SPE 147706

Improving Oil Recovery and Injection Strategy in Shallow Reservoir (Rindu

Reservoir) of Area 3&4 Duri Steam Flood
Sudianto Lumbantobing, and Sandra Natalia, and Henri Sahat MP Silalahi, Chevron Pacific Indonesia

Copyright 2011, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Asia Pacific Oil and Gas Conference and Exhibition held in Jakarta, Indonesia, 20–22 September 2011.

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
The Rindu reservoir of Area 3&4 Duri Field has been
developed economically since 1994. Since it is shallow
reservoir and penetrated by existing PK producer wells,
the Rindu reservoir is started producing commingles with
PK sand in 1997 (Rindu Vertical Expansion/ RVE well).
In alignment to this effort, the Rindu injection also started
at the same year. After 8 years production-injection, these
commingle producers have poor production performance
due to back pressure, cross-flow, and scaling.
Through heat management review, it was found that
Rindu in the southern Area 3&4 has longer maturity
response, poorer injection (200 BSPD), and longer
heating time (~80 years).
An integrated geological and reservoir engineering study
of Rindu performance is carry out to define improvement
production and steam injection in southern part Area 3,4.
The study resulting some key finding:
1. Rindu and PK have significant reservoir pressure and
temperature different.
2. All poor performance RVE producers are facing
cross flow/backpressure and scaling problem
3. Low drainages point in Rindu pattern resulting high
reservoir pressure
4. Low injection performance controlled by high
backpressure
5. Southern Area 4 Rindu sand has a poorer reservoir
quality compare to Northern part.
The project team proposed to drill separate Rindu only to
avoid cross-flow, back pressure, or scaling issues. At the
same time it improved Rindu pattern drainage points. The
project also proposed to increase steam injection rate
gradually after reservoir pressure decreased. After
completing the project, the Rindu reservoir deliver 50
BOPD/well, increase steam injection by 500% (1000
BSPD), accelerate pattern heating process (~2.5 years),
and better reservoir management.
This study recommends developing Rindu and PK
reservoir separately. Injectivity improvement will be
resolved through completing pattern drainage points and
deliver better reservoir heat management.
Introduction
The Area 3&4 are located in the southern part of Duri
Field Anticline. These areas composed by Pertama-Kedua
(PK) sandstone and Rindu sandstone reservoirs. The
entire reservoirs are deposited from estuarine and shallow
marine deposits. The PK reservoir as the main Duri
reservoirs is characterized by thick, clean, high porosity,
high permeability (> 1000mD) and relative deeper
reservoir (> 600’TVDSS). This reservoir is form by
amalgamated channel deposits. This interval has been
injected and produced in early Duri field steamflood
development.
The Rindu sandstone is part of the shallow reservoir of
Duri Field (300-500’TVDSS). The Rindu sandstone is
deposited as product of estuarine to shallow marine
depositional environment. The Rindu reservoirs are
formed by delta mouth bar, offshore marine bar sand and
channel bar deposits (Johansen et al, 2009). The Rindu
reservoir quality distribution is highly varied across field
that follow stratigraphic trend. This study is focused on
area that composed by channel edge deposits. This
deposit is characterized with muddier sandstone, thinner
sand (30-60 feet) and relatively moderate permeability
(500-1500mD).
2 SPE 147706
The Rindu reservoir was started producing commingles
with PK sand in 1997 under RVE (Rindu Vertical
Expansion) project. Although it has a very different
reservoir properties, the RVE wells were completed
comingle between PK sandstone (Open Hole Gravel
Pack) and Rindu interval (Cased Hole Gravel Pack) as
shown in Figure-1. The Rindu injection started at the
same time that targeting pattern with RVE wells. Rindu
reservoir is continuing developed several years after RVE
project until now a day (Figure-3).
Problems
The RVE (Rindu Vertical Expansion) project was
completed with lower cost. Despite it contribute to capital
cost reduction, but these wells are not performing as
expected. Through detail review of all RVE wells in study
area, several problems are founded as listed below
1. Poor Production Performance
There is no significant contribution from Rindu Sand after
produced commingle between Rindu and PK sand. This
finding was proved with the result of finger print study in
Area 5, 8, 9 Duri Field. Based on an oil finger printing
study, Rindu sand in RVE wells only contributes 17% in
average of the total production or only could deliver 30%
recovery factor until the end of the field life (Figure-
2a/2b).
2. Poor Injection Performance
Beside the poor production, injecting in Rindu sand also
creates its own problems. Most of the injectors only able
to take under 200 BSPD steam injection rate (1000 BSPD
injection target) since the wellhead pressure has reached
their maximum wellhead pressure/fracture gradient
(Figure-4.a).
3. Cold Reservoir/ No Temperature Response
Surveillance in the reference observation wells proved
that no temperature response in Rindu sand of southern
part and some saturation logs show that Rindu reservoir
properties (net pay and oil saturation) didn’t change much
compare to its original, even though it had been injecting
and producing more than 8 years (Figure-4.b).
4. High Pressure Reservoir
The recent drilling activities were facing with high
pressure during penetrating Rindu reservoir with
equivalent mud weight (EMW) ~15 ppg. The high
pressure reservoir is also indicated by high shut in casing
pressure of some injector in the drilling location (Figure-
5.b). This finding is the evidence of accumulation
reservoir pressure in Rindu reservoir. The reservoir
pressure data was taken from new drilled well shows that
Rindu pressure tends to increase up to 1.5 – 2 times
higher than its original reservoir pressure (Figure-5.a).
Data & Analysis
A comprehensive approach of reservoir engineering and
geological review is conducted to address the problems
above. Numerous data is collected and utilized to support
this review as follows:
1. Static reservoir properties
This data included porosity; permeability; net pay;
and structure map (Figure-6).
2. Production-Injection Profile
The production-injection data consist of Area 3&4
production-injection history (Figure-3); RVE
production-injection performance (Figure-2); Rindu
production-injection profile before drainage
improvement (Figure-4), Area 3&4 North and South
Rindu production-injection performance comparison
(Figure-7), and Rindu production-injection profile
after drainage improved (Figure-11).
3. Cross-flow and scaling problem data
This data consist of Rindu Vertical Expansion (RVE)
completion type (Figure-1), cross-flow problem
example (Figure-9), and scaling problem example
(Figure-8).
4. Saturation and temperature data
The saturation data is taken from saturation log in the
observation well, open hole log from new drilled
well, and calculated saturation from Neutron-Porosity
log in the observation well. The temperature data is
taken from surveillance program in all observation
wells. The data consist of the example of No Heating
Response and low saturation changes in TOW before
drainage improvement (Figure-4), the example of
heating response after drainage improvement (Figure-
11), and Rindu temperature map (Figure-4).
5. Reservoir pressure data
This data is taken from RFT log of new drilled well
(Figure-5.a) and combined with shut in casing
pressure of Rindu injectors and calculated reservoir
pressure from drilling mud weight data (Figure-5.b).
The analysis of this study is conducted through
production performance analysis, heat efficiency
evaluation on each injected pattern, multiple-year
saturation changes review, fluid scaling tendency,
reservoir properties control to production-injection
performance review, Rindu producer density review, and
reservoir pressure evaluation thru recent drilling
activities. These comprehensive reviews carry out to
address all Rindu reservoir problems in the southern part
of Area 3&4. The flow chart was generated to simplify
the study analysis conducted by project team as shown in
the Figure-10.
Discussion
This section is representing data analysis result that focus
to address the underperformance Rindu reservoir.
Throughout assessment of problematic Area 3&4 Rindu
development, there are four main factors that lead to poor
performance of Rindu reservoir as follows:
1. Reservoir Quality Controls Steam flood
Performance
The reservoir quality in Area 3&4 is very heterogenic in
term of permeability and net pay. Although the Rindu
SPE 147706
porosity of Area 3&4 spread out homogeny by average
30%, but the southern part of Area 3&4 has lower quality
compare to the northern part. The Rindu reservoirs in this
location have less thickness, less permeable and tend to
vertically separated (multiple lobe sand). As part of
channel edge deposits, this area has more shale content
compare to northern area.
The lower quality of reservoir in the study area controls
the amount of steam injection. The lower sand quality has
limit the amount of steam injection rate related to lower
permeability and high reservoir heterogeneity (vertically
and horizontal). The average steam injection in southern
part only 200 BSPD while in the northern part could reach
1500 BSPD. As the consequence, the northern part
reaches faster maturity and better production performance
compare to the southern part (Figure-7).
2. Commingle Completion Implication to Production
Performance
The commingle completion of Rindu and PK reservoir
allow a direct contact of two different reservoir behavior
in the same borehole. Understanding this interaction is a
critical study to explain the well behavior. This study
reveals three reservoirs characteristic such as fluid
properties, reservoir pressure and temperature that
controls the wells performance.
The temperature data play important rule on scaling
problems. The Rindu commonly has colder temperature
compare to PK reservoir. This temperature different is the
best environment that allow both reservoir to create scale.
A previous study on comparison of Rindu and PK fluid
properties concludes a different fluid chemical
composition in both reservoirs. Based on laboratories test,
the mixtures of both reservoirs’ fluids tend to create scale,
especially in a contrast temperature borehole
environment. Since Rindu and PK are produced
comingled in RVE wells, the scaling problem is common
found in the wells. It shows by a steeper decline of oil and
total fluid production after Rindu started produced
commingle with PK for a while. As direct evidence to this
study, Area 3&4 RVE look back found severe carbonate
deposits inside liner during work over or well service
activity (Figure-8). The fluid analysis of related wells also
shows from water sample of the RVE wells, 70%
indicates tendency of scale deposition in Rindu sand.
Among of them have been treated using hydrochloric acid
and have been successfully increased the production.
However, after a while, the scaling problem returned back
and holds on the production again. The scaling problem
was repeated until the end life of well (Figure-8).
In addition to scaling problem, the reservoirs pressure
different is also significantly impacting the oil production
from RVE wells. This review conducting by observed the
pressure data contrast in low performance RVE wells.
The recent pressure data from new drilled well shows
reservoir pressure in Rindu become 1.5-3 times higher
than PK sand. This finding implies the occurrence of
3
cross-flow or pressure hold up from Rindu to PK sand.
This indication became stronger when the team found the
steeper production decline trend after opened the Rindu
intervals in the existing PK wells (Figure-9).
3. Poor Drainage Point Implication into Rindu
Production-Injection Performance and Reservoir
Pressure
The pattern steam flood mechanism will be working if
injection and withdrawal (reservoir drainages) is balance.
Both injector and withdrawer (producer wells) play
important rule to maintain gravity drainages. The
withdrawal (reservoir drainages) efficiency is function of
number wells in the pattern and also the completion types.
The Rindu injection in the study area has been started
since 1997. The injection is targeting the RVE wells
patterns with no Rindu dedicated producers. Moreover,
the RVE wells completed as cased-hole gravel pack. This
completion type reduced the inflow area and impacted to
lower productivity compare to the open-hole completion.
The Rindu initial injection rate is low (~averages 200
BSPD) that controlled by lower sand quality and lower
withdrawal compare to northern Area 3&4 Rindu area.
Similar to other Rindu well, the injection rate is expected
to increase in alignment to increasing of reservoir
depletion (reduced the reservoir pressure). After more
than 10 years, the injection rate remains the same. This
injection rate is never improved because no pressure
depletion occurred related to production. The depletions
never happen because poor drainages of pattern steam
floods.
After several years facing with drainage problem, the
continue of low rate Rindu injection lead to pressure
accumulation up to 1.5 – 2 times of its original pressure
(Figure-5a). This review results is proved by number
drilling result that shown a significant high pressure in the
cold Rindu reservoirs. The high reservoir pressure
conditions contributes to low steam injectivity and
limiting the production performance and pattern recovery.
4. The Relationship of Injection Performance and
Heating Efficiency
In relation to previous point discussion, the Rindu
reservoir could only injected with averages 200
BSPD/pattern. This rate is lower from injection target
(about one fifth of its target rate) and is not sufficient to
heat up the reservoir (heat loss is equal to heat injected).
With condition, the pattern heating time will be reached at
least 75 years from target 3 years. It is extremely longer
than expectation and yielding a failure in operation and
project economic. Several measurement data have
confirmed this calculation result such as temperature
survey and saturation log in observation wells (Figure-4b)
that remain the same (cold and no depletion) after 10
years injection. The surface measurement data also
indicates a cold wellhead temperature and production
profile is following primary production trend or no steam
4 SPE 147706
flood impact (Figure-4b). With the current practice, this
area will never performed in term of production and
injection as planned. The longer heating time means poor
reservoir heat management that leads to failure in steam
flood project economic.
Results
Throughout previous discussion, this study clearly define
that the poor reservoir quality, scaling/cross-flow
problem, low drainage point, poor injectivity lead to
longer maturity time and un-optimized Rindu recovery
across the field. A unique solution have been proposed to
resolve above concerns by drill separate Rindu open-hole
gravel pack producer in adjacent to the existing RVE
wells and complete pattern drainage point. To maximize
steam injection and maintain reservoir pressure, all in-
efficient injection (below 200 BSPD) is recommended to
shut in until drainage improvement or reservoir pressure
is reached.
The drilling new Rindu producer is aiming to improve oil
recovery and reduce accumulated pressure in the reservoir
at the same time. After pressure depleted to sufficient
level for injection, the steam injection rate could be
increased to ideal target.
This recommendation has been implemented into a series
of drilling package. A number of drilling activities has
faced high pressure that confirmed the presence of
accumulated pressure. The shut-in of poor injector is also
applied as one of strategy to reduce additional
accumulated pressure and it helps drilling execution. The
additional producers prove depleting reservoir pressured
significantly after 3-6 months production. As implication,
project team succeeds to increase averages injection rate
to 1000 BSPD. The significant injection improvement
leads into inclining production to 50 BOPD/well of target
area (Figure-11). Heating efficiency also improved
rapidly as shown in temperature growth in observation
and producer wells (Figure-11).
Conclusion
As conclusion to this integrated study, some key lesson
learns and finding on steam-flood operation and practice
are summarized as follows:
1. The comingle completion practice is not
recommended for a significant different reservoir
condition. This study shows that Rindu and PK
have significant reservoir properties different
especially in reservoir quality, pressure and
temperature that lead to cross-flow and scaling
problems
2. The steam flood injection design will only work
when the pattern has a complete drainages point
and sufficient reservoir pressure. This case study
shows that the low drainages point in Rindu
pattern and the low injection over 8 years yields
to the high pressure accumulation in reservoir
3. The reservoir quality strongly controls the
pattern injection performance. This study reveals
that Southern Area 4 Rindu sand has a poorer
reservoir quality compare to Northern part. The
injection performance also shows significant
lower in southern part compare to northern Area
3&4.
4. The time to reach maturity and pattern maturity
stages controls the reservoir production
performance. Failure to reach the maturity target
in time resulting a underperform production and
economic.
5. This study shows that drilling dedicated Rindu
open-hole producer is improving oil recovery
and injection
Recommendation
This study recommends continuing re-develop Rindu
reservoir that has the similar problem across Duri field
by:
1. Drilling separate Rindu Open-hole gravel pack
close to the existing RVE wells.
2. Increase injection after several month
productions/ after reservoir pressure decreased.
Acknowledgement
Special thanks to my team (Dijan Prijambodo, Milla
Amlan, Ardi Nazamzi, and Yustinawati) for their
contributions to complete this paper.
Reference
1. Neumann, C. H.: “A mathematical model of the
steam drive process, Application”, Paper SPE
4757.
2. Johansen, S. and Habash Semimbar. : “Tide-
dominated deltaic systems of lower Miocene,
Central Sumatera Basin”, AAPG Hedberg
Conference, 2009, Jakarta, Indonesia
SPE 147706 5

Figure. 1. Rindu Vertical Expansion (RVE) Completion Type

Normalize Production
Group: AREA-04RVE, PID: DURI-AREA-04-RVE
Group: AREA-04RVE Entities In Group: 173 Format: [p] BG - Monthly Normalize

200000(L1) 34000(R1)

187000 31500

174000 RVE 29000

161000 26500

148000 24000

135000 21500

122000 19000

109000
Rindu=17% 16500

96000 PK 14000

83000 PK=83% 11500

70000 9000
-60 -48 -36 -24 -12 0 12 24 36 48 60

(L1) CDALLOC_FLUID (R1) CDALLOC_OIL

VS NORM_MONTH

Figure. 2.a. Actual Rindu Reservoir Contriution in RVE Producers

6 SPE 147706

Result of Oil Finger Printing

% Contribution
Well Name
R PK
6M61A 15 85
6M52B 15 85
6M53A 27 73
6M63A 15 85
5M79A 15 85
5M58B 8 92
5M59A 21 79
5M67A 23 77
7M58B 39 61
7M67A 24 76
7M69A 4 96
8M53B 20 80
6P64A 14 86
6P55A 29 71
6P35A 4 96
6P44A 20 80
6P43A 2 98

Figure. 2.b. Actual Rindu Reservoir Contriution in RVE Producers from Oil Finger Printing

4 5 6 7

1
1. Primary Production (PK) (1985)
2. Steam Flooding Event (PK) (1988)
3. Shallower Sand Pilot Project (Rindu) (1994)
4. Rindu Vertical Expansion (RVE) Project (1997 – 2000)
5. Rindu Development (2000 – 2001)
6. Rindu Sweet Spot (2007 – 2008)
7. Rindu Re-Development (2009 – 2011)

Figure. 3. Area 3, 4 Historical Production-Injection Performances

SPE 147706 7

Existing
Rindu Producer

RVE Producer

Figure. 4.a. Rindu Poor Injection Performance

TPHI-SIGMA
RST_2010_JUN.TPHI_JUN10_1
Temperature
0.6 V/V 0
TDT2009_OCT.TPHI_2
0.6 V/V 0
Steam
GR-SP-
WELL : 5P-12BResistivity
TDT2007_EVEN.TPHI_2007AUG_1
0.6 V/V 0
chest-Oil
NUMBER : 2B #4262
Caliper
TDT2006_ODD.TPHI_2
0.6 V/V 0 Saturation
RTE : Ground Level TDT2005_EVEN.TPHI_1

STATUS : P RHOB- 0.6 V/V 0

TDT2007_EVEN.TEMP_2007AUG DEGF

RST2005.TPHI_1

NPHI 0.6 V/V

TMDL2004_JUL.PHIT_2
0
TEMP_2010.TEMP_2010OCT_1
TEMP_ALL.TEMP_MAR01 DEGF

TEMP_ALL.TEMP_MAR10 DEGF
TEMP_ALL.TEMP_MAR10 DEGF
TEMP_ALL.TEMP_NOV97 DEGF

TEMP_ALL.TEMP_MAY00 DEGF

TEMP_ALL.TEMP_MAY02 DEGF

TEMP_ALL.TEMP_OCT09 DEGF
TEMP_ALL.TEMP_APR03 DEGF

TEMP_ALL.TEMP_APR06 DEGF
TEMP_ALL.TEMP_SEP99 DEGF

TEMP_ALL.TEMP_SEP08 DEGF

0.6 V/V 0 0 DEGF 400

TDT2005_EVEN.TEMP DEGF

TMDL2003_APR.PHIT_2 TEMP_ALL.TEMP_JUN10_1
0.6 V/V 0 0 DEGF 400
TDT2002_MAY.TPHI_2 TEMP_ALL.TEMP_MAR10_1
0.6 V/V 0 0 DEGF 400
PHIE_DURI

EVAL.PAYFLGD_1 TDT2001_ODD.TPHI_2 TEMP_ALL.TEMP_OCT09_1

0.1 500 0.6 V/V 0 0 DEGF 400
EVAL.VSH_1 WIRE_RAW.LLS_1 WIRE_RAW.CALI_2TDT1999_ODD.TPHI_1 TEMP_ALL.TEMP_SEP08_1
KAIR

0 V/V 1 0.2 OHMM 2000 0 IN 10 0.6 V/V 0 0 DEGF 400

DEPTH TVDSS
EVAL.GR_1 FEET FEET WIRE_RAW.MSFL_2 WIRE_RAW.RHOB_2TDT1997_EVEN.TPHI_1997NOV_1 RST_2010_JUN.SO_1 CHEVAL.SO3P_PLUS_JUN10_1TEMP_ALL.TEMP_AUG07_1
0 GAPI 200 0.2 OHMM 2000 1.65 G/C3 2.65 0.6 V/V 00 V/V 10 V/V 10 DEGF 400
WIRE_RAW.CALI_2 WIRE_RAW.LLD_2 WIRE_RAW.NPHI_2 WIRE_RAW.RHOB_2 EVAL.SW_1 EVAL.SW_1 TEMP_ALL.TEMP_APR06_1
10 IN 0 0.2 OHMM 2000 0.6 V/V 0 1.65 G/C3 2.65 1 V/V 01 V/V 00 DEGF 400

B_U_RN

T_RN1

300

-250

350
B_RN1
T_RN2 -300

B_RN2

T_RN3
B_RN3

T_RN4 400

-350

B RN4

Figure. 4.b. Rindu Poor Temperature Response

8 SPE 147706

Rindu Formation Pressure Vs Time
240

220 y = 0.0276x ‐ 929.18

R² = 0.7094
200

180

160
Pressure, psi

140

120

100

80

60

40
Jan‐02 Jan‐03 Jan‐04 Jan‐05 Jan‐06 Jan‐07 Jan‐08 Jan‐09 Jan‐10 Jan‐11

Figure.5.a. High Pressure Phenomenon in Rindu Reservoir from Historical RFT

170 80

130 140

210

220 190 230

100 180
185

140 200 160

150 SICP vs Approx. Depth
280
130 120
270
260
250
240
230
220
210
200
Reservoir Pressure Psi

190
180
170
160
150
140
130
120
110
100
90
80
70
190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360
11
Depth TVDSS

9 10 11 12 13 14 15 SICP

Figure.5.b. High Pressure Phenomenon in Rindu Reservoir from Recent Injector Shut in Casing Pressure (SICP)
SPE 147706 9

Figure.6. Static Reservoir Properties

10 SPE 147706

Permeability Map 700
Production‐Injection Performance
Northern Part
2000

1800
600

North
Northern Part
500
1600

1400

1200
400

BOPD

BSPD
1000
300
800

200 600

400
100
200

0 0

South
Southern Part Jan‐00 Jan‐01 Jan‐02 Jan‐03

Oil_Allo
Jan‐04 Jan‐05 Jan‐06

Actual Steam
Jan‐07 Jan‐08 Jan‐09

Target Steam
Jan‐10

Production Injetion Performance
Sothern Part
270

800
240

700 210

600 180

500

PSI/BOPD
150
BSPD

400 120

300 90
• Northern part: injection= 1500 BSPD/pattern, Oil  200 60
production=500 BOPD/Pattern
100 30
• Southern part: injection= 200 BSPD/pattern, Oil 
0 0
production <30 BOPD/Pattern May‐02 May‐03 May‐04 May‐05 May‐06 May‐07 May‐08 May‐09

Injection Rate (BSPD) Injection Target (BSPD) Actual WHP (Psi) MAX WHP (Psi) OIL

Figure.7. Northern Vs Southern Part Area: Rindu Production-Injection Performance Comparison

4P-49A

Figure.8. Scaling problem in Rindu Vertical Expansion (RVE) Producer