SPE-171469-MS

Ichthys Field Development - Implementation of a Multi-Disciplinary Well

Optimisation Workflow for a Large Gas-Condensate Reservoir
A. Kawasaki, S. Batiot and A. Ion, Inpex

Copyright 2014, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Asia Pacific Oil & Gas Conference and Exhibition held in Adelaide, Australia, 14 16 October 2014.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Optimisation of development well locations is a critical field development objective to ensure maximi-
sation of hydrocarbon production and gas plateau duration. In a large reservoir with relatively sparse well
control, significant inter-well uncertainty can exist, resulting in a high degree of subjectivity in the
selection of production well locations. Following the implementation of a comprehensive workflow to
construct multiple geologic realisations of the Brewster reservoir, a two-step process was developed to:
1) objectively identify optimum production well locations that suited a selected range of geologic models,
and 2) subjectively refine these locations to satisfy practical requirements.
Optimum well locations were initially identified using a systematic approach covering a wide range of
geologic scenarios based on an objective to maximise gas plateau duration. Resulting well locations for
each realisation were then consolidated into a single well pattern using a frequency mapping technique.
Subsequently, a subjective approach was adopted to refine these well locations based on considerations,
including targeting of structurally high points, minimum inter-well spacing, step-out from subsea drill-
centre locations, maximising the value of data acquisition (distance from existing well control) and
reducing well construction complexity (minimising lateral step-out and drilling in non-preferred azimuth).
Concurrently, optimisation of well geometry and total depth was investigated which indicated the
potential for drilling/completion cost savings with negligible impact on predicted production performance.
The integrated, multi-disciplinary nature of this methodology was paramount in effectively commu-
nicating and satisfying well planning/execution objectives between key project disciplines.

Introduction
Following the acquisition of exploration permit WA-285-P by INPEX CORPORATION in 1998, the
Ichthys Field was discovered in 2000 during an initial three-well drilling campaign, and was appraised in
two subsequent campaigns between 2003-2007. The Ichthys LNG Project is expected to produce 8.4
million tons of LNG and 1.6 million tons of LPG per annum, along with approximately 100,000 barrels
of condensate per day at peak. The final investment decision for the Ichthys LNG Project was made in
January 2012, and production licenses WA-50-L and 51-L were subsequently awarded over an area of 1,
077 km2 (Fig. 1).
2 SPE-171469-MS

The Field will be developed using two offshore

production facilities. Well fluids will flow through
subsea infrastructure for preliminary processing by
a central processing facility (CPF). Produced water
and the majority of hydrocarbon liquids will be
extracted and routed to a floating production, stor-
age and offloading (FPSO) vessel, whilste the re-
maining lean gas stream will flow via an 889 km a
gas export pipeline (GEP) to a processing plant at
Bladin Point near Darwin, Northern Territory, Aus-
tralia (Fig. 2).
Approximately 50 subsea wells are expected to
be drilled from multiple drill centres to provide
production over a period of approximately 40 years.
Well construction will occur over a number of
phases, with the first phase expected to commence
Figure 1Location of Ichthys Field
in late 2014.
Broadband 3D seismic data was acquired in
2010, and geological modelling of the Brewster Member (one of two productive horizons in the Field) was
subsequently conducted to facilitate development well planning. Geologic models were created by
integrating several petrophysical, geophysical and geological studies that have captured the reservoirs
complex setting1.

Well Optimisation Workflow

The subject of well optimisation has been covered extensively within industry literature2, 3. Fig. 3 shows
an overview of the well optimisation workflow applied for Ichthys Field development. Numerical
reservoir simulation was used to identify the optimum number and location of wells for development.
Incremental economic analysis was conducted to confirm optimum well-count. Well locations were
objectively selected using an iterative reservoir simulation method with the objective of maximising
production plateau duration with the optimum well-count. In parallel, an alternative methodology of well
location selection using estimated permeability-thickness (KH) and initial gas-in-place (GIIP) was
evaluated.
These reservoir simulation studies were carried out using multiple geologic scenarios, with resulting
locations from different models consolidated for a further subjective refinement step, which included
consideration of well trajectory design, geologic/drilling risks, step-out from drill-centres and data
acquisition requirements. This process was repeated until reservoir simulation results indicated satisfac-
tion of production objectives.

Reservoir Simulation Base Well Optimisation

An objective reservoir simulation methodology (well thinning) was used to determine optimum
well-count/location as an initial screening process. This involved a step-wise reduction in well-count from
an initial high-density well pattern by eliminating low-ranked candidates using pre-defined well perfor-
mance criteria. Simplified production assumptions were applied to reduce computational time.
This optimisation workflow was as follows:
1. Position vertical wells with regular, high-density spacing to cover the whole gas accumulation area
(Fig. 4).
2. Conduct reservoir simulation, and rank wells by gas-production potential at plateau-end (primary
SPE-171469-MS 3

Figure 2Ichthys Field Offshore Development Concept

Figure 3Well Optimisation Workflow

criteria) or by cumulative gas production (secondary).

3. Eliminate low-potential wells and re-run simulation.
4. Repeat Steps 2-3 until realistic minimum well-count achieved.
5. Repeat Steps 1-4 for selected geologic realisations.
4 SPE-171469-MS

Figure 4 Initial Well Pattern

Figure 5Well-Count vs Field Plateau Duration (left) & Incremental Plateau (right)

Two different initial well patterns were evaluated in Step 1 in order to reduce dependency on initial
well location assignment and impact of local heterogeneity, as shown in Fig. 4. In Step 3, some process
rules were applied: the maximum number of wells eliminated in a single step was set to be less than 10
per cent of the total well-count, and adjacent wells were not eliminated in the same step.
While several criteria could be applied to rank well performance in this process, well gas production
potential, which represents maximum production rate under prevailing production constraints, was
selected as the primary parameter of interest. This process was conducted using an automated, in-house
system which enabled multiple geologic cases to be run simultaneously.
Fig. 5 shows an example of relationships between plateau duration and well-count obtained from this
process.
SPE-171469-MS 5

Figure 6 Well Location Refinement for Subsea Design

Well Location Refinement Based on Subsea Design

Analysis of well locations obtained from Step 1 occassionally indicated excessive clustering of wells
within one drill-centre location, which was partly driven by the simplified production assumptions applied
(Case A well locations in Fig. 6). Based on consideration of drill-centre layout and number of available
well slots, some wells were relocated to other drill centres (next-best locations at that drill centre, as
identified in Step 1), represented simplistically as Case B well locations in Fig. 6.
Production forecasts (with full subsea network included) were generated for both Cases A and B, and
a confirmed production benefit of Case B in terms of gas plateau duration and cumulative gas production
for all geological scenarios.

Subjective Well Location Selection (Check)

As an alternative methodology, and in parallel with the reservoir simulation base optimisation in Step 1,
well locations were subjectively selected based purely on a fundamental assessment of well spacing and
key reservoir parameters:
structural elevation;
reservoir thickness; and
gas saturation and permeability.
Well spacing can be important in gas-condensate reservoirs. While wells concentrated in highly
productive areas could achieve higher gas production rates, any localised pressure sinks can induce earlier
in-situ condensate drop-out, which can reduce well productivity and hydrocarbon recovery.
Reservoir simulation was performed to test these alternative well layouts for the selected geologic
scenarios. Results did not show any improvement in gas production; however, a minor improvement in
condensate production was observed in some scenarios.
6 SPE-171469-MS

Figure 7All Well Location Candidates

Well Location Candidate

Consolidation
Given the inherent geologic uncertainty across a
large field area, optimised well locations from Step
1 differed between geologic realisations. More than
100 well locations were selected from various geo-
logic cases. Fig. 7 shows a representation of all
possible well location candidates, including nota-
tion of frequency of selection for each well (for
example, a frequency of five indicates that the same
well location was objectively selected in five differ-
ent geological realisations).
Consolidation of highest-frequency locations
was conducted within a regular spacing, and a rep- Figure 8 Consolidated Well Locations
resentative well was set in the centre of this spacing
(Fig. 8). The resulting well pattern was then carried
forward for multi-disciplinary review.
Multi-Disciplinary Review
In previous steps, well locations were initially selected to maximise plateau duration and hydrocarbon
recovery. Although a wide range of geologic scenarios were evaluated, capturing a realistic level of
variation in reservoir structure and rock properties, multi-disciplinary judgement was ultimately para-
mount to ensure that final target locations met all subsurface and drilling requirements. The parameters
considered in this review are outlined below.
The philosophy during this stage of well selection was to conduct on a well-by-well basis an extensive
review of each target, incorporating subjective perception of geological risks not captured at the resolution
of the dynamic model in addition to project considerations for field development. In addition, proximity
to existing well control was also considered as a factor for well location refinement, particularly for early
wells in the drilling campaign, to maximise the value of data acquisition. An assessment of wellbore
stability for candidate locations was also conducted to ensure drillability based on geomechanical
modelling and well azimuth relative to its drill-centre location.
SPE-171469-MS 7

Figure 9 Volcanics-Related Seismic Anomalies

Reservoir entry points were geographically refined to ensure coincidence with the highest local
structural elevation within the target area. In order to increase confidence in reservoir intersection depth,
all targets were aligned with appropriate seismic lines. A geophysical review was conducted to check
targets located in the vicinity of faults where a secondary faults network can be visible or where volcanic
material may be detected as strong impedance signatures in the main fault plane similar to a volcanic
conduit (see Schematic A in Fig. 9).
Volcanic material can be present as lava flows above the Brewster Reservoir in the Upper Vulcan
Formation (see Schematic B in Fig. 9). Based on previous experience and the expected high inclination
8 SPE-171469-MS

in some development wells, drilling is expected to be challenging in this type of environment. In addition,
due to the vertical resolution of seismic data (and as proven during the appraisal campaign), predicting the
thickness of these seismic features can have a high degree of uncertainty. Some reservoir features were
also observed due to intrusions originating from the underlying Ichthys Formation (see Schematic C in
Fig. 9), which may compromise the Brewster Reservoir interpretation. Following a review of these issues,
some wells were relocated to adjacent, lower-risk areas.
Detailed analyses of each target by the Geology, Petrophysics and Reservoir Engineering Teams
resulted in pragmatic definition of optimum Total Depth (TD) for each of the Brewster development wells.
The Petrophysics and Geology Teams proposed a practical TD based on operational considerations (LWD
string length and sensor locations, survey uncertainties), geological features (depth and thickness of
intra-formational mudstone marker, reservoir quality variation within the Brewster Member) and other
subsurface uncertainties.

Conclusions
Preliminary development well locations were selected using an objective methodology developed in-
house which enabled rapid, automated screening of multiple geologic realisations. Subjective refinement
of these locations was then conducted across multiple disciplines, including assessment of the following
issues:
Optimise reservoir intersection relative to local structural interpretation;
Avoid complex geological features that could cause drilling problems (e.g. large faults, volcanic
rocks);
Refine TD to minimise drilling cost and minimise risk of water production;
Optimise well spacing and offset from existing well control;
Consider drillability (borehole stability, azimuth, etc.), lateral step-out and optimum well incli-
nation; and
Optimise sequence of higher-/lower-risk wells for production assurance and data acquisition.
The well optimisation workflow enabled a comprehensive and logical assessment of multiple well
design variables relating to geoscience, reservoir engineering and well construction, and promoted
discussion/alignment across all disciplines.

Acknowledgements
The authors wish to acknowledge TOTAL, Tokyo Gas, Osaka Gas, Chubu Electric Power and Toho Gas
as Ichthys Joint Venture participants for their permission to publish this paper. INPEX management is also
acknowledged for support provided during the preparation and review of this document.

References
1. The Ichthys Field: Challenges of Geological Modelling for the Field Development, Nakanishi
et alet al., Australian Petroleum Production & Exploration Association (APPEA) 2014 Confer-
ence.
2. Optimal Well Placement for Production Optimisation, Wang C., Li G. & Reynolds A.C.,
SPE-111154, presented at SPE Eastern Regional Meeting, Lexington, Kentucky, 11-14 October
2007.
3. Improved Methods for Multivariate Optimisation of Field Development Scheduling and Well
Placement, SPE-49055, Pan Y. & Horne R.N., presented at SPE Annual Technical Conference
& Exhibition, New Orleans, Louisiana, 27-30 September 1998.
SPE-171469-MS 9

AUTHOR BIOGRAPHIES
Akikazu Kawasaki is Senior Reservoir Engineer at INPEX in Perth. He spent five years working on
reservoir engineering projects in Indonesia, Iraq and Australia. He gained a BS degree in 2003 and MS
degree in 2005 from Kyoto University.
Sebastien Batiot is Senior Development Geology Advisor at INPEX. He gained an M.Sc in Petroleum
Geology in IGAL France in 2000, and worked as a lead geologist with Schlumberger on Venezuela,
Algeria, Angola and Malaysia E&P projects. He joined INPEX Perth in 2012 to manage subsurface well
planning for field developments.
Andy Ion is General Manager, Reservoir Engineering at INPEX in Perth. He gained an M.Eng in
Petroleum Engineering from Heriot-Watt University in 1989 and has worked at INPEX since 2006.