --- , ~ ..... \
-.
-~-
:::---=::~::::-:--------
I
\ --- 0 --- / \ I
(!
.....
.... " --- "\ /1/
I I
100
.....
T Cue)
RHO 21
fiG 5 ~-
..... :t
NW __-..T9____... TlI
Ol~---- ~ _ L_ _ _ _.... '1 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.. TtO
~ -L~~~~~~
~
i'-; NT" ...16 __________ ~L_
SE
_ _ _ __ _ _
- - - - - - - - !.~
•, ••0
\
, ,,-------ioO'----------·-
~--------------------~-
I
' / /1
/
/
/
....
"\
\
" " "" ________"!"!. ________ _ ,.-.
.' .--.. "",
" ,
10
/1
. ...... . ,, , ,
. .., ,, "" ..
\
\
... .....
.
'~',
, , ,,
\
.;-- ',
100 >00 ,, ' '
T (. .<I
,. '"
.- ..- -- ,\ " ,
,, , .
, '
FIG 6 0 ... ~ .. RHO 12
"\ •,
-910-
MAGNETOTELLURIC MEASUREMENTS AND GEOMAGNETIC DEPTH SOUNDING
IN THE AREA OF THE URACH GEOTHERMAL ANOMALY
Summary
-911-
In figure 1 the central part of the Drach geothermal anomaly
is presented with the main tectonic features and eruption si-
tes. The geothermal anomaly is mapped by the isolines of tem-
perature increase in m/1 oC (1). The measuring sites are mar-
ked on the map by different symbols. At sites of magneto-
telluric measurements the preference direction of the induced
electric field is indicated by a bar (for the period T = 100
sec). The length of the bar is a measure for the degree of
polarization. The regional preference direction of the indu-
ced electric field is about perpendicular ~o the strike of
the Swabian Alb. It is locally disturbed in the area of vol-
canic pipes. The strike of the Swabian Alb can be seen in
figure 1 by the line S - st.
---
+
10 km
~-
Figure 1
f
10
1 10 10 2 10 3
period T (sec)
"'£lST
coordinates for
figures 2 and 3
Figure 3
8
51
90
~
E
.t:
o
I'
, /
fI
-914 -
N
- 0 • 100 5@ C
-X - 10
E
9 10
0.1 ...
0.1
-~
~5W -20 o 20 1. 0 km - 'IE
X
Urach
Figure profile P
p ,. ,] -~c---+-Ir-it-+~~-~-~
11.1
!
t
60
~----
. ~ ... ~... ~ -Y!' ' - - - - _ _ _ _ _ _ _,
......
-.
+_
... .-J<---
-_ . • ° 1'VU "'" 2- ... ,.
'="=. 3-~- ~ _0.. -- -<>-- _ ..0.... ... + _____ .0- _ _ _ _ _ _ ...0
n+:: to.. "'3
-- - - - :
---.-.:~ - -'>" -
10 10 20 )0 km 50
-~ ~-NW '0 ~5E
u roer. Oo nou
profil S
,"'-"'& k
ia .
Le a ./ ~
Q"'"
j k
I
Z/H"
Figure s
- 915-
In figures 4a and 5a the lateral variation of the apparent
resistivity Sa and the phase difference ~ between the elec-
tric and magnetic field variations (in fig. 5a only) as well
as of the preference direction P and the degree of polari-
zation I are shown for the periods 10 and 100 sec along the
2 profiles. In figures 4b and 5b the induction arrows i.a.
and the amplitude ratio Z/H It (Z = vertical component of the
magnetic field, Hit = horizontal component, which is correla-
ted maximal with Z) as well as the phase difference ~between
Z and Hit are shown along the 2 profiles. The symbols II and J.
refer to the strike of the Swabian Alb: Sail' for instance,
is calculated for the currents flowing parallel to the Alb.
The preference directions and induction arrows are plotted
with respect to the geographic coordinate system. The extent
of the area of volcanic eruptions is marked by a thick line.
The extent of the geothermal anomaly, however, is much larger.
The lateral distribution of the parameters plotted in fig. 4
and 5 is smooth outside the volcanic area but disturbed in-
side this area. As a consequence the lateral distribution of
the conductivity must be more or less homogeneous outside
the area of volcanic eruptions but inhomogeneous inside this
area. Concerning the conductivity-depth-distribution we can
conclude from the results for larger periods of T ~ 100 sec
(like the rather constant preference directions and induction
arrows) that there must exist a 2-dimensional conductivity
anomaly in greater depth below the Swabian Alb. From the re-
sult for shorter periods of about 10 sec (which are more
disturbed along the profile~ we can conclude that the con-
ductivity-depth-distribution varies stronger near the earth
surface than in greater depth. A clear correlation between
the lateral variation of the temperature and the conductivity
within the geothermal anomaly could not be found. There is,
however, a clear effect of the area of volcanic eruptions on
the lateral variation of the electrical conductivity.
We will discuss now in more detail the results in the area of
volcanic eruptions. The results of geomagnetic depth sounding
(i.a., Z/H It ) are rather constant throughout the volcanic area
-916-
for periods of T ~ 100 sec, whereas clear lateral variations
occur for periods less than about 50 sec. See figures 4b, 5b
and the induction arrows in figure 6. The induction arrows
for T ~ 100 sec are caused by a 2-dimensional conductivity
anomaly below the Swabian Alb striking in direction of the
Alb. The induction arrows for T less than 50 sec are caused
by a local anomaly of increased conductivity within the vol-
canic area. The location and extent of this anomaly could
not be determined fully by geomagnetic depth sounding as the
distribution of the measuring sites was not optimal for this
purpose. Z/H" is large at the stations c, d, 3 and 4. It is
small at the stations f, g and 5. From the lateral distri-
bution of the ratio Z/H" we can just conclude, that the local
conductivity anomaly must be situated east of the stations
c and 4. The boundaries of the anomaly in the other directions
cannot be determined clearly by geomagnetic depth sounding -
neither from the ratio Z/H" nor from the direction of the
induction arrows for short periods.
More informations about the presumed anomaly may be deduced
from the lateral distribution of the preference directions.
This can be seen in figure 1: The induced electric field is
strongly polarized in the central part of the volcanic area
and it is only weak polarized outside this area. The trans-
ition zone between high and low polarization coincides in the
SW with the western boundary of the conductivity anomaly in-
dicated by geomagnetic depth sounding (maximum of Z/H" at
stations c and 4). The eastern boundary of this anomaly may
be near the station 11. The width of the anomaly thus should
be about 15 km. The preference direction of the electric
field is about NNW-SSE. Geomagnetic depth sounding, the degree
of polarization and the lateral distribution of the preference
directions indicate an elongation of the presumed anomaly in
NS or NNW-SSE direction. The continuation of the anomaly to
the NW and SE is, however, not fully known.
Further informations about this anomaly can be deduced from
the lateral variation of the apparent resistivity ~a plotted
in figures 4a and 5a: For T = 100 sec Bat! is small at the
-917 -
stations 5, 6 and 11 and it is large at the stations 3 and 8
(This is the B-polarization case with respect to the striking
direction of the presumed anomaly.). In contrary, 0 for T •
oaJ,.
100 sec is rather constant along the profile (This is the
E-polarization case with respect to the striking direction
of the presumed anomaly.). Both results indicate a small con-
ductivity anomaly near the stations 5, 6 and 11 striking
roughly in NS direction. The western boundary of the anomaly
may be near station 4, the eastern boundary near station 7.
In figure 5a the southern continuation of the anomaly may be
supposed near the stations f and g. We find similar results
for periods of about 10 sec. The lateral variation of Sa is,
however, disturbed in this period range by small near sur-
face anomalies.
The anomaly of increased electrical conductivity in the cen-
tral part of the volcanic area may be caused by fractures
within the sedimentary cover and the cristalline basement
which are filled with highly conducting fluids. The conduc-
tivity anomaly cannot be limited to the sedimentary cover.
The conductivity must be increased also within the cristalline
basement. This can be concluded from the results of geomag-
netic depth sounding where an increased amplitude of the
vertical component has been measured to periods of at least
50 sec. This effect cannot be produced by an increased con-
ductivity just in the sedimentary cover as the skin depth for
periods of about 50 sec is large compared with the thickness
of the sedimentary cover.
The presumed conductivity anomaly strikes about parallel to
the "Filder" lineaments marked by the symbol L in figure 1.
The conductivity anomaly may probably be a continuation of
these lineaments in SE direction. Consistent surface struc-
tures might be the Ermstal and the Lautertal (small river
valleys) •
Figure 6 shows typical examples of the parameters discussed
in this paper for stations representing the 4 different types
of curves: Type "SW" (BRE, Bremelau), transition type "SW" to
"CW" (BERN, Bernloch), type "CW" (GAECH, Gachingen) and type
-918 -
"CE" (WIES, Wiesensteig). For the localization of the stations
see the numbers and letters in the upper left hand corner of
the graphs and the numbers and letters referred to in figure
1.
! V..· ":,,
,. " .. , '" IEtE'A!>'I" IEm'"
······71!Effi'\
, i i #' '\ I
IHm
..
8a(T) " r~'
.
I,!
! !
, I I
'f,'
preference d. I~'- .,~~.~:~ ~''''':l::
.. -.; ';'~ £EE-~"'to~.~~~ ~·:·~·;~=
phase diff. I ,.,\ ; - _ .. I .. ;0;. I ....... .
(- - . ,_._ " ,_ ..:. 1:': L _ ....' _
E - H _ ...l+-=....:.....:~·,L..·-II ',/ ::.: .::..
.
±+~1 !j +,~" I:1
~ In rllU l
induction arr.
l• "\'
!~ t +~'I
IZ/H"I .
,
IIIIff'!
-- 'g'g'g1[11"3.32-".
\AlII III ,. ,
"'[C"eM.""
""'I" "I''' \,(11[516
WOIo I..
.I ~ '"
_1U_
:i -t7!~1, :t-EJ:J:t-tlj:t-tJ~j
phase diff.
Z - H"
!
1 10 10 2 10 3 (sec) period T
Figure 6
- 919-
mentary cover and upper part of the cristalline basement).
s* becomes maximal in a z*-depth of about 30 km. Differences
in the conductiYity-depth-distribution indicated by the
different S*Cz*)-curves may partly be caused by the dis-
torting effect of the lateral conductivity distribution and
partly by different conductivity-depth-distributions inside
and outside the volcanic area.
E-P • SJO
• GAE
• APF
o HU
o BRE
o MUN
• ING
z. IKMJ + MAS
• APF
o HUN
co BRE
.. MUN
• I NG
..,
+ MAS
15
. ,~
Figures 7 and 8
-920 -
SOME RESULTS OF VERY LOW FREQUENCY
MAGNETO-TELLURIC SURVEY IN THE MONT DORE AREA
(MT 5 EX AND HARMONIC SOLUTIONS)
G. MALERGUE
Bureau d'Etudes Industrielles et de Cooperation
de 1'Institut Fran~ais du Petrole
Abstract
This survey of the Mont Dore geothermal area was carried out mainly
with a view to prospecting before locating the first exploratory borehole.
The aim was to locate, in both depth and horizontal extension, a
possible conductive body, which in known geothermal fields is related to
hot water reservoirs.
Despite the high measurement polarization level, two conductive
anomalies have been mapped with longitudinal parameters.
One is particularly promising because of the very low resistivities
measured in the medium range of depth.
But, as the calculated parameters proved to be only roughly r€pea-
table at some particular stations, the results cannot be reliably inter-
preted from a quantitative angle.
-921-
INTRODUCTION
-922 -
Therefore, in this type of environment one should aim at finding a
block of these sub-caldera rocks offering both a strong fracturation at
the top (good permeability of the reservoir) and an extension of the
fracturation at depth (high rate of heat transfer favouring high reservoir
temperatures). Fig.1 is a sketch diagram of such a geometry of a caldera
environment.
2. The use of the Magneto- Telluric method as a direct prospection tool
From our experience gained in previous surveys carried out on various
geothermal fields, it appears that the fracturated blocks, defined above
as the objective, give anomalous low resistivity measurements in the 10
to 100 seconds (0.1 to 0.01 Hz) period range of analysis. If the Total
Conductance parameter is used for mapping, the conductive anomalies
in water-plus-steam geothermal fields often reveal producing areas
directly.
The low resistivity of the reservoir can be explained by :-the very low
resistivity reached by the hot and saline water contained in the reservoir
(For exemple 0.03 n.m for a 5g/1 Nacl water a T = 200°C).
- The geometry of this water in the reservoir rock. Filling the fractures,
the water forms a continuous film. Contrary to porous sedimentary rocks,
most of this very conductive water is involved in the final conductibility
of the rock.
-The low resistivity of the geothermal deposits and alteration products.ln
the upper parts of the reservoir conductive hydrothermal deposits have
accumulated. The "self sealing", that gives a cover to the reservoir, is
a result of this phenomenon. A brine accumulation can exist, and also
give a very low resistivity level.
The Magneto-Telluric measurements in the lowest frequency range integrate
and cumulate the effects of both the reservoir and its fracturated deep
extension.
Fig.N°II and III show the frequency range in which the apparent resisti-
vity measurements are affected by a deep conductive body. These theore-
tical curves apply to a tabular distribution of the levels, but they
can give a rough indication for real measurements.
3. The field survey
As the main target was not only to detect but also to delineate possible
conductive anomalies, we intended to carry out the measurements on a
maximum of points. About 110 "stations" were set up on the field within
and close to the supposed boundaries of the caldera (Fig.IV). These
stations were regularly spaced at a distance of between 0.7 to 1.5 km
apart.
Two kinds of recording laboratories were used: A truck mounted model
and a portable unit. Both amplified 5 components in the 5 to 100 second
range which were numerically recorded. The 3 magnetic sensors used were
CNRS CM10 Models.
- 923-
A high level of noise and perturbations were encountered in this area.
More than 80 stations gave records suitable for processing. Some of
these stations were recorded several times in order to study the repea-
tability of the calculated results.
4 - The apparent longitudinal conductance map:
Most of the stations recorded in this area show a high level of
"Polarization". The ratio of resistivities measured for the same frequency
with only two components in two perpendicular directions ranges from
1.5 to 30. The highest ratios are found near the edges of the caldera.
Owing to this polarization (typical of volcanic deposits and structures)
only the calculated longitudinal parameters can be mapped with certainty.
We use the "longitudinal apparent conductance"as a parameter for the
mapping of a possible anomaly. We recall that this parameter is cal-
culated in the lowest frequency range of the resistivity curves. It inter-
grates the effects of all the conductive levels from the surface to a
resistive basement. In tabular cases,this apparent conductance is the
measurement of the total conductance of the deposits
ement hi
Ca = ~ u face ~ p1
where hi and pi are respectively the thickness and the resistivity of
each level.
The Ca lonparameter mapped here is calculated with a software involving
exponential solutions of Maxwell equations.
The map is based on the results of 84 stations, taking into account their
polarization levels and their reliability control. It shows: (Fig.V)
-The extend of the medium conductive deposits inside the caldera,
and therefore the caldera boundaries beneath the last lava cower.
-Two very conductive areas, where calculated conductances reach
anomalous levels in this geological environment. One is located
East of the Mont Dore valley near the eastern edge of the caldera.
This anomaly is mapped from 6 results placed along a N.S. line.
The other is located South-east of the town La Bourboule. This
anomaly gives conQuctances of overl100 Mho at three different
places and extends over at least 4 km2.
For the sake of comparison the conductances measured in another
caldera, about 1,300m deep, which was revealed to be a high
temperature geothermal field, range from 350 to 2200 Mho.
Productive wells have been located in conductive areas of more
than 900 Mho, which extend over several square kilometers.
The interpretation of the conductances is not easy. A high value of
conductance measured on a single station can indicate a real
low resistivity body. at depth, but can also be the result of an
"edge effect", typically encountered over faults or structures that
separate two geological areas of very different resistivi ties.
-924 -
One can only attempt to interpret a group of measurements using two or
three dimention models. If the distance between the stations is short
enough, the evolution of both conductance and polarization ratio
measured at these stations can be connected to a theoretical calculated
model of a simple geometric distribution, or to a scale model studied
with high frequencies.
In this study of the Mont Dore area the spacing of the stations which
gives reliable results is loose, mainly in the valleys because of in-
dustrial perturbations.
The elongated anomaly, east of the Mont-Dore valley (Sto n020, 54,74,
75 and 79) could represent a North-South faulted zone, with possible
water circulation but also with probable edge effect. This structure
indeed belongs to the main fault system that limits the caldera
at the East and separates conductive deposits from the highly resisti-
~ basement.
-925 -
We did not record the frequency range which could be used to study the
shallow rock resistivities or depth variations. So, for our theur't!1:1C
models we gave constant depths and resistivities to the caldera deposits.
The total depth (700m) and resistivities (300 nm, then 2 n.m) correspond
to the results of electrical and wide range magneto-telluric surveys in
the area South-east of La Bourboule. A very conductive level was supposed
at various depths.
Fig. II and III are examples of the theoretical curves obtained from such
a model.
In comparison with these curves,the experimenal ones should indicate
at some stations inside of the caldera such a conductive level. The
depth ranges from 1.5 km to more than 3 km. On the conductance anomaly
located South East of La Bourboule, the vertical extent and the con-
ductivity of this level should increase greatly. The top of this very
conductive body should be about 2 km deep.
Most of the interpretable curves show high resistivities at greater depths.
This means that the remains of the magmatic chamber have now ceased to
be conductive.
For a more direct comparison, we processed with the software developped
for the Mont Dore survey a few records from others surveys. Fig.VIII shows
some curves obtained on other geothermal fields that proved to be productive.
6 - Some results of reliability and accuracy tests:
Some test stations (N°89-68-69-117, .. ) were recorded several times,with
different signal levels, during the 3 field survey periods.
Fig.IX shows the curves we obtained for two of these stations, St.N°89
was recorded twice the same day, and gave interpretable curves. St N°69
was recorded twice at a monthly interval and only gave a cluster of
resistivity points.
Such tests indicate that the resistivity measured with two componemts
are only roughly repeatable (with the software we used). The longitudinal
calculated parameters are more reliable but here do not reach a high
accuracy. St N°69, for example, gave longitudinal conductances ranging
from 900 to 1300 Mho. These values remain nevertheless anomalously
conductive.
We normally get better repeatability in less industrialized areas. In
the Mont Dore area there are many disturbing sources, such as electrical
fences. The telluric field is the more perturbed of the two fields at
most of the stations. This explains why we chose to set an energy
threshold on only the magnetic components to select the best parts of a
record.
CONCLUSIONS
This survey reveals out and maps two anomalous conductive areas inside
the Mont Dore caldera. One, located south-east of La Bourboule, seems
to be the most promising. The very conductive body which creates this
-926 -
anomaly extendsin both horizontal and vertical dimensions. This conductive
body should be in our interpretation a hiohlv fracturated zone, full of
hot water,in the sub-caldera hard rocks (lava flows or basement).
This interpretation is mainly based on the experlence gainea on otner
geothermal fields which have already proved to be productive. In similar
geological environments, we learned that a hot water reservoir in a
caldera is favourable for the use of the MT method as a direct prospection
tool. The reservoirs and the main convection cells that canalize the
anomalous heat flow are here often superimposed. They cumulate their
effect on very low MT frequency measurements . However, owing to both the
complicated geometry of the structures at depth and the poor accuracy of
the measurements, we cannot attempt to obtain a true quantitative inter-
pretation of the results. This work method is probably not very satis-
factory for the scientist, but it is often sufficient for the driller.
Meteotlc •• ler .
Fig. I.
CALDERA DIAGRAM
With poIllble loe.tlona
of high .nergy g4oth..rmal
r... rvol .... .
-927 -
" r:::- .. .-
nm~~-4-4~++++--~~~4-~~~-H4---~,--
"' 1
~"'!-,
1- 1 /--.
Lf-..
IA
-:-~ ".
/~. i'. r}
~" ..... I'i '.~
t- I'.. ""'/
t-'
r--L
I
Fig II
I
vr l.o ~
~Uc m"g:rtelo.leUutk curvet
fo< • condLX:U.... ""dy _ f .
,25 • 700m calo.,.. .
A 50 19" ~ T•
.!i .\0 04 ,I n .61 F,..
"
': -:::::::::
= ""' C+I"'---;~:
' -l rr-'- - .
if":: I::::
1,1:: '
...
.'"
~:.. . ,l' i"~
,... . ( t ·,. ._.
•
:.-
, --'
.- ---,-,
'-''''- ~
• --. '''. too- "
14 ~
I S--
...
• •.n._
2. ,, - 2.
...
11 - 4. l\AII
10 '1_ 1. lJ- .6
17 - .4 :u- .4
1,_ .2 ZI- .2
Fig III Th*Oletic
ot YI".""
m.gMlo.,.lIurk: eurw, 'or • conduct I " bOd,
, ••J,tlWity . a • • IIhpth 0' 3km . Itte'" 5 km
2, . .1
benuil'l • 1'OOm calder. .
- 928-
629.00
'12.00
10.00
65.00
SO.OO
55.00
629.00
-930 -
S~LC 1" t t lur - lC"'[l.l( lOM~n H. I \,Iut.
,,
FOR THE HARMONIC
,
11 11
, " , PROCESSING
115>-
as-
I!.S-
uI ' s-
~
FREQUEC I ES
1 750 -
la- -----lI.~ TO AVO I D
II~ -
...
0; .
0>.
Ex
,>.
S>.
'!Io - ••
l~- •••• • ••••••••• , ••••••••••••••••••••••••••• , . . . . . . . . . . . . . . . . . . . . . . . . . . . . • ••••• • •
1~- ••••• • ••••••••••• • • , • •• •••••••••••••••••••••••••••••••••••••••••••••••••••••••••• • ••
! - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . • •••••••• P • • • •
Hy
"
"5-
,,
", t'! + .., ~ ", ", ", .", '", "',
U,S·
ISS-
l ' !-
ns-
Il~-
Il'!i-
1'5-
". t Ey
". . .t.
......................,....,.........,.......................
,~
Hx
12~
us-
In-
".
0>.
".
...
".
"
lS·
.
•••••••••••• , •••••••• " " •••••• ' •••••••••• ' •••••••• " •••••••••••
J.
1~-
I!- . , ••••••••••••••••••• , ••••••••••••••••••••••••••••••• , ••••••••••••••••••••••••
::.: : :: : ::~::::: :~: ~ : ~~ :~ : :: ::: : :::::::: : ::: : :: ::::::::::: :::::~:::~::::::~~::: :~:~~::: . .l.. -.A. __ _____ ~ •.•••••• ________ ••• _
I I I I I t i t I I I I
I II II Jt ~1 '1 ~":..-=-='.:.'__;'_:__:.,,',.,.,.-:--:':":-::-=___'_;
IO;-'=,.-;-.:',,"==:-:'':i''
,, .,, Sl
, , ,
lUi
,
lil
,
III
~
11:'-
'6!.-
I~~
I'~
us-
Hz
In-
II!' ••
l l
I'S- I ••
-931-
Aa /
Ra
.17.."" .n.m
Y
Y I
,0
/
100
/
...
.if·
100
f
\ Y.
"\' .
f-*'f
tf-.,/
/'
100
10
"92 9 25 100 T
®
/
./ Ra
,../
• ~'l
\~
\~-I
If
I
"
4 9 25 100 T(S)
6 100
)I(
1.
65
9 100 T
68
4 9 25 100 T
- 932-
Ra @
100 41 I
R.
100
/i)
31 0 '(
J/ 36 ~I
I
/
.(j)
i
'f ;(
)(
,{ I
/~y /
~ 10
". '"
I
yIC. ~JI('"
I
>/
I~ • I I
• Y y /
.
l! y x
1O x'X I
y x x
/ -!,c.~_~ l f
to y
A -,.:-~-®
9 25 100 T
9 25 100
.. .
I
I
~I
~ ~
..... X I
'< ~--.-'
100 T
- 933-
R
,,;
y
I
i : ...A~
i ,,:I¥
./
O.10E03
(TOO)
~ t•
'\
'~
"
x
x
O.10E02 II
(10.) ltX
~
y
y
x y
O.10E01 T
(1. )
O.10E01 O.10E02
t - -_ _ _ _ _ __ ....&...---.;T 69 AOJ
Some results of the r epeatability tests
69 A02 Fig . IX
carried out in the Mont Dore area .
-934 -
MAGNETOTELLURIC PROSPECTION OF THE MONT DORE AREA
-935 -
I. INTRODUCTION
2. LES EQUIPEMENTS
-936 -
Nous disposons pour notre part en dehors de cette M.T. lente, de
deux appareils destines a l'etude de frequences plus rapides :
-937 -
de son remplissage, la presence eventuelle de conducteurs profonds, la
fracturation interne.
Des les premieres mesures, il est apparu que nous disposions d'un
appareil bien adapte a la definition du contour de la caldeira (figure 2)
les courbes etant fort differentes de part et d'autrt d'une ligne de
contour probable indiquee par R. Brousse dans la region du lac de Guery.
De plus la presence d'un terrain tres conducteur (quelques nm) etait
confirmee sous un recouvrementresistant de quelques centaines d'nm ou
parfois de quelques dizaines d'nm.
4. CONCLUSION
po
".. !lm
1I
21
)'
5)
62
150
Un profil ex'cutf avec 18 traina8U at mont rant l' anomali a cr'~e par 1a
bordure de 1a caldeira du c8t~ de 1a Banne d'Ordanche
FIG. 1
- 940-
- __ I!'t) . • ,11'1.
~"".
_S1.11.'MI
.. ...
"'"'-'"
. . . r..... 1
•• tv...........
Figure 2
_ un profil execute avec Ie resistivimetre et montrant Ie passage
de l'exterieur a l'interieur de la caldeira
_ quelques courbes obtenues avec Ie resistivimetre dans la region
de Charbonniere.
10000
p.
n.
looob-- - - - + - -
. ......... --
, .. ' .
-'
,x ;~ RI,htr cit ( A.gJt
" f.W.
1Or-------+-------~----~_+--~--_J
Figure 3
- une courbe M.T. d'un sondage execute a l'exterieur de la caldeira:
Banson
- une courbe M.T. d'un sondage execute a l'interieur de la caldeira:
Ie Rocher de l'Ai&le.
- 941 -
• y •
....
........
.'"
.... ....
.... .... .'" .... .."
.... .."
-942 -
/,
ISUlLU1IBS 1111 '1'01'1' 1111
c:O;~nlJf:TBlI H l .. mFO;~n ;......,~_",;I••
'!
- 943-
SESSION VIII - Hot Dry Rocks
-945 -
HYDRAULIC FRACTURING OF HOT DRY ROCKS.
TRIDIMENSIONAL STUDIES OF CRACKS PROPA-
GATION AND INTERACTION BY PHOTOELASTIC
METHODS
o
G. F. CAMPONUOVO, (+) A. FREDDI, (") M. BORSETTO ( )
Abstract
- 947-
1. INTRODUCTION
2. EXPERIMENTAL PROCEDURE
-948-
quasi-static pressurization (only for a few polyester models): after
setting the pressure to the fixed value the hydraulic circuit was closed;
the procedure was repeated every 15 minutes;
cyclic (or fatigue) pressurization: the hydraulic circuit was driven
by a simple flip-flop electric circuit. Pumping rates ranged from 70 to
80 cycles per minute.
The calibration made for the case of single crack demonstrated
that the fatigue propagation tests were perfectly interpreted in terms of
the classical laws of fatigue propagation (2). These laws have been used
in fact for determining the material parameters of the two kind of models
(epoxy and polyester). No difference in behaviour has been recorded be-
tween static or cyclic pressurization if cases of non stable propagation are
disregarded.
The front evolution was recorded (hand plotting) by means of the
projection polariscope (Fig. 3) used as a profile projector. In a few
cases also a TV System and/or a film camera were used.
For the boundary stress 6'0><' a steel frame was used with a rubber
camera loaded by water and feeded pneumatically through a separator bag.
(Fig. 3). In the room temperature propagation tests the constant value
(3""... = 1. 02 MPa was mantained.
3. RESULTS
The case of single crack was only analyzed for calibrating the
effectiveness of the experimental procedure. A s already mentioned the
results were perfectly satisfactory and in accordance with the theory.
The case of two collinear cracks (parallel and co-planar) is the
most favorable situation from the point of view of possibility of linking,
though its feasibility in the case of an actual HDR project is not so likely.
The tests have confirmed that in this case there is no doubt that
the two cracks do link (Fig. 6) either in presence or in absence of
boundary stresses; in all cases a delay of a few thousands cycles of
pressurization was recorded between a geometry of contact and the actual
linking, due to a small overlap zone. From the theory <Oe.... should not
have any influence on the behaviour in this case.
The case of two parallel non co-planar cracks isthemostprobable
case in the realization of an actual plant, and then the most deeply studied.
The following cases were analyzed:
In absence of boundary stresses ( S",,= 0) no problems arose in
getting linking when ~~q_oE~5!.~C!_i!.l?!.~s~~~!~~ provided the distance
between the crack planes is not too large (h/a:s O. 2), in order to have
effective interaction. The trajectory for getting linking is straight with
very good approximation. In fig. 5 the distortion of the isocromaticloops
of the pressurized tip are reported for two different distances between
crack planes when only one crack is pressurized.
When .!>~~_c:E~~k~_a':~.l?::.~s~~~!~d_ ( Sox = 0) a characteristics b~
haviour occours in all the tests on non co-planar cracks: after overlap
each crack front bends towards the other (Fig. 7) while the propagation
diagram records a drastic deacceleration, up to a stop. The rest of the
front goes on overlapping and bending in the same way, so that finally the
overlap zone becomes a strip of constant width (the width is a direct
function of crack distance), but the process of approaching cannot go on
any more.
This behaviour has been confirmed by the photoelastic evaluations
of SIF (in the overlap zone the values of Kr were lower than elsewhere),
-950-
and by the tests reported in (7). This feature has also been recorded in
cases of great instability in propagation: due to a level of pressurization
too high, a few cracks collapsed, coming out of the model, but also in these
cases they did not link.
Given the constant width of the overlap strip the situation in the
overlap zone can be schematized as 2 D plane strain problem.
When boundary stresses (<0= 0 0 >< ) are applied the behaviour is
mainly determined by the ratio 6/P (where p = internal pressure):
For values of this ratio ~ 1 about the same behaviour as for the case
6" = 0 was recorded, with little differences in the angles ~ of bending in
the overlap zone.
When (5" prevails over p the possibilities of linking seem to reduce
.
rapidly, particularly when only one crack is pressurized (see Fig. 5 d for
6:' /p ~3) •
"
In fig. 8 some of the photoelastic evaluations of SIF in the overlap
zone are reported. For more details on the experimental results reference
is made to report (8).
In order to supportthe interpretation of experimental results, two
dimensional plain strain finite element computations have been performed.
Reference geometry and a detail of the mesh used are presented in Fig. 9.
Second order displacement interpolation was used all over the
mesh, and the quarter point technique applied at crack tips (9). The evalu
ation of stress intensity factors was performed by fitting the crack opening
displacement at various distances r from the crack tip and extrapolating
to r = 0 (10).
The K-factors were computed for unit load cases and then used
for any load combination provided that displacement superposition gave no
interpenetration of the two lips of a crack. In Fig. 9 the results of a few
admissible superpositions are presented. Non linear analysis is requested
in a general case.
CONCL USIONS
-951 -
Fig. I Two circular predlap>led Fig. 2 An epoxy model (100 x 100 x 200 mm) with three
defectaina polyeoter model circular cracks by impact load
(11 0 x 200 x 300 mm)
-952 -
Fig. 5: Typical isocromatics patterns near crack tips (overlap zone)
a) Isolated tip; crox = 0 (from Modell O. 11). Same pattern for single crack
- 953-
o~- . d
d2a . 0.91
No. of cycles
1 6,200
2 12,700
3 11 ,750
4 21,300
5 24,500
6 32,000
7 41,500
8 45,500
9 56,000
10 70,200
11 82,000
12 92,000
13 100,000
~~o~--- Proclldun: Alter •• ntnc th_ pn ..un to the lixed value the hydraulic
circuit .... elo •• d. Thl. procedure ..... rflipeilltd Iyery 20
minute I,
Fig . 7 Propagation diagram for Model P. 8 (110 x 200 x 300 mm) in polyester.
-954 -
Tip A Tip B
PI Pl eTO)(
~od.l ~ b/_ i,! b c.ia
Cii I"-r, Ib_r, \bar, ~II ~I!I ~II MU
O. OS 0.52 O. '2~
j (
0.15 0.52 0.22 1. 21 0.13
,§.~ I
0.06 0.53 0.23 0.11 O. U
F
10-28 0.40 0.00 0.110
0.14 0.53 lJ. 2J :. JI 0.16
~
,.
0.06
0. 12
0.53
0. 53
0.27
0 . 25
1. 14
1. 45
0.241
0.11 I .. I
0 . 07 0 ,.501 0 . 26 0.53 0.10 0 . 32 O. 12
10 - 1 1 1. 85 1. 65 0.00
0.07 O.H 0.20 O. S2 0.13 0. 32 0. 18
II ~ KIKo
O. H
0. 07
0.56
O. i l
0. 17
O. '9
O. ,6
0.62
0 . 00
O. • c
K
0
.. 2p .,Ja!TT
10-l9 o. : 2 O. 00 ! . Z1
O. I: O. H 0. 13 O. 17 0 . 00
0.10 0. 65 D. 5S O. JI 0.00
--
t=O.66
f=0.33 t - Oox
--- 'I 1
2a
t=o.s I ®
t-
'A' , , , , ,,~
t-
Af
---
It r t ttl I I
t-
I ", I-
til
I I 'I
@ tip A tip B
I I I I , v.X'
, , , , , , , , 'I' ,-,
I I I 14 I-
t- ctx MI Mn MI MU
-- f'l t t-
I-
-
23 C o. 0 1. 55 -0. 76 2.31 0.04
1 '2b t-
t- 0.5 1. 54 -0. 77 2.02 0.41
I ~~.,..JffiI
I 0.7
1.0
1. 54
1. 53
- 0.77
-0.77
1. 90
1. 72
O. 56
O. 78
/ ~ P": ~ b? I/'.,
1"'/
l/'.- '"
U- ~
V 11\
7
Pl = P2 = 1. 0
V~ 1/\V 1/ M = KIKo
Fig. 9 Schematization, detail of the mesh and results of FEM analysis in the overlap zone.
- 955-
and "in situ" stresses must be chosen in such a way that attraction
between the fronts of the cracks can prevail against the opposite effect of
the maximum principal stress. which tends to straighten the propagation
pattern and to inhibit connection.
The tests have clearly demonstrated the consistence of fracture
mechanics in dealing with interaction and connection possibilities of circu
lar cracks in an elastic medium pre loaded by an "in situ" stress. -
As the basic assumptions of the simulation were in fact of perfect
elasticity and isotropy of the medium. the results obtained have to be
considered as restrictive: permeability effects in the rock and/or presence
of natural discontinuities can in fact determinate linking even against the
theoretical and experimental previsions. Neverthless the proposed ap-
proach seams to the authors to be resulted as powerful and effective. even
if the research cannot be considered as conclusive in results.
REFERENCES
- 956-
SCALE MODELS STUDIES OF WELL LINKING BY HYDRAULIC FRACTURING
AND FATIGUE MICROFRACTURING
Abstract
I - Introduction
-957 -
II - Scale-Model Investigation of Well Linkage by Hydraulic Fracturing
1 - Introduction
on a hyperbola
-958 -
These equations show that there is a region in the plane where the
stresses are tractions. The orthogonal paths of these tractions are the
hyperbolas, and they intersect the crack. There is thus a theoretical
possibility of linking this first crack with a second one created by a
traction break.
Two important comments can also be made on the basis of the equations
for the stress fields :
a) The stresses are proportional to the fluid pressure inside the crack.
The above calculations were performed for the simple case of a conti-
nuous medium containing only a single crack. They thus cannot be used
for forecasting the growth of a fracture in the stress fields created by
pressure inside a second fracture.
For the same pressure inside the crack, the stress field obtained
after the reduction of the crack length is similar to the stress field
around the full scale crack, with the similarity ratio being equal to
the geometric reduction ratio.
-959 -
can be evaluated by the conventional criterion
K
(~P) _IC_
r:;;:c-
=
rea 1
with KIC real being the intensity factor of the critical stresses of the
real material.
~p = propagation pressure - miminum horizontal stress.
The mixture finally choosen for the rest of the experiments contained
10 %bentonite. The corresponding critical-stress intensity factor was
1. 6 x 10 5 ffll- 3/ 2
- 960-
Compared to the critical-stress intensity factor of a rock for which
the well linking process can be applied (e.g. a granite such that
KIC = 3.5 x 106Nm-3/2 ). the geometric reduction factor obtained is then
about 2 x 10- 3
Two wells were drilled in each block. The distance from one well
to the other was 15 em. With the geometric reduction factor given above,
a model was thus made of the experiment to link two boreholes 75 meters
apart.
One well is fractured and the pressure is kept constant while the
second one is fractured. Then the two fractures are alternatively deve-
loped by pumping the fracturing fluid in one well and keeping in the other
one a pressure lower than the propagation pressure. A dye is used in
order to observe the fractures after the experiment.
-961-
(photograph n° II). and the connection is obtained with a low incidence
angle.
Injection pressure varied between 220 and 250 bars for the first
test. At the end of the fourth cycle the sample was fractured.
For the second test, pressure varied between 178 and 210 bars.
No improvement in the permeability was found during the experiment.
-962-
-..."
~
N
u
:!
-..."
~
u
:!
-963 -
A ?iT
,/ A? :;J 1
I I
I - - -L
e .40mm
1.16Omm
a.l0mrn
b .<1Omm
Re:corder Pressure Volume
Fig I Surface energy measurement
21K .J05
Ie
I
>Q o
0-
... 1,51 Wells
I ; :lOx 30
I '
r-- 15----;
. I
I
C.ment block
1,01 ~
50- -
o~*------------+-------
01 1 1 1
10 20 Bentonite %
Fig.2 Influence af bentonite on the stress intensity foetor of cement
Leak off flow
regulation
Recorder F
Buffer
Oil 130 cp cell
Flow rate
I
0-
V-
""
I
Pump
Pump
t
void void
Pressure Regulation
REFERENCES
(2) BUI,
Mecanique de la rupture fragile - MASSON - 1978.
- 966-
MICROSISMIC AND ACOUSTIC ACTIVITY ASSOCIATED
WITH HYDRAULIC FRACTURE PROPAGATION
F.H. Cornet
Abstract
Sismic and acoustic sources activated by the development of hydrau-
lic fractures are discussed. It is shown that acoustic activity asso-
ciated with crack extension should be high frequency j its existence
depends upon pressure distribution at crack tip. For permeable fractured
material, the increment in pore pressure in the vicinity of the hydraulic
fracture may lead to some local slips which, in turn, can generate signal~
The frequency spectrum of these signals depends on the amount of injec-
ted volume and injection duration. In situ experiments carried out to
investigate the possibi lity of mapping this activity in order to deter-
mine orientation and extent of a hydraul ic fracture are presented. By
comparison with previously published results, it is concluded that hydro-
phones are not appropriate for monitoring this type of acoustic activity.
Further no low frequency induced seismicity has been observed although
up to 12 m3 meters have been injected.
It is proposed that only three components high frequency geophones
(I inear response up to 5000 Hz) tightly coupled with the rock in deep
boreholes should be used for "this type of investigation.
-967 -
I. INTRODUCTION
Hydraulic fracture orientation can be determined at the wei I bore wi~
satisfactory accuracy but with much more difficulty away from the injec-
tion wei I. For this purpose, many geophysical techniques have been pro-
posed, or are under investigation (e.g. induced electrical potential or
self potential measurements; determination of induced magnetic fields
active sismic and acoustic sounding). The present paper is concerned
with the mapping of seismic and acoustic events associated with the
development of hydraulic fractures.
-968-
dx
(2)
11
-969-
Without a better understanding of the pressure distribution along
the fissure, only experimental results can help answer this question. It
can only be concluded that if a jerky fracture propagation occurs, the
radiated energy should be associated mostly with compression waves in the
I - 10 KHz frequency domaine.
where ~WE(X) is the elastic strain energy associated with any virtual
crack configuration X ;
llWB(X) is the free surface energy absorbed by this virtual con-
fi gurati on
Xo is the actual configuration.
Equation (6) implies that, for materials with an anisotropic free
surface energy, a fracture may propagate at an angle with the direction
of the major principal stress. For such conditions, a numerical inves-
tigation has shown that the influence of pressure distribution on frac-
ture stability is similar to that 'described for the classical theory.
However, in this case, shear maves are generated in addition to compres-
sion waves.
where I; is the shear component of the stress vector applied on the dis-
continuity under consideration;
op is the normal component of this stress vector
I;
o
is the apparent cohesion of the discontinuity
1J is the internal friction coefficient
P Is the pore pressure.
-970 -
An increase in pore pressure induces a decrease in the normal com-
ponent ( 0 -P) whilst the shear ~ remains constant. As a consequence,
n
sliding may occur along favorably oriented faults and an induced seis-
micity may be observed.
N = V I (-.L) (8)
s
3
where N is the frequency, Vs is the shear wave velocity, and e is the
discontinuity length.
3. IN-SITU EXPERIMENT
Experiments described here after were conducted within the frame
work of an in-situ investigation on Hot Dry Rock geothermal energy
potentials undertaken at Le Mayet de Montagne test site (25 km SE of
Vichy).
This site is characterized by a very homogeneous granite; P wave
velocity measurements obtained at depth larger than 15 m is the same as
that determined in the laboratory for 5 cm long rock samples (5600 m/sec).
The upper 15 m layer is composed of altered granite (P wave velocity
decreases progressively from 5600 m/sec. at a 15 m depth to 1000 m/sec.
in the upper meter).
-971 -
Table I Hydraulic fracture experiments undertaken at Le Mayet de
Mon t a ne test site.
)I(
Test nO Well Depth maximum breakdown fracturi ng Total
m f low rate pressure ~Iuid injected
11m c.NI b. vol ume
I INAG 3-2 27 60 330 water -
~ for tests 2 to 6 the maximum flow rate was maintained during the
first 20 minutes of injection; it was dropped afterwards.
(band pass of 30-50 Hz, 60-80 Hz, 100-150 Hz) so as to try to screen
-972 -
events from background noise. During the first fracturing test (27 m
depth), spatial filtering was experimented (e.g. Smith 1956). Arrays of
nine vertical geophones were used as single sensors. This technique
revealed very efficient to eliminating stationary surface waves gene-
rated by the pumping unit. However no signals associated to the crack
extension were recorded. In fact, the major difficulty encounteredwilhihis
mapping technique results from the fact that signals are weak and high
frequency so that coupling with the ground is a very serious problem.
Previous experiments (L.A.S.L. 1975, Power et al. 1976) revealed that
seismic and acoustic activity is stil I very noticeable after pumping
has stopped, thus it was decided to concentrate on the coupling problem
and not on the spatial filtering.
From Power et al. results, it was considered that, for the high
frequency domaine (100-5000 Hz) hydrophones located in boreholes deep
enough to reach the water level would be appropriate. Accordingly ten
shal low boreholes (up to 8m depth) were dri I led in various azimuth at
distances from the injection wei I ranging from 10 to 100 m. In addition,
since two 200 m deepwells were available, a hydrophone was located
in the borehole which was not used for injection at a depth equal to
that of the straddle packer in the other hole. Thus a sensor was placed
30 m away from the injection point.
Two types of hydrophones were tested. The first ones were buil~
by Institut Fran~ais du Petrole ; the response is not I inear but their
pic of sensibi I ity are sufficiently wei I distributed in the 100-5000Hz
domaine to insure accurate determination of first arrival times. The
second type was provided by BrUel and Kjaer (Hydrophones 8104) ; it is
I i near ina Iarge frequency range (Q.I Hz to 50 KHz). Its sens i b iii ty
is equal to -205 db for a IV/ ~Pa reference. After amplification its
apparent sensibility was 42 mv/Fa (4.2 V/mil libar) for an electronic
background no i se of 5 mv.
Finally a geophone (Mark Products LOO, vertical component only,
65 Vim/sec) with a 100 Hz natural frequency was tightly fixed on a
granite outcropping 20 m away from the injection well.
The background noise recorded with these sensors is shown on table 2.
Beside the initiation of fracture for experiments number 6 and 10, no
signals were recorded. Recording periods occured either during the first
30 minutes of the injection or at the end of the fracturing experiment
-973 -
(most of them lasted between 3 and 4 hours). For the second recording
period, observation spanned on 10 minutes of injection, 10 to 20 minutes
during which the wei I was kept pressurized whi Ie injection had stopped,
15 minutes during which the hole was opened.
In situ stress determination indica~5 that fracture did not propa-
gate in the major principal stress direction and that the maximum shear
stress was of the order of 20 bars (for more informations on this deter-
mination see other paper by Cornet in this volume on multiple linking
between two wells). These are sufficient conditions to insure that signals
were generated during the various fracturing experiments. The question
then arises as to why no signals were observed whilst both the Los Alamos
Scientific Laboratory and Power et al. (1976) were successful in their
observation. It is proposed that for high frequency signals, hydrophones
were not appropriate because of the velocity contrast that exisffibetween
water and granite and because of the low energy radiated. No signals were
generated probably in the 1-100 Hz frequency range because of the rela-
tively sma I I amount of injected fluid.
Table 2 : Background noise recorded with the various sensors used at the
Mayet de Montagne test site.
Sensor gain fi Iter recorded background sensor
noise noise at the sens i b I I i ty
sensor
Hz mV ].IV ].Im7s
Geophone 2-15 40 80 .68
Mark Products 30-50 15 30 .25 117V /m/sec.
2.'
2 Hz 70-80 10 20 .17
100-150 3 6 .05
Geophone
Mark Products 42500 100-2000 350 8 • 12 65 Vim/sec •
100 Hz
Hyal rophone 2-5000 60 600 .94 0.42 mV/Pa,
630 Vim/sec
Bruel & Kjrer 100 for a 1500m/
8104 160-6000 40 400 .63 sec. P wave
velocity in
water.
-974 -
4. CONCLUSION
It has been shown that when hydraulic fractures propagate in the
maximum principal stress direction, if the fluid pressure is not uniform
up to the crack tip, the fracturing process is stable. However if the
fracture is inclined with respect to the major principal stress direction
or if some fluid can percolate in the rock mass, a significant sismic and
acoustic activity can be expected. For high frequency signals, hydrophones
do not seem appropriate; the only feasible sensing technique seems to
be thru a three component high frequency geophone tightly coupled with
the rock mass deep into a borehole. Low frequency signals are expected
only if large amounts of fluid are injected.
Acknowledgements.
This work was partially funded under EEC contract number 100-76
EGF and INAG, ATP Geothermie. I would like to express my sincere grati-
tude to J.Audric and C.Vial Ie for their help In this unrewarding
experiment.
BIB LI OGRAPHY •
AKI K., M.FEHLER and S.A.DAS ; 1977 "Source mechanism of volcanic tremor:
fluid driven crack models and their application to the 1963 kilauea
eruption". Jou.Volcanology and Geothermal Research; Vol.2 p.259.
BARENBLATI G.I. ; 1962 ; "The mathematical theory of equi I ibrium cracks
In brittle fracture" Advanced appl. Mech. Vol.7, p.55, academic Press.
CORNET F.H. ; 1978 ; "Appl icatlon de la fracturation hydraul ique a la
mesure in-situ des contraintes" ; colloque sclentifique J .Goguel.
B.R.G.M. memoire n091.
CORNET F.H. ; 1979 ; "Comparative Analysis by the displacement disconti-
nuity method of two energy criteria of fracture" ; Jou. of app. Mech.,
Vol.46 nb.2 p.349.
GRIFFITH A.A. ; 1921 ; "The phenomenon of rupture and flow In solids" ;
Phi I. Trans. Roy. Soc. London. Vol. A 221.
HAIMSON B. and C.FAIRHURST ; 1969 ; "In situ stress determination at greet
depth by means of hydraulic fracturing", 11th Symp. on Rock Mech.,
Berkeley. Somerton editor, p.559.
-975 -
HUBBERT M.K. and D.G.WILLIS ; 1957 ; "Mechanics of hydraul ic fracturing
Petrol. Trans.A.I.M.E., Vol.210, p.153.
IRWIN G.R. ; 1957 ; "Analysis of stresses and strains near the end of
a crack traversing a plate" ; Jou. App. Mach., Vol 24, p.361.
POWER D.V., SCHUSTER C.L., HAY R. and J.TWOMBLY ; 1976 ; Jou. Petrol.
Techn., p.1116.
-976 -
PERMEABILITY STIMULATION OF CRYSTALLINE ROCKS BY CHEMICAL LEACHING
ABSTRACT
-977 -
-1he feasibility of the geothermal process in dry rock was partially
demonstrated by the first pilot test performed at the FENTON HILL site, near
LOS ALAMOS. The development of the method requires a sufficient flow rate
to be maintained unde~ reasonable pressure through the fractures linking
the injection and production wells. The permeability of these fractures can
be ensured by a propping agent if the downhole temperature were not so high.
Since a propping agent cannot be used, leaching appears to be the only
solution.
1 - CHOICE OF ROCKS
- quartz,
- alkaline feldspars,
- calcosodic feldspars.
2 - EXPERIMENTS
3 - PRELIMINARY TESTS
Rock plates were placed in each chemical solution under pressure and
temperature. The weight loss and permeability were then measured. The
results obtained with Na 2C0 3 , HCl and NaOH are shown in Table 1. Sodium
hydroxide was the only one that led to both a weight loss and increase in
-978-
permeability at the same time. Other tests also showed that the combining
of two of these reactants can produce a considerable increase in weight loss
and permeability.
2
0.12 mg/hr/cm
2
0.02 mg/hr/cm
2
0.011 mg/hr/cm
The weight losses obtained with the different rocks are shown in
Figure 8. It is interesting to compare this figure with the modal
-979 -
classification of granitoids (Fig. I). The weight loss does not seem to
depend on the quartz content alone but also on the proportion of alkaline
and calcosodic feldspars. Therefore with and identical quartz content the
weight loss increases with the percentage of alkaline feldspars, as is shown
by the comparison of LANDEDA with PENARCREACH as well as that of PORSPODER
with LAFORGE and FLAMANVILLE. This means that the effectiveness of sodium-
hydroxide treatment should be increased in alkaline and suba1kaline granites.
In addition, with just a few reference rocks it should be possible to pre-
dict the weight loss obtained in a new rock with a known composition.
5.1.- Description
The solutions gathered were also analyzed, and their pH was measured.
During the first tests, the permeability of the rock matrix was
-~-
measured at room temperature and at 200°C. The permeability at 200°C was
found to be systematically lower. Typical values for an effective stress
of 20 bars were : 2 x 10- 7 darcy at room temperature, and 2 x 10- 9 darcy
at 200°C. The same decrease in permeability with an increase in tempera-
ture can be seen for samples separated in two by a fracture.
5.4.- Permeabilities
-982-
The solution must be injected in the space of several hours. In as
much as possible, this injection must be followed by that of salt water. The
injection of 100 m3 of 10% sodium-hydroxide solution into a formation at
200°C would dissolve from one to two tons of rock, depending on the alcohol
content.
CONCLUSIONS
-983 -
~
PRELIMINARY TESTS
loo'C - 100 bars - 144 hour.
TABLE II
Na2 C0 3 0.3 'f. 6.10- 7 10- 6 NaOH molar Ethanol/soda Maximum Activity , I
'" '" . Concentration molar ratio permee.bi Ii ty mg Si02 /hr/cm2 I
....'" :z: Darcy
~ HCl 3.1 'f. 10-6 -6
. 1.7.10
:z: ...
II< 0.5
'" 6.8 2.10- 2 9
Na.OH 10-6 1.4.10-4
5.14 " 2.5 0 1.10- 1 25
2.5 0.8 -1
0.5.10 21
Na2 C0 3 0.14 'f. 2.4.10- 6 5.6.10- 6
2.5 1.36 -1
~ 1Qd.
I fIl
~ HCl 5.9 'f. 1.2.10- 6 10- 6 2.5 1.78 1.10- 1 20
a!
'"~ 2.5 2.47 7.10- 2 12
.. ~
....
NaOH 10- 6 10- 3 2.5 3 4.10- 2
. 4.2 " 8
0
...
Na2 C0 3 0.14 'f. 9.10- 8 10-7
0
'"
:z:
~
~ HCl 10-7
.'" 5,5 'f. 5.10- 7
/v
~blre
PIo9 .. Colae plOV iodolt
Anof"lhhe
~n ;s ~norcreoch 40 /
o Lan ... londedo
For- • Por-spoder
L.F • La Forve
Aa :;: FIomonvilie
A,I , ;; AbM Hdul
/
30
-" --
V
20
AUTOCLAVE TESTS
AClI.leOUlI No OH
144 hI"" - IOO·C - 100 bars
/ POf'lpoder GranodiorITe
/
10
o 10 20 30 40
Soda conCentrglHln in weight %
/
50 50
AIZOJ AcHY'''Y I<z 0 Actlviry
(mg AI20., to-'Jly/cm 2 ) CmO KZO a 10·3/hr/cm Z )
7 V
40 40
/ /
30 30
/ V
20 20
AUTOCLAVE TESTS AUTOC L AVE TESTS
Aqueoul No OH Aqueous No OH
V
Porspoder Granod lorit. Porspocler Qranod iorile
10 10
o 10 20 30 40 o 10 20 30 40
Soda conCentration In weight % Sodo c.oocenfrOI.on In welCJhl 'Y.
Fig . 3 Fig . 4
-985 -
50
Si 02 Adivily AI 2 0 3 "" I.ll y
(mg 5;02.10-2 IlY Ian 2 ) Img AI 2 0 3 .10-3..."",2)
40 20
30 5
V
20
/ 10
1\
/ ~
AUTOCLAVE TESTS AUTOCLAVE TESTS
10 % Hydroolc:oholic No OH 10% HydroolCOl'lOlIC ~OH
---
10 5
~
-.:r
o 2 3 4 o 2 3 4
EthonoU sodo molor rolio Ethanol I $00(1 molar 1"'0110
Fig. 5 Fig. 6
Quortz conlenl
/v
0' rock
40
OLANOV' PENARCREA'"
"'''''4.-...jII''-.
PI.Phllll!"~. 30
/
~
,... ....... wn.t ..... __
"'~. '1 .
~.'"
LA:'OFIGE
6PII.,.._ ....
iII'II
f~MANV/ POASPOOER
20
ZLDur
10
AUTOCLAVE TESTS
10% H)'droolconoUc Na OH
144 hr - 100· C - 100 bors
o
i Elhonol/SOda molor rotlo: 3.43
10 20 30 40
We ioM loss in %
Fig . 8
- 986-
a> ~ G>
~5~
:i1
Q 210.1
.'"
FIO.9 e • 200'C - p. 155 bars
LE ACH I NG TEST o dl$l,lIed woler
10% hydroaleoholic No OH @ 10% aqueous No OH
e • 200'C - p: ISS bors Gl 10011 Na CI solutoo
0->
Elhanoll 5Odo motor rollO I: 1, 37 o town waler
PQRSPODER GRANOOIORITE PORSPOOER GRANODIORITE
510'"
2.10·'
o 200 400 600 800 1000 1200 1«Xl 1600 1800 2000 2200
1()-I' I I I
Volum. Inj.cted. in tJn3
500 1000 1500
Volum. '"j.cred • In cm:)
,
00
'"..., ~I,-------~------,----------------
FlO. 11 510, 0"..1t, LEACHING TESTS
(moSI021t1t/cm ZI IO~. Hydroolcoholic No OH
STUDY OF HEAT CARRIERS
EVOLUTION OF THE ~ PH AFTER ALKALINE
e· 200'C - p. 155 bar.
LE40UNG PorspocJer GronOdicriht
e • 200'C - p. 155 bars
>:
Q.
<D .... by passing a 10911 sodium chloride solution
®...... by pouinO woler
tol '%;: I
1
0 ',- 2 3 4
Elhonol/ ..odo molol'roM
5 FlO 12
~
/ ~'"
~
LEACHING TESTS
to % hydroalcoholic No OH
e =200°C - P =155 bars
Porspoder Granodiorite
i
1,3 2 3
Ethanol/soda molar ratio
-988 -
HEAT FRACTURING PHYSICAL MODEL
Abstract
The stability of a deep fracture under the influence of thermal fac-
tors is a controversial question. Although theoretical approaches exist,
there is no doubt that the physico-chemical complexity of this problem
leaves wide scope for an experimental approach. The purpose of this study
is to perfect test apparatus for rock samples which will be placed as close
as possible to the conditions encountered in the neighbourhood of a frac-
ture located at a great depth and destined to be used as an exchanger in a
hot dry rock environment. This contribution gives a detailed description of
this test apparatus and the main perfomances to be expected of it.
-989 -
I. INTRODUCTION
Test apparatus is being perfected to enable a rock sample to be ex-
posed to conditions of stress and temperature close to those that exist
in the neighbourhood of a deep geothermal exchanger in hot dry rock. The
purpose is to make an experimental contribution to the understanding of
the thermomechanical process. Does heat fracturing occur?
-m-
COOLING
liN 1'1'
BIAXIAL CELL
IN THERMOSTATICALLY
CONTROLED HOUSING I I
I I
: I I
I
I Po
REG ULATION
L.
Figure I RECORDING
-991-
3.2 Peripheral apparatus
3.2. 1 P_I!..t:.t_i..nJL1:.h_e.._~a_t:..e..r__I!..~<!.e..r
.p..r..e..s..s_u..r..e__a..n_<!.'p_u.,!D.PJ_I!lt_t_~_o_~1!._t:..h_e__s..~
%)_e~Water runs under pressure in cir-
cuit A (Figure I) by means of a URAKA
KE 201 pump fitted with a safety val-
ve, manometer with stopcock and a
tank to absorb water hammer and regu-
late pressure. A forged steel NOR I
globe valve placed at the end of the
circuit allows the flow to be regula-
ted, delivering directly to the tank
under head above the pump. This tank
enables the circuit to be cleared of
any solid particles that may have
been stripped from the sample. Sampling
of the water is carried out from this
tank for the analysis of the water in
function of time and to detect any
water-rock chemical reaction during
Figure III
the test.
3 . 2•2 J>..r_eJlJlJl..r~_.9_n__t.h_e__~~t_e..r_n_aJ.. J_1!.~JULt_tJl.~_EL¥I...2.l..e
-992 -
The oil used in
the biaxial unit Accumulo tOf'
18l1k
to put the exter- H
nal surfaces of
E0 oto300
Ga<l<Je
Figur e IV Va 20F
~~ a;; C
5. PROSPECTIVE USES
The scope for experiment provided by this test apparatus is thus very
wide. The biaxial unit is designed to give a parametric answer to the pro-
blem of heat fracturing.
-993 -
ERFAHRUNGEN MIT FRACBEHANDLUNGEN IN TIEFEN,
GERINGPERI~EABLEN GASLAGERSTJtTTEN
Sununary
- 994-
EINLEITUNG
LABORUNTERSUCHUNGEN
1. Petrographi e
2. Petrophysik
3. Physik der StUtzmittel.
Als Beispiel fUr den starken EinfluB der Kompaktion auf die Permea-
bilitat mogen die in Abb. 3 dargestellten Messungen an trockenen
Kernproben aus einer Karbonlagerstatte dienen. Wahrend bei 10 bar
MeBdruck die Luftpermeabilitat zwischen 0,136 und 2,44 md lagen,
wurden bei 450 bar Luftpermeabilitaten gemessen, die urn ein Viel-
faches darunter lagen. Xhnliche Permeabilitatsreduktionen konnten
anhand von Uber 50 Proben auch aus anderen Karbonbohrungen nach-
gewiesen werden. Da die "insitu"-Permeabilitat fUr das Design einer
Behandlung von entscheidender Bedeutung ist, jedoch bei Druckauf-
baukurven selten die effektive Permeabilitat eines einzelnen Sandes
bestimmt werden kann, sind bei der Beurteilung der Behandlungs-
moglichkeiten Kernmessungen unter BerUcksichtigung der Kompaktion
verwertbar.
-996 -
Die physikalischen Eigenschaften von StUtzmitteln spielen bei ber
Planung von Fracbehandlungen eine entscheidende Rolle. Untersuchungen
an verschiedenen StUtzmittelarten verdeutlichen wie in Abb. 4 dar-
gestellt, daB gerade in tiefen Bohrungen Bauxit und Zirkonium dem in
flachen Bohrungen gebrauchlichen Ottawa-Sand stark Uberlegen sind.
Das bei groBen Teufen, d.h. hohen OberlagerungsdrUcken, hier zwischen
420 und 600 bar verursachte Zerbrechen der Proppants bzw. ihre Ein-
bettung in das Gestein bewirkt ein teilweises SchlieBen bzw. Ver-
stopfen des Fracspaltes und fUhrt damit zu einer erheblichen Permea-
bilitatsreduktion.
Die Frage nach dem Einsatz eines geeigneten StUtzmittels ist somit
nicht nur eine Kostenfrage geworden. sondern hangt wesentlich von
den teufenabhangigen Erfordernissen, wie Gesteinsbelastbarkeit und
Druckfestigkeit des jeweiligen StUtzmittels abo
FELDPROJEKTE
Anfang 1977 fUhrte die BEB gemeinsam mit flobil Oil im SUdoldenburg-
Konsortium erstmalig in der Bundesrepublik einen GroBfrac auf das
tiefliegende Oberkarbon der Bohrung Goldenstedt Z 7 durch. nachdem
ausgiebige Voruntersuchungen, die sich Uber 1 1/2 Jahre erstreckten,
vorausgegangen waren. Seither sind weitere 8 r,1HF-Behandlungen durch-
gefUhrt worden.
PETROPHYSlKALISCHE DATEN
-997 -
BEHANDLUNGSDATEN
Es wurden die StUtzmittel mit max. Konzentrationen von 720 gil einge-
preBt. Die Raten lagen bei 2 - 2,6 m3/f.1in. bei EinpreBdrUcken von
360 - 660 bar.
-998 -
Die Fracgradienten lagen zwischen 1,83 und 2,26 bar pro 10 m.
TESTERGEBNISSE
Wie auf Abb. 6 erkenntlich, waren die Forderraten vor den Frac-
behandlungen z.T. auBerst gering. Die Bohrung Goldenstedt Z 7 er-
brachte einen ZufluB von 400 m3/h bei 1 bar. In einigen Fallen
konnten erst nach den Break Down Testen etwas hohere GaszuflUsse
erzielt werden. Die durch die Fracbehandlungen erzielten Kapa-
zitatssteigerungen liegen beim 5 - 10-fachen der Ursprungsrate.
Das nachste Bild, Abb. 7 zeigt das typische Logprofil einer Folge
vO,n Karbonsanden in der Bohrung Goldenstedt Z 7, von denen zwei
Sande nacheinander einer Fracbehandlung unterzogen wurden. Es
wurden in Abstanden von 20 Monaten zwei Produktionskontrollmessungen
nach der Behandlung durchgefUhrt, die zeigen, daB der ZufluB im
wesentlichen aus zwei relativ kurzen Intervallen der oberen und
unteren Perforation erfolgt. Des weiteren zeigt das Bild, daB das
Produktionsprofil Uber die bisherige Forderzeit konstant geblieben
ist. Diese Feststellung ist eine wichtige Voraussetzung fUr eine
Lagerstattensimulation.
FORDERPROGNOSEN
BEHANDLUNGSKOSTEN
Das nachste Bi1d (Abb. 10) zeigt Ihnen die Behandlungskosten, die
wir bisher aufgewendet haben. Man kann davon ausgehen, daB pro
Job die Kosten zwischen 0,7 und 1 t~io m" liegen. Hinzuzurechnen sind
natUrlich noch die Kosten fUr die Bohrung selbst, als auch die not-
wendigen obertag;gen Einr;chtungen, was alles zusammengenommen etwa
12 Mio OM betragt.
-wru-
kommt den MeBverfahren zur Bestimmung der Fracorientierung zu; diese
Oberlegung wird mit einem eingehenden Studium der Tektonik verknUpft
werden mUssen. Ausgehend von gesicherten Eingabedaten wird noch eine
erhebliche Entwicklungsarbeit an Simulationsmodellen erwartet.
-1002 -
Abb.1a DUnnschliff-Sandstein Goldenstedt Abb.1b DUnnschliff-Kaolin Aggregat
(x Nikol, 40:1) Goldenstedt Z 7 (XNikol, 200:1)
- 1003 -
'"'"
~ '~
...'"
~
ou
-....
E
a: ~ a: =>
e
'"
<0
~o::
Z~
-g
e
N "'Ill
.... UJ
~
!::
'",,;.
~
N "'a:
~
'"'"
- M' ~ a: a.. ffiz
u
:::>
a:
N - W
.... '"
Za: ~~~
Z
cr :::>", ~;::!:!
°'" 0
0
,.... ""'<0 cri:S:
-
C>
a: -0 - ~2 WILUJ
Q...1La:
iii ~ .....
W (/)
C) 00 a:Z .
UJ:::>
'"
C>
Z
:~w
"'''l~
>Wa:
to
ILC>
f- tii
a: Z
'"
.... 00..
UJ
> d- :4 ~ ~l:lz
;::
~
!::z -
..... 0 ~3~
""
UJ
IL
-
<OUJ
<1<0
mILa:
<z=>
IL
::EO ww~
UJ - UJ a:
~.
a: <0 ~ UJ
~-
~
UJ", cr i)
0.. ..... w :
..
Q...
... - i--=--
~
E~
~
~
~
~
~
~
~ ~ ;;;
...
~
~
~.
~~
..
!:I't8 OS7 NOli >ton!:lo 1'13N13 138 l::1nl !:I!1::1 1'!f1l118't3I'1!:13d
SI
Abb. 3
40
DURCHLAsSIGKEIT (DARCY)
1000~m~Bm~=~~
BEREICH DES ::t:::::t::=:
OBERLAGERUNGSDRUCKES ~
__
+-+-+-+-+~~~~-+-+-'-1KA~-~N
\ '3)
Abb. 6
- 1005 -
GOLDENSTECT Z 7
UPPER CARBONIFEROUS
PRODUKTIONS -
PROFIL
50%
8/79 \
~~--,
1/78
l J
- 1006 -
GOLDENSTEDT Z7
PRODUKTION NACH FlAC
II J\
-.. \
~II(
"
z
~ r\
~ ~
0: 0:
U. U.
+ +
J
If , 11 'I , . II I II III I U ru If , ~ ~ ~ II I U W I U III I' , VI YU lUI II I 11 III
Abb. 8
:.0::
U
::J-
a::: a::: OHNE
0« KOMPRESSION
l1.1D
a..-
o
:.0::
10000
a::: 8000
~r SEP JEMJR
1979
19)8
~) ~ULI
:z: MARZ
...::f:
...... 1979
"
6000
--
~
........
-
a:::
a::: 4000
r-- r-- ..................10..... GESAMT
r--
w
o. (OBERER·UN1ERER
a::: I- SANO )
:0
l1. 2000 I\. ......... OBERER SAND
UNTERER SAND
o
o 2 4 6 8 10 12 14 16 18 20
ZEIT (JAHRE)
Abb. 9
-1007 -
KOSTEN FUR MHF- BEHANDLUNGEN
(Mio OM)
Fig. 10
SCHWERPUNKTE
DER
ZUKUNFTIGEN ENTWICKLUNGSARBEITEN
• FRACFLUSSIGKEITEN
• STUTZM ITTEL
• RECHENVERFAHREN
• MESSVERFAHREN
• TECHNOLOGIE
• WIEDERHOLUNGSBEHANDLUNGEN
Fig. 11
- 1008-
THE ENHANCEMENT OF THE PERMEABILITY OF GRANITE
BY EXPLOSIVE AND HYDRAULIC FRACTURING.
Summary
Geothermal energy is potentially available from rock at elevated
temperatures in the Earth's crust but which lacks circulating fluids
to transport the energy to the surface. This resource has become
known as the Hot Dry Rock energy source. Two al ternative
extraction concepts have been proposed and this paper detai Is the
results of a field experiment designed to stimulate many fractures
at the wellbore. Unfocused explosive charges were used to produce
optimum radial fracture patterns that could then be enhanced by
hydraulic stimulation. The resulting system is extremely complex
but the results indicate that this combination of fracturing
techniques is essential to overcome the strength and impermeability
of the granite fabric.
- 1009-
1. INTRODUCTION
The successful exploitation of geothermal energy from hot, dry rock
sources will have to extract the heat from a volume of 3.6 million
cubic metres of rock to support a 100MWt extraction for 25 years,
cool ing this volume by 1000 C. Simple models, (1) , show that
extraction ratios of 1%-2% could be achieved and the impl ication is
that a volume of 360 mi II ion cubic metres would need to be
connected into the circulation system.
Single, plane fracture reservoir geometries use the concept of
cooling the fracture surface by circulating a cool fluid and,
thereby, extracting the heat. A 25 year circulation would cause a
one-degree temperature drop at 35m from the surface, therefore, the
area associated with the extraction would have to be 5.2 mi II ion
square metres or the equivalent of 18 fractures, 300m in radius, (2).
An alternative geometry of a highly fractured sphere of rock, 160m
in radius, could support an extraction ratio of 20% for the same
power rating. These two geometries are extreme cases and the likely
form of a reservoir will have some features of each type.
The work completed during Phase of the EEC Hot Dry Rock
Programme examined various facets of the problems associated with
stimulating a fractured volume between two wells in an attempt to
make maximum use of the existing natural fractures. The work was
undertaken in 4 wells, 300m deep and 150mm in radius; these are
located in the Carnmenell is Granite pluton which forms part of the
Cornubian Bathol ith of South West England. This area has a heat
flow of 100-130mW/m 2 , (3).
2. THE RESEARCH OBJECTIVES
The well bore is the sole component in the HDR System that can be
directly affected by the stimulation engineer. This impl ies that the
stimulation procedures must be such that they do not impede the
growth of a large swept area, low impedance heat exchange system.
The objective of the research programme was to demonstrate that the
destruction of the well bore stress concentration by decoupled
explosives would facilitate the growth of a multifractured reservoir
from a predetermined point. This should permit individual flowing
systems to be stimulated in the desired positions.
Conventional hydraul ic fracturing theory assumes that the fracture
volume is glven by subtracting fluid lost to the formation from
- 1010-
fluid injected. In granite, with its permeability of nanodarcies, the
fluid lost to the matrix is essentially zero so all the fluid is
involved in the stimulation process. In the I imit, the volume stored
is simply the volumetric compression of the individual blocks. The
required volume of fracturing fluid is given by evaluating the
volumetric strains associated with a given pressure and then, using
the desired intersection distance as the radius of an equivalent
cylinder, the affected volume can be evaluated.
It is essential to ensure that the volume is injected from the
wellbore in as uniform a manner as possible; this means that a
major set of axial fractures is necessary. These must be emplaced
prior to the hydraul ic stimulation and would be, at least, 4
borehole radii in length. The obvious choice for this pre-fracture
treatment is a procedure that produces tangential tension at strain
rates in excess of fracture energy dissipation rates and the ideal
solution is explosive fracturing with carefully designed charges.
The temperature problem is readily overcome with modern detonators
and explosives, (6).
The remainder of the paper presents the results of an experiment
designed to demonstrate the validity of the above arguments.
3. THE CONCEPT AND THE THEORIES
It has been demonstrated that hydraul ic fracturing in massive
impermeable rock will produce one or two axial fractures over the
stimulated length of hole, (4). Hypothetically, in an unconfined
joint-free formation the fracture grows ell iptically unti I it passes
the packer and it can be assumed that the length of the stimulated
zone must exceed the desired fracture diameter. Furthermore, any
one 50MWt doublet system may contain 8 or 9 individual fractured
zones, each producing 6-7MWt at a flowrate of 10-15 kg/s
(4Bbl/min). The flow velocity into anyone fracture O.lmm wide and
600m long will be approximately 0.2m/s which would give
sufficiently low Reynolds numbers to ensure a low pressure loss at
the well bore.
Unfortunately, the Cornubian Granite is known to contain marked
discontinuities at I-2m spacing to depths of over 800m and there is
no evidence to indicate that these disappear with depth. The aver-
age length of intersection of these joints with a 24 0 incl ined
borehole is seven metres. These discontinuities form zero tensile
- 1011 -
strength planes and will be opened when hydraulically pressurized
with water if the fluid pressure within the fissure exceeds the
normal stress across it. Many authors (e.g. (5) ) have shown that
high viscosity, high wetting angle fluids will not penetrate a
narrow, existing fissure so it is possible to fracture the wellbore
and override the fissures. However, these fluids degrade rapidly
and are not stable at production temperatures of 2000 C.
If a single flow exit opens in the fully packed off zone then the
fluid velocities could reach 30m/s leading to very high head losses
at that point. It is clear that a reliable method of generating
propped, axial fractures over the entire stimulated zone would be a
significant improvement on the very uncertain hydraulic stimulation
of fissured, unconfined and impermeable granite.
Away from the wellbore a multipl icity of fractures would form the
most certain method of connecting into the existing fissure system.
Ultimately, the in-situ stress field. will govern which of the Joints
will dominate the flowing paths and dictate the reservoir geometry.
4. EXPERIMENTAL PROGRAMME - OPERATIONS AND METHODS
Table One shows the logic of the experimental approach used to
study the problem. The following sections give the experimental
methods adopted for each task.
4.1 Drilling
4 boreholes, known as RH6A, RH8B, RH7C and RH9D, were drilled by
"down-the-hole" percussive air drilling to depths of approximately
300m. at a diameter of 150mm.
Three cores, 3m. long were taken from the last two holes. Two rigs
were used for this work and these are shown in Figure One.
Unfortunately, the last two holes were severely deviated.
4.2 Surveying
Eastman Whipstock were engaged to survey the holes by gyroscopic
and magnetic methods. Both of these methods gave identical results
but the magnetic tool was certainly easier to run. The tools did
show signs of being affected by running in open hole because of
the wall roughness and abrasivity of the granite.
4.3 Geophysical logging before stimulation
Seismograph Survey Ltd. (a division of Birdwell) were engaged to
run caliper, natural gamma, gamma-gamma density and various
lengths of sonic tool. The Institute of Geological Sciences ran
-1012 -
TABLE ONE PROGRAMME LOGIC - TASKS COMPLETED
TASK DESCRIPTION COMPLETED
1 Drill holes by percussive method Sept. 1978
2 Survey holes by Gyro and Magnetic Method Oct. 1978
3 Geophysical Logging before stimulation Oct. 1978
4 Hydraulic tests on in-situ conditions,
(a) Complete hole injection
(b) Drawdown interference
(c) Packed off lower 30m, low pressure
(d) Packed off lower 30m, high pressure April 1979
(e) Stress measurements
5 Hydrau I ic Fracture on Borehole RH7C Apri I 1979
6 Hydraulic tests after hydraulic fracturing
(a) Low Steady Pressure
(b) High Steady Pressure
(c) Pulse or slug tests
(d) Venting tests
(e) Shut-in decay
(f) Interference
(g) Radio-active tracers
(h) Long, circulation - 14 days -
between wells at 3kg/s. Nov. 1979
7 Explosive Stimulation of Borehole RH8B
(a) Emplacing charge
(b) Cleaning operation Oct. 1979
8 Post-Shot Geophysical logging Oct. 1979
9 Hydraulic tests on system (see Tasks 6,a,b,c) Nov. 1979
10 Hydraulic stimulation after shooting Jan. 1980
11 Hydraulic tests after stimulation
(see Tasks 6, a - g) Feb. 1980
12 Circulation at flow rates up to 10kg/s Mar. 1980
-1014 -
to measure the total volume of injected fluid and derive an average
flowrate for use in a superposition decay analysis.
Pulse or slug tests were also used by rapidly injecting 12 - 20
litres of fluid and closing the well to permit decay measurements to
be made. The system elasticity (pipe-work etc.) required careful
monitoring to ensure that accurate values were derived.
The actual pump hydraul ic schematic is shown in in Figure 2.
(e) Stress Measurements
Several measurements were made by Bochum University staff and the
remainder were run by Camborne personnel. The latter set were
made using a small triplex ram pump (0.15 kg/sec) running through
a turbine flowmeter assembly to the tubing installed in the well.
Only one straddle packer test was conducted with this equipment
because the mandrel fai led when pressures of over 300 Bar were
needed to fracture the rock. Great care was taken to select the test
horizons accurately to ensure no natural joints were in the test
zone. A pulse test was run at 5 MPa, followed by the fracturing
run itself. Pumping was then stopped to allow a shut in pressure
to be measured and then restarted to measure the "refrac" pressure.
4.5 The Full Hydraul ic Fracture of Borehole RH7C
The lower 30m. of RH7C was pressurized with the small pump until
breakdown occurred and, subsequently, a much larger pump (rated
at 3.5 kg/sec at 10 MPa) was used to expand the system.
Connections to RH8B and RH6A were obvious by the small return
flows. However, these wells were not equipped with necessary
hardware to isolate the lower sections of the hole. Repeated
injections to 10 MPa increased the specific acceptance of the well
but at no time could more than 2 kg/sec be injected at 10 MPa
(Specific Acceptance, Ymax = 0.2 m3/sec/GPa).
4.6 Hydraul ic Tests after Hydraul ic Fracturing RH7C
Test Series RT 204 - RT 222
(a) Constant Pressure below packer < 5 MPa
A run was made holding the pressure constant at less than 3 MPa
and measuring the injected and return flows whi Ie temperature and
conductivity logs were run in the return wells. A slug of sal ine
solution was run ~ 400 kg) but it was not detected during
logging. The injected flow was low ~ 0.1 kg/sec) and held
constant with time.
- 1015 -
(b) Constant Pressure below packer >6.0 MPa
Injections at up to 10 MPa were conducted at various flowrates up
to 1.8 kg/sec to ..examine the effects of repeated stimulation on the
system compressibility.
(c) Pulse tests
The hydraulic energy stored during an "instantaneous" inflation
dissipates as permeation loss occurs. The rate of loss can be
related to the various system parameters.
However, as the system grew, the response became meaningless
because of the slowness of inflation and the rapidity of loss of
pressure.
(d) Venting
The volume vented at any overpressure represents the volume stored
in the rock at that pressure. At values less than the effective
earth stress this volume is finite, but, above the earth stress, the
well head pressure represents the effective earth stress plus the
head losses into the reservoir. After several high flowrate injections
and large vents with more than 10 m3 of water the system near the
well responded at lower overpressures (7 MPa as opposed to 8 MPa).
(e) Shut-in Decay
Conventional injections and decays were used to examine the
applicability of the usual "diffusion" analyses methods (e.g. (10)).
The principal difficulty with "diffusion" is that the results are
only valid if the system properties are not changed by the
injection. This is not true when at least one effective earth stress
is reduced to zero or even becomes negative. The only way to apply
the tests is use low flows and low overpressures to ensure the
maximum pressures reached do not change the system. Unfortunately,
the impermeability of the rock is so low that even flows of .1kg/sec
reach high pressures before the "storage transients" are over. The
rock is so impermeable that no shut in pressures were identified.
(f) Interference T es ts
Typical injections and shut-ins were run whilst measuring the flows
and/or the wellhead pressures in the other wells. At the time of the
tests on RH7C the other wells could not be pressurized to more than
.3 MPa but, notwithstanding this limitation, the other wells did not
respond until the injection pressure exceeded 6.9MPa.
-1016 -
(g) Radioactive Tracer
A Br82 tracer was used to examine the system porosity and other
reservoir parameters. Unfortunately, the return flow had to be
throttled to less than 0.3 kg/s to permit the use of a prefabricated
counter system. The flow regime was quite unl ike the "normal"
conditions. However, the test did show the applicability of
available expertise to the problem.
(h) Long Circulation Tests (2 x 7 days)
Water was circulated at flow rates up to 2 kg/s and a complete
water balance was measured. The specific acceptance was so low
3
(.1 m /sec/GPa) that the injection well had to be pressurized to
over 10 MPa and fluid loss was still continuing at 1 m3 /hr after
300 hrs, approximately 16% of injected flow. However, at the end of
3
the test, nearly all the 600 m lost was recovered during 12 weeks
of return flow at 2-3 kg/min on each well.
It was clear that the specific acceptance must be raised by
3
stimulation methods to better than .4 m /sec/GPa to reduce these
losses to a small value.
4.7 Explosive Stimulation of Borehole RH8B
(a) Emplacing the charge
A 30m long charge of Submarine Blasting Gelatine weighing 38 kg
was fired at the base of the hole by suspending the charge on a
rope of synthetic fibres. The charge was contained by 60m of sand.
The instruments indicated that the shot fired correctly but, later,
it was discovered that it had not fired at a high velocity of
detonation.
(b) Cleaning the hole
The charge used booster PETN pell ets (30g) to ensure safe
detonation but, unfortunately, the special detonators were suppl ied
with the wrong charge weight of explosive due to a clerical error.
This left partially fired explosive with live boosters in the hole.
This was not discovered until 20m of the charge zone had been
cleaned using a workover rig.
The rig was released and high pressure water jetting at 10 kg/s
was used to clear the remaining debris. The estimated velocity of
detonation was less than 2500 m/s from experiments after the shot.
4.8 Post-Shot Geophysical Logging
Although the shot did not fire at its design point, significant
- 1017 -
changes to the lower 12m of the shot zone were indicated by the
sonic log. The cal iper tool showed the smallest of perturbations
through this zone. The T.V. log showed some wall damage and
3-5m. long axial fractures through this zone.
4.9 Post-Shot, Pre-fracturing Hydraul ic Tests
The steady pressure, build up and slug tests were used to examine
the post stimulation hydraul ic characteristics of the well.
4.10 Hydraulic Stimulation after Shooting
The well broke down at 8.4 MPa after failing to respond at 10 MPa
before shoot i ng. No shut-i n pressure was observed but the "refrac"
3
pressure was 6.3 MPa. 80 m of water was injected and vented 5
times to stimulate the well, see Figure 3.
4.11 Hydraul ic Tests after Stimulation
Many of sub tasks 6 (a) - (fl, Table One, were not repeated but a
selection of build-ups and decays were used to determine the
effectiveness of treating the well. A specific acceptance of .3
m 3/sec/GPa was achieved during this stage.
4.12 Circulation at Flow-Rates up to 10 kg/s
Twin Halliburton no pumps (150 kW each) were used to start
circulation at up to 10 kg/so These ran for four days and then
released in favour of our own HB 550 pump, rated at 50 kW, to
maintain a constant 2.7 kg/so The specific acceptance rose to better
than 0.5 m3/s/GPa and 100% returns were achieved. These results
are now presented in detail.
Each of Sub-Section 4.1 - 4.12 appears as a full report in the full
project report (ref. 12) but Sections 5 and 6 in this paper deal
mainly with 4.6 and 4.12.
5. ANALYSIS PROCEDURES
The previous section described the various tasks used to provide
data on the problem of reservoir stimulation. It is not practical to
present the results of every task relating to the research programme
but the essential features of, and the differences between, the
hydraul ic and explosive stimulations were highl ighted during the
two long circu lation experiments, RT 219/220 and RT 239/240. It is
these experiments that are presented in more detail. The analysis
methods are described below.
5.1 The Importance of Specific Acceptance
The entire system depends upon being able to inject large volumes
-1018-
of liquid at an inlet pressure just below the minimum effective
-1019 -
:. the flow rate to storage, is given by:- dVr
dr
where Vr stored volume
time
therefore dVr = QI - nQI = QI - ( ~PI - ~Pr)
dt I
but the volume in storage is a function of the mean reservoir
pressure because it is due to the compression of the reservoir rocks.
:. Vr = f (( ~P I + ~Pr )/2) = f (Pm)
dVr
--
=f (P
m
)
dP
m
:. d (Pm) = Q I· I - { ~P I ~Pr)
~--~I.~f-(~P~)~----
.
•• n
dt
=1 - __ ) • f (P )
d(P m_
m
m
dt
The results should show 100% recovery when d {P
m
)/dt = O. If other
well bores are connected to the system the value will tend to a
constant percentage.
5.6 Net Power Extracted in a Commercial System
Net Power = Thermal Power - Pump Power
Pe = n.OI.YCp. ~T - 01' (~PI - n ~Pr)
where
P Extracted Power
e
Y Mass density
Cp specific heat
~T tempera ture changes
An alternative form is given by:-
Pe = n.OI.Y.Cp. ~T - °2 1 ,( 1 . (1 -
2
n) + n .1)
Y
The term related to pump power is a function of the square of the
flow rate but notice that the specific acceptance is less important
in the analysis when n -1. Both parameters are highly dependent
on the return side back pressure.
6.COMPARING THE RESULTS OF HYDRAULIC AND EXPLOSIVE STIMULATION
Section 5 outlines the critical reservoir parameters that can be used
to compare the effectiveness of stimulation. Table Two is a summary
of the analyses that are presented graphically.
-1020 -
TABLE TWO
- 1024-
9. REFERENCES
1. Batchelor, A.S., Pearson, C.M.P; 1979: "Preliminary Studies of
Dry Rock Geothermal Exploitation in South West
England," Trans • B. Inst. Min. Met., Vol. 88, pp. B51
- B56.
2. Cummings, R.G., Morris, G.E; 1979: "Economic Modell ing of
Electricity Production from Hot Dry, Rock Geothermal
Reservoirs: Methodology and Analyses," EPRI EA-630
RP1017, Final Report September 1979.
3. Wheildon, J., Francis, M.F., Thomas-Betts, A', 1977:
"Investigation of the SW England Thermal Anomaly Zone,"
In Seminar on Geothermal Energy, Brussels, December
1977, Vol. 1 (Luxembourg: Commission of the European
Community, 1977) pp. 175 - 88 (EUR. 5920)
4. Haimson, B.C., 1978; "The Hydrofracturing Stress Measuring
Method and Recent Field Results," Int. J. Rock. Mech.
Min. Sci. and Geomech. Abstrs. Vol. 15, pp. 167 - 178.
5. Zobach, M.D., Rummel, F., Jung, R., Raylelgh, C.B; 1977:
"Laboratory Hydraulic Fracturing Experiments in Intact
and Pre- fractured Rock" Int. J. Rock Mech. Min. Sci.
Geomech. Abstr., Vol. 14 p. 49 - 58.
6. Patterson, W.W., Deam, D.R., Macdonald, H.J., Rochester, R.H;
1979: "Capacitive Discharge Firing System for Providing
Acoustic Sources in the Hot Dry Rock Geothermal Energy
Development Project," LA-7761-MS, UC-66a, Los Alamos
Scientific Laboratory.
7. Ramey, H.J., Earlougher, R.C., Kersch, K.M; 1973: "Well bore
Effects in Injection Well Testing" Jnl. Pet. Tech., Nov.
1973, p. 1244 - 1251.
8. Black, J., 1979: "Results of a Multiple Borehole Pumping Test in
Low Permeab iii ty Granite" OECD Conference on Lowly
Permeable Rocks, Paris, March, 1979.
9. Tester, J.W., Albright, J.N., 1979: "Hot Dry Rock Energy
Extraction Field Test: 75 Days of Operation of a
Prototype Reservoir at Fenton Hill, Segment 2 of Phase
1" LA-7771-MS, UC-66a, Los Alamos Scientific Laboratory.
10. Earlougher, R.C; 1977: "Advances in Well Test Analysis,"
Monograph Series, Soc. Petro Engrs. of AIME, Dallas
- 1025 -
11. Myung, J. I., Helander, D.P; 1973: "Borehole Investigation of
Rock Qual ity and Deformation Using the 3-D Veloci ty
Log" 4th Int. Symp. on Salt, AstroWorld Houston, Texas,
April 8 - 12, 1973.
12. Batchelor, A.S., Pearson, C.M., Ha"aday, N; 1980: "Project
Report on Permeability Enhancement Studies Relating to
Geothermal Extraction in South West England" Covering
EEC-497-78-1 EGUK, SRC-GR/A/27971, U.K. D of E
E/SA/Con. 111, Camborne School of Mines, U. K. (I n Press)
- 1026-
APPENDIX
Selecting Packer Horizons From Geophysical Logs in Granite
Fracturing, and other high-pressure injections with packers set in
open hole, is one of the least reliable downhole operations that can
be attempted. However, the strength of granite is such that packers
can be anchored successfully provided that there is no wallcake
from the drilling fluids and no high angle discontinuities pass the
packer horizon.
(a) The Caliper Log
No washouts occur in granite and the hole is usually within a few
mill imetres of gauge. However, discontinuities running through the
hole tend to break up at the wall causing slight increases in
gauge. They are readily spotted and can be avoided even with a
normal caliper. A multi-independent arm tool is even easier to
interpret but it is not vital.
(b) The Density Log
In regions with no caliper anomaly it is still possible to have
density values indicating lower than average densities. These zones
are avoided because they may indicate the proximity of a fracture
parallel to the hole axis. A zone of uniform, average density for
twice the packer length is required.
(c) The Total Natural Gamma Log
Zones of uniform average countrate are another feature confirming
the absence of discontinuities in granite. Joints with any alteration
tend to be depleted in Uranium if they have hy.drothermal alteration
or enhanced in Uranium if ground water has leached the Urananite
near the fracture. Alternatively, the potassium and thorium will
tend to be in higher concentrations in small intrusive dykes.
Consistent uniformity is a sign that the granite is unfractured.
(d) The Sonic Logs
Finally, the lack of attenuation and disturbance in the shear and
compressional wave arrivals will confirm that no fissures are within
.5m - 1m (see Ref. (11) ).
-1027 -
- 1028-
e.Y.
Filtered 'nlet Water
•
Chriahe 2S I
ChOl'9o Tonk Accl.!mulotor
400 I. (C .. e.g ot
2.1 MPo) NY
50 mm Inlrt
"BY
BY
Pur- p PC
MOWltod on Pump RI9
o
-
'C
'" NBY
J2 mm Hydraulic Hose
BV - e.H-type laolator Valves
SV - Saf.,)" Valve Set ,lot 10 MPa
NRV - Non-Return Valve 50 mIT', PIPit'. Down Hole
I'V - Neo}dle 'J.lve
P:; _ Snubbed F1uld-D.mp~d Pr~aSure
Cauge-.
SEPT - Sur(oct Pr~ssure Transducer
F''(O - Pock_oil Vent Tank
£3;-' _ a.1l .nd Howell Downhole Tranaducc 1600 I
10 20 30 40 50 5 6 7 B 9
VOLUME, INJECTED (LiTRES) INJECTION WELLHEAD PRESSURE RHBB (MPa)
...,c
<:>
FIGURE 5 IMPEDANCE AGAINST RETURN SIDE (RH6A) FIGURE 6 EFFICIENCY AGAINST WELLHEAD PRESSURE
3J "- WELLHEAD PRESSURE I RH6A. RT240
<J)
<'l'
e2
~
w
~1
i
2 3 4 5 6 7 B 9 ro 23456 7 B 9 ro
RETURN SIDE WELLHEAD PRESSURE (RH6A) (MPa) RETURN SIDE WELLHEAD PRESSURE RH6A (MPa)
12
w
0:
;;)
VI
VI
w
__+-______ ~~~~~~AN~~ 0:
0-
4 c<I:
CONTfK)U..EO IRTO 12Mh
w
J:
....J
....J
RESIJl.TS FOIf R7220 ' WELLHEAD PRESSURE 2 W
/ CONSlANT I TOS#05MhI 3
w ~ 00 ~ m 00 ~ ~ ~ 200
ELAPSED HOURS
w ~ 00 00 m 00 ~ ~ ~ ~
ELAPSED HOURS
3 100
0
I(
~8Q
w
r
HVORAUI..IC FR~CrUl1£D RESERVOIR
~6Q RT220 AVERAGE INJECTED FL(N.I RATE 137 Kg /s
~ WELLHEAD PRESSURE '0 MPil
~4O
0:
w
e5 w
u
w
0:
- 1031 -
ANALYSIS OF HYDRAULIC FRACTURE PROPAGATION
A FIELD EXPERIMENTATION
F.H.Cornet.
Abstract.
Numerical results and laboratory work on hydraulic fracture propa-
gation is first briefly recal led.lt is shown that fracture orientation
depends upon in-situ stress state, injection flowrate and rock mass
resistance anisotropy.
Field results are presented which illustrate the applicability
of these results to granitic rock masses. A new stress measuring method
based on shut-in pressure measurements conducted for fractures develop-
ped in various orientations, is presented j dispersion of results deri-
ved from 8 fractures is within 3 bars. This stress determination has
been used to compute the orientation of a propped hydraulic fracture
away from the injection wei I. Intersection of this fracture with a
second borehole, drilled 30 m away from the first one, demonstrated
the accuracy of the determination.
-1032-
I.INTRODUCTION
It is usually accepted that hydraulic fractures propagate perpen-
dicularly to the least principal stress direction (e.g. Hubbert and
Wi I I is 1957 , Scheidegger 1962, Haimson and Fairhurst 1969). However
numerical and laboratory investigations (Cornet 1978) have shown that
it is theoretically possible to develop fractures perpendicularly to the
major principal stress direction, at least for sma I I distances. The
purpose of this investigation is to determine whether such results are
still applicable to large scale test conditions so that multiple linking
between two boreholes could be achieved by hydraul ic fracturing techni-
ques. Further, a new stress determination technique based on the inter-
pretation of hydraul ic fracturing pressure data is presented; results
obtained in two 200 m deep boreholes are presented.
-1033 -
(4)
With this numerical model it has been found that, if the fluid pres-
sure is maintained uniform up to the crack tip, a fracture can propagate
perpendicularly to the maximum principal stress. However the fracturing
process is unstable (the strain energy release rate increases as the
crack grows) so that the pressure must be monotonically decreased if
quasistatic conditions are to be maintained. Because of this constant
pressure decrease, it was found that for long enough fractures, the an-
gular variation of the strain energy release rate becorres so small that
crack propagation is control led by local rock heterogeneities. If flow
rate is such that a pressure gradient develops, the numerical model indi-
cates that the fracture gets oriented perpendicularly to the minimum prin-
cipal stress. For fractures oriented at an angle with respect·to the
major principal stress direction, their propagation is found to be such
that they get perpendicular to the minimum principal stress.
Laboratory experiments conducted on smal I granite cubes (30 cm X
30 cm X 30 cm) corroborated these results. It was found that for a given
appl ied uniaxial load, a fracture innitiated perpendicularly to the ap-
pi ied load direction,propagated in its own direction when the pressuri-
zation rate was slow enough; that the fissure got al igned with the ap-
I ied load direction for pressurization rates faster than a critical
value, which depends on the uniaxial load magnitude and that a new frac-
ture developped in the direction of the appl ied load orientation for pres-
surization rates faster than a second critical value.
-1034 -
3. INFLUENCE OF THE STRESS FIELD
In order to investigate experimentally the feasibil ity of develop-
ping fractures perpendicularly to the maximum principal stress direction,
a tool combining, a Goodman Jack type instrument and a straddle packer
has been bui It. It has been designed to operate in shal low boreholes.
Theoretical solution for the state of stress and displacement
field caused by a curved jack loading device in an infinite borehole
dri I led in an elastic medium is wei I known (Jaegger and Cook 1971).
At the well bore wall
:x:~
orr
099
P
P
4a P
iT
for t"<'H"
1T-a<9<1T +a
(5)
orr 0 { a<9<7t-a
4aP 1T+a<9< 1T-a
099 for
1T
- 1035 -
of It). Four major fissure orientations have been identified
N300E ! 10°, N600E :!: 10°, NIOOoE:!: 10°, NI55°E:!: 5°.
AI I these fissures are subvertical (dip lies between 70 and 85°).
Thermal, electrical and video logging were used to identify fractured
zones. Their orientations, determined with the borehole TV camera were
found to be those identified by the surface mapping.
Eighthydraul ic fractures were generated with a straddle inflatable
packer in area thoug~to be homogeneous i results are presented in
Table I.
Table I : Data from hydraulic fracturing test at Le Mayet de Montagne
test site (25 km SE of Vichy, in the center of France) for borehole
INAG 3-2
Depth 27 42 54 65 84 90 174 186
m·
breakdown 22.3 frac. frac. frac. 10. 34.7 15.1 ~9.5
pressure by by by
3333.3 Packer Packer Packer
mPa
reopening
pressure * 5.4 5.9 - 8.5 5.1 - -
shut-in
pressure - 2.1 4.2 3.2 4.6 4.4 5.6 5.4
MPa
Injected
v3 1ume in - 4.88 1.19 4.58 2.54 3.59 .02 13.0
m
recovered
volume in 04 .24 • 12 .26 .14 - 2•
m3
-
Pumping 60 60 60 60 60 60 I 320
rate
I/min.
fracturing waier water gel water water water water gel
fluid
...
~ Pressure required to reopen the frac generated during the first InJec -
ti9n {measured after the pore pressure has dropped back to its original
va ue 1
-1036 -
It can be noticed that orientation of these fractures is quite
variable and that, for very shal low depth (27 and 42m) more than one
fracture occures. For the 27 m frac a first breakdown pressure was
measured immediately after permeability tests had been conducted (pore
pressure was not al lowed to return to its original value) ; after 200
liters have been injected, the well was left opened so as to let the
pore pressure return to its original value. Pumping was started again,
a higher breakdown pressure than the first one was recorded. Impression
taken with a packer, after the test, revealed two vertical fractures.
This effect is interpreted in a similar manner to that obtained in the
laboratory experiments: for the first frac, build up of pore pressure
al lowed a preexisting fissure to be reopened; for the second injection
the fast injection flow rate, as compared to fluid flow velocity in the
preexisting fissure, induced a new crack to be developped.
Thus, this variability in fracture orientation can be accounted
for only if anisotropy exists: surface energy for some specific di-
rections are much lower than for other directions. An extreme case is
given by a preexisting fissure for which the free surface energy is nul
partial cementation induces a slight increase in surface energy, complete
healing may stil I imply a lower surface energy value than that of the
intact rock. This anisotropy may coincide with that observed for the
elastic behavior (schists or gneiss for example), it may also be inde-
pendant of it (granite exhibit isotropic elastic response but the free
surface energy is often very much direction dependant as is well known
in granite qLarries).
A precise knowledge of main fracture orientations along with a
detailed determination of the rock "strength" anisotropy must be obtained
if fractures are to be developped in more than one direction; pressuri-
zation rate is also a significant parameter for this oriented fractur<rng
process. It is of interest to note that for the 174 m and 186 m deep
fractures, although they were oriented in the same direction, the dif-
ference in breakdown pressure Is much larger than what can be accounted
for by the depth difference. This difference may be attributed to a
property of the rock itself, it is much more I ikely that it is caused by
the difference in pressurization rates (flow rate of I Ilmln versus
320 Ilmin were used). This is confirmed by results obtained for the
fractures at 42, 54 and 65 m depth: the straddle packer was Inflated at
-1037 -
pressures of the order of 200 bj when injection was undertaken no break-
down pressure was recorded. Fracture examination conducted with the
borehole TV camera revealed that the fracture intersected one of the
packer on a 60 to 100 cm.1 ength. It is conc I uded that fracture was gene-
rated by the packer itself: the pressure appl ied by the packer, which
under fast flow rate conditions would not have induced any damage, was
sufficient to initiate fracture when appl ied for more than 30 minutes.
This is a clear demonstration that, as is wei I known (e.g. Haimson 1968)
the tensile strength of rock is rate dependant.
(7)
-1038 -
o
(8)
(~) ~ ( )
G1+ a(pgh 0 0
(~)
0 G2 +a~pgh 0
0 0 p(gh
For anisotropic materials, the superposition may not be as simple
since principle directions of both tensors may not coincide.
I n equation (9) four unknown must be determi ned: If. ,<fJ.' cI.. and'
(the angle between the magnetic north and the 07 direction) If ~ is the
azimuth of the normal to the fracture and n3 the vertical component of,
this normal, equation (9) can be rewritten
n2, \~
(~-') = 2r~
l-n
- pgh(a
_
----::;-]
l-n 2
( 10)
3 3
( " )
- 1039-
solved by a least square approach if more than three fracture orienta-
tions at more than two different depth are available. Error of the deter-
mination can thus be apraised.
-1040 -
Table 2 Comparison between computed and measured shut-in pressure in
INAG 3-2 borehole at Le Mayet de Montagne
54 4.3 4.2
65 3.5 3.3
-1041 -
A more detailed description of the various experiments conducted in this
dipole will be given elsewhere; it will only be mentioned here that a
40 days circulation experiment nas been conducted at the rate of 3 m3/h
and that fluid losses are of the order of 50 %.(see paper by Ledoux et al.
in 5alme vo I ume) •
Table 3 Comparison between computed and measured shut-In pressures in
INAG 3-3 borehole at Le Mayet de Montagne
CONCLUSION
Numerical, laboratory experiments and field tests have shown that
it is possible to develop hydraul ic fractures at an angle with respect
to the major principal stress direction. This oriented fracture propaga-
tion process depends on the In-situ stress state, the rock "strength"
anisotropy, the injection rate.
A new stress measuring technique, based on shut-In pressure measu-
rements for fractures at various depth in various azimuths, was proposed.
Measurements, obtained from 8 fractures developped at depth ranging from
50 m to 200 m in two boreholes distant from each other by 30 m yielded
a maximum di-sp~rsion of 3b (maximum error between computed and measured
sbut-in pressure).
Interpretation of shut-in pressure in terms of normal stress to the
fracture plane revealed also to be a powerful tool to determining fracture
orientation away from the wei I bore, once the stress state is known. Val~
dity of this interpretation was demonstrated by Intersecting a propped
fracture 30 m away from the injection well. If the role of depth on the
stress state is that described in o~r model, It becomes apparent that the
deviato.ric stress component remains unchanged with depth. If (01-02)
remains constant, results obtained at 200 m depth with respatto fracture
orientation should stil I be applicable at depth of the order of 3000-
4000 m. Itm.ay even be proposed 'hata!: the temperature an<l the spherical
-1042 -
component of stress increase, viscosity of the rock mass may become
more and more significant so that the maximum deviato\'\(. stress may
decrease with depth; as a consequence, chances of developping fractures
in chosen directions at depth should be better than in shal low conditions.
ACKNOWLEDGEMENT.
This work was partially funded by EEC contract number 564-78-1 EGF
and partially by INAG ATP Geothermie.
The thermal and Electrical loggings were conducted by Dr.Jol ivet
and Pro Mosnier. C.Vial Ie helped in the videologging. Pr • Drogue and Mr.
Gri I lot conducted the fracture mapping.
BIBL IOGRAPH IE
HUBBERT M.K. and D.G.WILLIS ; 1957 ; "Mechanics of hydraulic fracturing",
Petroleum transaction of AIME ; vol.210, pp 153-168.
SCHE IDEGGER A. E. ; 1962 ; "Stresses in earth's crust as determ i ned from
hydraulic fracturing data"; Geologie und Bauwesen, Vol.27, p.45.
HAIMSON B. and C.FAIRHURST ; 1969 ; "In-situ stress determination at
great depth by means of hydraulic fracturing" ; 11th symposium. tin Rock
Mechanics, Somerton editor, Berkeley Calif.
CORNET F.H. 1978 ; "Appl ication de la fracturation hydraul ique a la me-
sure in-situ des contraintes" colloque scientifique J.Goguel B.R.G.M.
Memoire n091
CROUCH S.L. ; 1976 ; "Solution of plane elasticity problems by the displa-
cement discontinuity method". Int. Jou. Num. Meth. in Eng., vol 10, p.301
CORNET F.H. ; 1979 ; "Comparative Analysis by the Displacement Discon-
tinuity Method of Two Energy Criteria of Fracture" ; Jou. App. Mech. Vol.
46 nb 2, p.349.
JAEGER J.C. and N.G.W. COOK; 1969 ; Fundamentals of Rock Mechanics,
chap.IO; Methuen and co.
HAIMSON B ; 1968 ; Hydraulic fracturing in porous and non porous rock and
its potential for determining in-situ stresses at great depth; Ph. D.
Thesis, Missouri River Division, corps of Engineers, Technical Report
4-68 Omaha Nebraska.
-1043 -
INVESTIGATIONS ON AN ARTIFICIALLY CREATED FRAC
IN A SHALLOW AND LOW PERMEABLE ENVIRONMENT
BGR-Hannover / RUB-Bochum
Summary
-1044 -
1. •. OBJECTIVES
-1045 -
Bavaria. In favour of this area are the geologic petrographic
characteristics of the "Falkenberger Granit Massiv" as well
as the low density of population and industry.
All cores from the three observation holes and the centre
hole (total core section 1200 m) were analysed regarding na-
tural fissures and cracks. The distribution of natural fissu-
-1046 -
res in the observation holes is shown in fig. 1 (see reports:
Rummel, Kassel, Teufel "Kernbeschreibung und TrennfHichenka-
talog der Bohrungen in Falkenberg",Bochum 1979).
-1047 -
The natural stress field in the underground determines
the direction and extension of artificially created fracs.
Informations about the stress field were obtained by creation
and observation of small hydraulic fracs. The volumes of the-
se fracs were some litres. The intersection of the fracs with
the hole were identified by impression packers. The results
of the experiments reveal the greatest horizontal stress com-
ponent in E-W direction below 150 m depth; the lithostatic
pressure is the minimum component down to 150 m depth, the
increase of the break down pressure with depth (frac gradient)
is about O.OJ M Pa m- 1 (see report: Rummel, Alheid "Hydrauli-
sche Fracturing Versuche in Falkenberg", Bochum 1978).
6. MAGNETIC INVESTIGATIONS
The distance between neighbouring boreholes and their di-
rection is determined by borehole measurements of the magne-
tic field which is induced by artificial magnets. This method
could have importance in deep holes for a fast determination
of distances between an injection hole and an extraction hole.
Three component borehole magnetometers (consisting of three
Forster probes) and magnetic transmitter coils with diameters
of 104 mm were built (see report: J.Pohl "Anwendung von Ver-
fahren der Magnetik zur Lokalisierung von Bohrlochern", Mtin-
chen 1979).
7. FRACTURE CREATION
For the execution of hydraulic fracturing experiments an
equipment with the following specifications was installed at
the test site: maximum pressure JO M Pa; maximum flow rate
200 l/min; straddle packer assembly, impression packers in-
cluding dipmeter and magnetic azimuth meter; electric induc-
tion flow meter; pressure transducer in the inje~tion inter-
val and at the borehole head; electrical thermometers in the
injection interval and at the bore hole head; analog recor-
ding unit with six channels; water reservoir with 30 m3 vo-
lume (see fig. 3).
-1048 -
A frac was created in 158 m depth. The break down pressu-
re was 9.5 M Pa; the pressure decreased to 9.1 M Pa within
2 minutes and dropped to 5 M Pa afterwards. The maximum injec-
tion volume was 4 m3 , recovery volume was about 50 ~. The
frac was identified by an impression packer, which shows a
horizontal intersection plane in the hole.
A. FRACTURE DETECTION
-1049 -
mic signals were created and observed in the observation ho-
les (Leydecker, BGR, Hannover 1980). An example of a seismic
event observed at four stations with three component seismo-
meters is shown in fig. 4. Eighteen such events could be used
for the construction of a model of the frac extension
(Leydecker, BGR,Hannover 1980). On the basis of this model
the locations of two intersection holes, each 15 m east and
west of the production hole were selected. The western inter-
section hole cut a fissure in 260 m depth which appears to be
the hydraulic frac.
-w~-
Fig. 3
Fig . 4
<:J)
03
<D
D2
"3
112
Folk.nberg HB 40
Ai
79OCT"/~.:t.AC -nl jgl 9·53.ca> I I I ~ .J.. el,
I w ~ ~ n~~
L.yd~cker. 8GR/PRA~A, Hannover 1980
ORGANIZATION OF' PROJECT
- 1053-
LIST OF PARTICIPANTS
AUBERT M. Observatoire
12, avenue Landais
F - 63000 CLERMONT
- 1055 -
BARTELSEN H. Institut fUr Geophysik, Universitat
Olshausenstr. 40-60
D - 2300 KIEL
BASSANI C. Joint Research Centre - Ispra Establishment
P.O. Box 1
I - 21020 ISPRA
BATARD F. B.R.G.M.
22, avenue de Lempdes
F - 63800 COURNON
BENDER W. Richard-Wagner-Str. 19
D - 7525 BAD SCHOENBORN
BERAUD-DUFOUR P. Societe Nationale ELF Aquitaine
7, rue Nelaton
F - 75739 PARIS Cedex 15
BERNARD J. B.R.G.M.
B.P. 6009
F - 45018 ORLEANS
BERRY P. Universita
Via Eudossiana, 18
I - ROMA
BERTHIER F. B.R.G.M.
B.P. 6009
F - 45018 ORLEANS
-1056 -
BLOOMER J. British Petroleum
Britannic House
Moor Lane
GB - LONDON
BORIES S. ENSEEIHT
2, rue C. Camichel
F - TOULOUSE
BOSCH B. B.R.G.M./MFA
B.P. 6009
F - 45018 ORLEANS CEDEX
BOSCHETTI Ministere de l'Industrie et Mines
97, rue de Grenelle
F - 75700 PARIS
BOULEAU J. C.N.R.S.
Rue du Loess
F - 67037 STRASBOURG CEDEX
BOURDIER J.-L. Laboratoire de Geologie et Mineralogie
5, rue Kessler
F - 63000 CLERMONT
BOWEN R. Exploration Consultants Ltd.
Harleyford Marlow
GB - BUCKS SL72DY
BOYER D. Ecole de Geologie Appliquee et de
Prospection Miniere
94, avo de Lattre de Tassigny
F - 54000 NANCY
BOZZONI T. ANSALDO
Via Pacinot ti 20
I - GENOVA
- 1057 -
BRAM K. Commission of the European Communities
Directorate General Research, Science and
Education
Environment and Raw Materials Programme
200, rue de la Loi
B - 1049 BRUSSELS
BRANGIER P. COFRETH
46, rue Letort
F - 75883 PARIS 18
BREDEWOUT J. W. Vening Meinesz Lab.
Budapestlaan, 4
NL - UTRECHT
- 1058 -
CAMPONUOVO G.F. ISMES
Viale G. Cesare, 29
I - 24100 BERGAMO
-1059 -
COSTANTINO F. ANSALDO
Via Pacinotti 20
I - GENOVA
CRAIG S. Houlder Marine Drilling Ltd.
53, Leadenhall St.
GB - LONDON
DABBOUS A. Hughes Tool
25-28, Old Burlington Street
GB - LONDON WlX lLB
D'AMORE F. C.N.R./Ist. Int. Ricerche Geotermiche
Vial del Buongusto 1
I - 56100 PISA
DANTINI E. Universita di Roma - Facolta d'Ingegneria
Via Eudossiana, 18
I - ROMA
DECKER H. Burgermeisteramt
D - 7580 BUEHL/BADEN
-1060 -
DOREL M. lnstitut FranQais du Petro1e
1 et 4, av.de BOis-Preau
F - 92500 RUElL MALMAlSON
-1061-
FAYON J. Christensen
Place la Gare
F - LA VERRIERE
FEHR A. Swiss Federal Energy Office
Kapellenstr. 14
CH - 3011 BERNE
FERNANDEZ R. Empresa Nacional ADARO
Serrano 116
E - MADRID (6)
FERRAND Direction Interdepartementale de l'Industrie
6, rue d'Ingwiller
F - 67000 STRASBOURG
FERRARA G. ENEL/Geoth. Research Center
Piazza B. Sassoferrato, 14
I - 56100 PISA
FONTAINE Cie Generale de Chauffe
310, avo d'Altkirch
F - 68200 BRONSTATT
FOUILLAC Ch. B.R.G.M.
B.P. 6009
F - 45018 ORLEANS
FRANCIS M. Imperial College of Science and Technology
Prince Consort Road
GB - LONDON SW7 2BP
FREDDI A. ISMES
Viale G. Cesare, 29
I - 24100 BERGAMO
FRIEDRICHSEN H. Universitat Tubingen
Abt. Geochemie
Wilhelmstr. 56
D - 7400 TUEBINGEN 1
FRITZ B. C.N.R.S.
1 rue Blessig
F - 67804 STRASBOURG CEDEX
FRITZ J. Ingenieurburo FRITZ
Am SchOnbl1ck 1
D - 7432 URACH 1
FUEHRER F. BEB
Riethorst 12
D - 3000 HANNOVER 51
-1062 -
FUHRBERG M. Gewerkschaften Brigitta und Elwerath (BEB)
Riethorst 12
D - 3000 HANNOVER 51
GLINTZBOECKEL B.R.G.M.
Domaine de Luminy
F - 13009 MARSEILLE
GOBLET P. Centre d'Information Geologique
Ecole des Mines de Paris
35, rue St. Honore
F - 77305 FONTAINEBLEAU
GODARD J.M. Geoservices - Hydrologie
87, rue Taitbout
F - 75009 PARIS
GOSK E. Geological Survey of Denmark
Thoravej 31
DK - 2400 COPENHAGEN NV
GOUT Ste Bordelaise de· Realisa
Domaine de Luminy
F - 13009 MARSEILLE
GRAILLAT A. Direction Interdepartementale de l'Industrie
6, rue d'Ingwiller
F - 67000 STRASBOURG
-1064-
GRINGARTEN A. Flopetrol
228, rue Einstein
F - 77530 VAUX-le-PENIL
GROSS IN R. BERTIN
B.P. 3
F - 78370 PLAISIR
GRUMPELT H. C. Deilmann - Abt. TE
Postfach 75
D - 4444 BAD BENTHEIM 1
GUDMUNDSSON J.S. Geothermal Div.- Iceland Energy Authority
Grensasvegur 9
REYKJAVIK - Iceland
GUEST P. Hughes Tool
25-28 Old Burlington Street
GB - LONDON WIX lLB
GULDENFELS W. Heimschule Lender
D - 7591 SASBACH
HAAK V. Institut fUr Geophys. Wissenschaften
Freie Universit~t
Rheinbabenallee 49
D - 1 BERLIN 33
HAENEL R. Nieders§chsisches Landesamt fUr
Bodenforschung
Stilleweg 2
D - 3000 HANNOVER
HALFON J. Secretariat d'Etat a la Recherche
Delegation Generale a la Recherche
Scientifique et Technique
35, rue St. Dominique
F - 75007 PARIS
HARMSEN G.J. Foraky S.A.
Place des Barricades 13
B - 1000 BRUXELLES
HAUERSTEIN G. Bundesministerium fUr Forschung und
Technologie
Heinemannstr. 2
D - 5300 BONN 2
HEIERS W. Commission of the European Communities
Directorate General "Scientific and Technical
Information and Information Management"
Jean Monnet Building B4/89
Plateau du Kirchberg
L - LUXEMBOURG
-1065 -
HERMINGHAUS Ch. Kernforschungsanlage - P.L.E.
Postfach 1913
D - 5170 JUELICH
HIROUX P. Foraky S.A.
Place des Barricades 13
B - 1000 BRUXELLES
-1066 -
JAUD Ph. Electricite de France
6, Quai Watier
F - 78400 CHATOU
KEPPLER H. BGR
Stilleweg 2
D - 3000 HANNOVER 51
- 1067 -
KLEINITZ Preussag
Erd8l & Erdgas
Arndstr. 1
D - 3000 HANNOVER
-1069 -
LOIDA A. Institut fUr Mineralogie
Universitat Bochum
Universitatsstr. 150
D - 4630 BOCHUM
LOOS F. Direction Interdepartementale de l'Industrie
6, rue d'Ingwiller
F - 6700 STRASBOURG
LOTT X. Geochaleur
4, place Raoul Dautry
F - 75741 PARIS Cedex
LOUWRIER K. Commission of the European Communities
Directorate General Research, Science and
Education
200, rue de la Loi
B - BRUSSELS
LUONGO G. Osservatorio Vesuviano
I - ERCOLANO - NAPOLI
MACCHI M. I.S.M.E.S.
Viale G. Cesare, 29
I - 24100 BERGAMO
MACHER J. S.C.I.
Les Thermes de Heiligenbrunn
4, rue Staber
F - MULHOUSE
MADSEN L. D.O.N.G. A/S
Dr. Tvaerg. 30
DK - 1302 COPENHAGEN
MAGET Ph. B.R.G.M.
B.P. 6008
F - 45018 ORLEANS-Cedex
MAHLSTEDE K. Gewerkschaft Walter
Am Handelshof, 1
D - 4300 ESSEN
MAKURAT Universitat Karlsruhe
Geologisches Institut
Kaiserstr. 12
D - 7500 KARLSRUHE 1
MERIGOUX J. M. Alsthom-Atlantique
141, rue Rateau
F - 93123 LA COURNEUVE
MOSNIER J. CNRS
24, rue Lhomond
F - 75231 PARIS
MUNIER-JOLAIN EURAFREP
64-70, rue de Ranelagh
F - PARIS 75016
-1072-
OTTE C. Union Oil of California
P.O. Box 7600
U.S.A. - LOS ANGELES, Calif. 90051
PARMEGGIANI A. C.I.S.E.
Via R. Emilia, 39
I - 20100 MILANO
PEANO A. ISMES
Viale G. Cesare 29
I - 24100 BERGAMO
PEZET C. EURAFREP
70, rue du Ranelagh
F - 75016 PARIS
-1073 -
PHAM VAN NGOL Institut National Poly technique de Lorraine
Ecole Nationale Superieure de Geologie
Appliquee et de Prospection Miniere de Nancy
Laboratoire de Geophysique Appliquee
94, avo de Lattre de Tassigny
F - 54001 NANCY Cedex
PICHERY L. Direction Interdepartementale de l'Industrie
6, rue d'Ingwiller
F - 67000 STRASBOURG
PLUM H. RWTH Aachen
Lehrgebiet Hydrogeologie
Lochnerstr. 4-20
D - 5100 AACHEN
-1074-
PRODEHL C. Geophysikalisches Institut der
Universit!t Karlsruhe
Hertzstr. 16
D - 7500 KARLSRUHE 21
-1075 -
ROCHE A. Institut de Physiqe du Globe
Universite Louis Pasteur
Rue Rene Descartes 5
F - 67084 STRASBOURG
ROENNEKE BUro Dr. Roenneke
Katharinenstr. 28
D - 1000 BERLIN
ROMBOUTS L. Rijksuniversiteit Gent
Geologisch Instituut
Krygs1aan 271
B - 9000 GENT
ROMAGNOLI P.L. ELC - E1ectroconsu1t S.p.A.
Via Chiabrera 8
I - 20151 MILANO
ROQUE C. Institut Francais du Petro1e
1 et 4, avo de Bois-Preau
F - 92506 RUEIL-MALMAISON
ROSSI A. Istituto Internaziona1e Ricerche Geotermiche
Via Buongusto 1
I - 56100 PISA
ROTERS B. Institut fUr angewandte Geophysik
TO Berlin
Str. d. 17. Juni 135, EB
D - 1000 BERLIN 12
ROTHENGATTER P. Universitit Karlsruhe
Kaiserstr. 12
D - 7500 KARLSRUHE 1
ROTTENBACHER K. Batte11e-Institut
Am RCSmerhof 35
D - 6000 FRANKFURT 90
RYBACH L. Swiss Federal Commission for Geothermal
Energy
ETH ZUrich
CH - 8092 ZUERICH
SATTEL G. Geophysika1isches Institut
Hertzstr. 16
D - 7500 KARLSRUHE
SAUER K. Geo1ogisches Landesamt Baden~Urttemberg
A1bertstr. 5
D - 7800 FREIBURG
-1076 -
SAUPE F. CNRS
C.O. N.l
F - 54500 VANDOEUVRE
-1077 -
SCHOELLER M. Universite Bordeaux I - Centre d'Hydrogeologie
Batiment de Geologie - Av. des Facultes
F - 33405 TALENCE
SCHROTH G. Geophysikalisches Institut
Universitat Karlsruhe
Postfach 6983
D - 7500 KARLSRUHE 1
SCHULZE-GATTERMANN R. Prakla Seismos GmbH
Haarstr. 5
D - 3000 HANNOVER
SCHWARZ G. Institut fUr Geophys. Wissenschaften
Rheinbabenallee 49
D - 1000 BERLIN 49
SCHWEIZER Geologisches Landesamt Baden-WUrttemberg
Albertstr. 5
D - 7800 FREIBURG
-1078 -
STENGER R. Mineralogisches Institut der
Univer~itat Freiburg
Albertstr.23 B
D - 1800 FREIBURG
-1079 -
THIERY D. B.R.G.M.
B.P. 6009
F - 45018 ORLEANS CEDEX
TONANI F. Via Monte Latici 1
I - 50121 FIRENZE
TONKIN G. University of Bristol
1.10 Physics
GB - BRISTOL
TORO B. CNR Centro Studio Geologia
Citta 2 Universitaria
I - ROMA
TOZZI M. AGIP S.p.A.
Piazzale Enrico Mattei
I - ROMA EUR
TRAONMILIN SNEA (p) ELF Aquitaine
Boussens
F - 31360 ST. MARTORY
TREPTE L. Dornier System GmbH
Postfach 1360
D - 7990 FRIEDRICHSHAFEN
VANDELANNOOTE U.I.A.
Universiteitsplein 1
B - 2610 WILRIJK
- 1080-
VASSEUR G. C.N.R.S.- Centre Geologique et Geophysique
Montpellier - Universite des Sciences et
et Techniques du Languedoc
Place Eugene Bataillon
F - 34060 MONTPELLIER CEDEX
VIALLE C. INAG
77, Av. Denfert Rochereau
F - 75014 PARIS
-1081 -
WERNER D. Institut fur Geophysik
ETH-H6nggerberg
CH - 8093 ZUERICH
WERNER J. Geologisches Landesamt Baden-Wurttemberg
Albertstr. 5
D - 7800 FREIBURG
WHEILDON J. Imperial College of Science and Technology
Geophysics. Dept.
Prince Consort Road
GB - LONDON SW7 2BP
WILKINSON TRW-REDA
1-7, Artillery
GB - LONDON SW 1
ZET'lWOOG J. ALFA-LAVAL
F - LESCLAYES-SOUS-BOIS
ZIER M. ERNO
Raumfahrttechnik GmbH
Hiinefeldstr. 1
D - 2800 BREMEN
-1082 -
ZILIOTTO M. ISMES
V1a1e G. Cesare 29
I - 24100 BERGAMO
- 1083-
AUTHORS' INDEX
-1085 -
MURAOUR P., 632 RAPOLLA A., 121 TAFFI L., 506
MUSMANN' G., 556,904 RICHARDS M.L., 893 TARDY Y., 340
RICHARDSON S.w., 447 TAZIEFF H., 37
RIPEPE 14., 875 THIERY D., 756
RISSE R., 109 THILLIER J.M., 516
NANNINI R., 141 ROCHE A., 89 THOMAS-BETTS A., 456
NAPOLEONE G., 875 ROCHON J., 735 TIETZE R., 45
NERCESSIAN A., 622 ROQUE C., 977 TONANI F.B., 428
NETH G., 253 ROSSI A., 506 TRAMACERE A., 833
NEURIEDER P., 911 ROTERS B., 525 TREBBI G., 694
NICOLICH R., 603 ROTH B., 803
NOIER M.L., 756 RUGO R., 823
NUNZIATA C., 121 RUMMEL F., 1044
UNGEMACH P., 9
-1086 -