IPTC-17791-MS

A Systematic Approach in Delivering Infill and EOR Opportunities to

Redevelop a Highly Complex Mature Field: A Case Study
R.A. Karim, H. Lee, C. Hong, M. Mohamed Latif, H. Anggono, N.H. Husain, and M.C. Tayok, EORC;
R.D. Tewari, PETRONAS Carigali Sdn. Bhd

Copyright 2014, International Petroleum Technology Conference

This paper was prepared for presentation at the International Petroleum Technology Conference held in Kuala Lumpur, Malaysia, 10 –12 December 2014.

This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s).
Contents of the paper, as presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s).
The material, as presented, does not necessarily reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers
presented at IPTC are subject to publication review by Sponsor Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the International Petroleum Technology Conference 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 where and by whom the paper
was presented. Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax ⫹1-972-952-9435

Abstract
This paper discusses about the systematic approach to identify and firm up the infill wells and EOR
opportunities as part of the redevelopment strategy in a multilayered mature offshore oilfield which has
been on production since the last 46 years. The field is located offshore East Malaysia at a water depth
of 186 ft. Production performance of the field indicates the current development strategy and practices will
yield a moderate recovery which is currently around 26%. It is in “mid-life crisis”, facing problems of
increasing water cut, high GOR, with bypassed/undrained oil areas and pressure depletion along with
aging facilities.
In 2010, the study team embarked on full field redevelopment study to evaluate the infill and EOR
opportunities. However, building dynamic model and identifying the remaining hydrocarbon from this
area are very challenging due to its highly complex nature; thin oil-rims in over 200 multi-stacked
reservoirs with individual thickness of 10 to 30ft, occurring at 2300 to 9800ft depth. The intensed faulting
subdivides the field into hundreds of small reservoirs in numerous fault-blocks, with varying degree of
fault-block communication and aquifer support. Limited reservoir data, including fluid contacts and fluid
types, coupled with ambiguity in production allocation due to commingled production, further complicate
dynamic modeling and history matching processes. These uncertainties will affect the remaining hydro-
carbon volume and distribution, hence the success of redevelopment project through infill and EOR.
A systematic approach to evaluate infill opportunities and EOR performance using dynamic modeling,
coupled with LTRO (Locate The Remaining Oil) studies will be discussed. The dynamic model captures
the complex fluid movement due to heterogeneity and inter-fault block communications, whereas the
LTRO provides an independent analysis on the current fluid contacts and remaining oil locations. Both
methods complement each other, hence providing better understanding on the targeted reservoir and
maturing the infill and EOR targets. Active surveillance and data gathering activities have also been lined
up to reduce subsurface uncertainties. All these prove to be a comprehensive workflow to unlock the
remaining oil in this complex field, resulting in several sizeable portfolios with potential to increase the
recovery factor by 4 to 9%.
2 IPTC-17791-MS

Introduction
The field is a geologically complex field, with hun-
dreds of multilayered reservoirs consisting of sand-
shale sequences, and heavily compartmentalized by
antithetic faults (Figure 1). Throughout the produc-
tion life, the field has undergone primary depletion
with a peak production of around 64,000 bopd.
Several phases of infill drilling campaigns have
Figure 1—General field cross section with initial fluid, illustrating large
been executed to arrest production decline and de- degree of compartmentalization and complexity.
veloped the remaining hydrocarbon, however the
average field recovery factor (RF) is only 26%.
It is important to understand why such a low recovery factor after 46 years of production. As depicted
in Figure 1, the hydrocarbon volume is actually scattered in more than 600 small reservoir compartments,
with 88% of them contain less than 2MMstb. Majority of the produced volumes have actually been
developed from the larger oil pockets which form only 12% of the total compartments. This imposes
significant challenges on how to develop these remaining reservoirs effectively and economically,
especially in the deeper sections due to low oil volume per compartment and often poorer property.
The RF also varies vertically due to variation in reservoir properties and aquifer support over a total
reservoir section of 7000 ft. The permeability is the highest at the top reservoirs, ranging from 1-2 Darcy,
and decreases in the deeper section, with permeability of 50-200md. Aquifer support is also the strongest
at the top section, and decreases in strength towards deeper reservoirs. As a result, the maximum RF of
45% is only observed at the top section whereas the deeper reservoirs could only achieve RF of 2-15%
due to large pressure depletion.
On top of the geological complexity, the team also faces challenges in understanding the reservoir
connectivity and fluid movement within a fault block and inter-fault block. There are very limited pressure
data and initial fluid contacts, where only 5% out of the thousands of fluid contacts were seen from well
logs. Production data which is a critical source of information to understand the reservoir behaviours, also
contains some ambiguity in production allocation due to commingled production, with very few PLT logs
to validate them. All these uncertainties have posed challenges in analysing the reservoir performance,
either using analytical method or using dynamic model. The quality of history matching would be affected
by these uncertainties, hence affecting the remaining oil volume and location, and thus the success of any
future infill and EOR project for this complex brownfield.

Building confidence through systematic approach

With multiple uncertainties and risks highlighted above, a systematic approach using a combination of
dynamic modelling and back to basics ‘Locate the Remaining Oil’ (LTRO) technique has been adopted.
Two workstreams have been set up to work in parallel. The dynamic modelling aims at capturing the
complex fluid movement due to heterogeneity and inter-fault block communications, whereas the LTRO
provides an independent analysis on the current fluid contacts and remaining oil locations. Both methods
complement each other, hence providing better understanding on the targeted reservoir and maturing the
infill and EOR targets. Figure 2 depicts the systematic approach for this complex field, which will be
discussed in the following sections.

Infill Screening through LTRO

The LTRO or ‘Locate the Remaining Oil’ technique is a back-to-basics analytical approach which
integrates all geological, petrophysical, reservoir engineering and surveillance data to come up with an
IPTC-17791-MS 3

inventory of remaining oil opportunities. This is a

separate workstream led by the Production Geolo-
gists (PG) with support from petrophysicists, reser-
voir engineers, and production technologists.
Since there are hundreds of reservoirs that spread
vertically and areally over many fault blocks, the
team has to come up with a strategy to prioritize the
reservoirs to focus on. First, a thorough screening is
carried out by classifying the reservoirs according to Figure 2—The systematic approach to deliver robust infill and EOR
STOIIP range (e.g. ⬎5MMstb, ⬍2MMstb, etc.). targets through combination of dynamic modelling and LTRO analysis.
Then they are prioritized according to the size of the
remaining opportunity. The selected reservoirs
comprise a group of sands in a particular fault
blocks, and not just an isolated sand. This is to
ensure they can be translated into economic infill
well targets with sufficient primary and secondary
opportunities.
For each of the reservoir, the LTRO analysis uses
various data inputs as illustrated in Figure 3. This
includes generating multiple maps such as depth Figure 3—Various data inputs for LTRO using back-to-basics analysis
maps with fluid contacts and logs, bubble maps
based on production data, cross-sections and stick
plots for fluid column analysis. Production trends
(e.g. watercut and GOR trends), as well as pressure
data also help to understand inter-fault block and
vertical fluid connectivity. On top of that, wellbore
diagrams, cement bond logs (CBL), and well inter-
vention database also serve as good references to
validate the allocated production data and investi-
gate any production anomalies that may occur, for
instance due to behind casing communication. Figure 4 —The LTRO methodology which produces multiple scenarios
Through LTRO analysis, the team is able to; on remaining oil location and volume, plus the de-risking strategy to
increase confidence on infill and EOR opportunity.
i. generate the low and high case realizations
of original and current fluid contacts. This will determine the range of remaining target oil
volumes.
ii. confirm the location of remaining oil and identify an inventory of behind casing and infill well
opportunities. This findings would be fed back into the dynamic model to ensure the robustness
of infill locations despite uncertainties in history matched model.
iii. validate no-further-activity (NFA) reserves as estimated from decline curve analysis (DCA).
iv. provide production forecasts and reserves estimates for infill well, as well as the range of recovery
factor to ensure consistency between LTRO and dynamic model.
v. determine data acquisitions requirement to de-risk the identified opportunities.
Figure 4 illustrates the methodology, where in this example there is uncertainty on existence of gas cap,
after observing only oil-up-to (OUT) from the fluid column chart. The high case map indicates the
possibility of having oil filled up to the crest, where infill wells are best placed at the crest to recover the
unswept volume via downdip injection. However, should there be any gas cap as illustrated in the low case
map, the infill wells are exposed to very high risks of not recovering the reserves as expected. This will
4 IPTC-17791-MS

trigger the need for running sensitivities in the dy-

namic model to see the impact on recovery. At the
same time, the team also put a priority for data
acquisition in getting cased-hole pulse neutron log-
ging (PNL) to get the saturation and ascertain the
presence of gas, thus improving the confidence in
the proposed infill well and EOR opportunities.
Dynamic modelling for IOR/EOR Figure 5—Improvement of history matching quality after LTRO work
especially in watercut and gas matching due to better definition of fluid
process contacts ranges.

Two types of models are used for modelling the

recovery process at different scales, which are the
field sector model (FSM) and the full field model
(FFM).
Field sector model (FSM)
The FSM is constructed to model the fine scale
displacement by capturing the heterogeneity of the Figure 6 —The low case and high case current contacts maps from
selected reservoir. This is very crucial to understand LTRO that would impact infill well location.
the recovery process during EOR screening, as dis-
placement efficiency and incremental EOR recovery are strongly related to the distribution of rock quality,
or heterogeneity. Presence of high permeability feature for example may cause early breakthrough of
injected fluid, and therefore limits the amount of oil volume being contacted and thus reduces the sweep
efficiency. If the coarse full field model is used, which is designed for primary depletion, the results are
usually misleading as the more homogeneous model will make the injection more favourable and
therefore provides better sweep efficiency.
For this field, 5 different FSM are constructed to represent 5 major sands targets with different
reservoir characteristics, mainly the coarsening upward and fining upward sequences. The vertical
resolution of the FSM is fine enough to represent similar heterogeneity as seen at core level, which is
calibrated by comparing the Lorenz and Dykstra Parsons coefficients of the models.
Through the FSM, the team is able to;
i. develop understanding of recovery processes (water injection, downdip gas injection, water-
alternating gas (WAG) and low salinity flooding) and the recovery efficiency for different
reservoir characteristics.
ii. assess impact of static and dynamic uncertainties using multiple static realisations to generate
multiple history matched models. With multiple models, the team can generate probabilistic
forecast for infill and EOR, and therefore providing the ranges of recovery and production for
concept select1.
IPTC-17791-MS 5

iii. generate the EOR type curve to scale up to

full field forecasts, and use it to discount the
forecast from the coarse full field model1.
Full field model (FFM)
The FFM is an upscaled geological model, and
therefore it is coarse and lack of details for EOR
prediction. It is mainly utilized to capture the inter-
fault block communication, which is one of the Figure 7—(Top) Remaining oil before infill well. (Bottom left) Remain-
major uncertainties in this field. Through sensitivi- ing oil after infill with downdip water injection, which indicate better
ties of fault transmissibilities, it allows understand- sweep if wells are placed at the crest, compared to the downdip location
(bottom right).
ing on fluxes scenarios between multiple fault
blocks, which is critical in optimizing the injector
locations. The injector-producer spacing can be dif-
ferent if the reservoirs in multiple fault blocks are in
communication, versus if they are totally isolated.
By building multiple scenarios, the team could
come up with de-risking strategy, e.g. by placing the
first few injectors in the critical locations and use
tracers to understand the connectivity1. Figure 8 —The range of incremental recovery factor over infill that can
be achieved through DDGI, WAG and WI.

Case study
Improvement in history matching
The G reservoir has been selected for the case study. It is mainly dominated by the coarsening upward
sequence, with strong aquifer support. A sector model is built for the reservoir using a base static
realization. During the history matching process, sensitivities are run on multiple static and dynamic
parameters, including the initial fluid contacts. It is found that the model could be better matched using
the fluid contacts ranges from the LTRO fluid column analysis, especially for the gas oil ratio (GOR) and
watercut. Previously, it was difficult to obtain such matching due to commingled production and high
uncertainty in interpreting the fluid type based on well logs only. The LTRO team has reinterpreted the
fluid type together with stick plot, production data, well correlation, depth maps, and bubble maps to
provide better definition on the possible fluid contact ranges. With improved matching, it increases the
confidence in model prediction especially for infill opportunity. Figure 5 shows the improvement in
history matching using LTRO fluid contacts.

Maturing infill opportunity

The LTRO analysis indicates there is good remaining oil opportunity in G reservoir that can be developed
through 3 infill wells for the low case, or 5 wells for the high case (Figure 6). The difference between the
low case and high case is due to the uncertainty in presence of gas cap, which would affect location of
the infill wells.
The impact is assessed in the sector model, where history matching indicates the most likely scenario
is the low case. The infill wells location are sensitized to either be at the crest and downdip. Results show
that the incremental recovery factor is the highest if the infill wells are drilled at the crest, with the strategy
of having downdip water injection and gas cap blowdown (Figure 7). It would give better sweep efficiency
and at the same time additional reserves for the well due to the produced gas.

1
Details on modeling work involving FSM and FFM are planned for future paper.
6 IPTC-17791-MS

Screening for EOR opportunity

The G reservoir is also screened for EOR opportunity using a field sector model (FSM). The vertical grid
resolution is 2ft to capture the reservoir heterogeneity with Dykstra Parsons coefficient of 0.74. Three
types of recovery processes are assessed, water injection (WI), downdip gas injection (DDGI) and
immiscible water alternating gas (iWAG).
Figure 8 shows the incremental recovery factor over infill that can be achieved by each process versus
cumulative injected pore volume. The preliminary results indicate there is EOR opportunity for the G
reservoir, with the highest incremental recovery obtained by WAG, closely followed by WI and lastly
DDGI. The low incremental from DDGI could be attributed to the coarsening upward feature of the
reservoir, which would not be very favourable for gas injection. The injected gas preferentially moves
through the higher permeability layers at the top, causing early breakthrough and inefficient sweep. WAG
could address this by better mobility control and thus able to contact more unswept volume.
The overall findings from the LTRO and FSM show there are both infill and EOR opportunities for
G reservoir, with incremental recovery factor between 4 to 9% (over infill, unconstrained). The EOR
modeling is still ongoing and detail modeling work is to be discussed in the future paper.

Conclusions
This paper discusses a systematic approach for assessing the infill and EOR opportunities for a very
complex mature field. Several conclusions can be made from the study;
i. The approach is suitable for any mature brownfield, especially for the complex ones with high
risks and uncertainties. It gives a clear and systematic strategy to prioritize, assess and mature each
opportunity without being overwhelmed by the amount of uncertainties associated to each
reservoir (which could be hundreds of them for highly compartmentalized field).
ii. LTRO analysis is an excellent technique to identify the remaining oil location and infill oppor-
tunities. For a complex mature field where building dynamic model may take years, coupled with
the challenges for history matching due to limited data, LTRO may give a faster and better solution
to identify infill locations.
iii. In absence of fluid contacts information, LTRO analysis can generate multiple scenarios of
original and current fluid contacts, as well as possible reservoir connectivity, which could be
sensitized in dynamic model to improve the history matching.
iv. Field sector model (FSM) is required to understand the recovery processes and screening of EOR.
It must be fine enough to capture the heterogeneity, otherwise results can be misleading. It is not
advisable to use the full field model (FFM) for EOR prediction, which is usually coarse and
designed for primary depletion. For complex heavily faulted field, FFM should be used to
understand the inter-fault block connectivity and for optimizing the injector locations.
v. By combining the results from LTRO, FSM and FFM, each infill and EOR opportunity can be
ranked according to its confidence level, where an opportunity is considered high confidence if
results are consistent in the 3 approaches. Any inconsistency should trigger for more data
acquisitions to de-risk the opportunity.

Acknowledgement
The authors would like to thank the project team for their valuable contributions and great support to
complete this IPTC paper. Special appreciation to EORC and PETRONAS Carigali for reviewing and for
granting the permission to publish the paper.

References
1. A. Wefzelaer, A. Hildebrandt, S.D. Coutts, SPE and S.W. Veeman, Shell U.K. Exploration and
IPTC-17791-MS 7

Production: “Locating the Remaining Oil in the Brent Field Prior to Depressurisation”, paper SPE
36891 presented at SPE European Petroleum Conferences, Milan, Italy, 22-24 October 1996.
2. Laurent Costier, Vitaly G. Mitroshkin, Hermelia Mbadinga Hayes*, Shell Gabon (*now with
Brunei Shell Petroleum): “Locating the Remaining Oil (LTRO) and Delivering Opportunities in
a Mature, Declining Oil Field”, paper SPE 140636 presented at 34th SPE International Confer-
ence and Exhibition, Tinapa-Calabar, Nigeria, 31 July – 7 August 2010.
3. Moreno, Jaime; Flew, Steve; Gurpinar, Omer Schlumberger: “EOR: Challenges of Translating
Fine Scale Displacement into Full Field Models,” paper SPE 143568 presented at SPE Enhanced
Oil Recovery, Kuala Lumpur, 19-21 July 2011.