SPE 59335 Increased Oil Production From Bachaquero-01 by Steamflooding Using Horizontal Wells

M. G. Rodriguez, SPE, PDVSA, and D. D. Mamora, SPE, Texas A&M U.

Copyright 2000, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2000 SPE/DOE Improved Oil Recovery Symposium held in Tulsa, Oklahoma, 35 April 2000. 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, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes 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 where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

injection wells with horizontal sections of 1280 ft to 1560 ft long were infill-drilled in 1995-1997 in areas of the reservoir containing vertical cyclic steam injectors. Three separate simulation studies were performed to evaluate the performance of the three horizontal wells under cyclic steam injection and steamflooding. Dimensions of the Cartesian models used were 11224, 11275, and 1220 6. Reservoir Description Geologically, the Bachaquero-01 sandstone reservoir has been divided into nine intervals, namely (from top to bottom), Arena Principal, HH, GG, FF, EE, DD, CC, BB, and AA (Fig. 2). Arena Principal contains 75% of the reservoir OOIP, while the upper four intervals jointly contain 95% of OOIP. Arena Principal is the thickest and most important reservoir interval. Thickness can exceed 200 ft and values of 150 to 200 ft are very common. Excellent lateral and vertical continuity are evident within this interval. Arena Principal consists mainly of well-developed, highly porous and permeable point bar and braided stream sands with alluvial sediments being present. The HH is the next thickest interval. Net oil sand thickness rarely exceeds 90 ft with an average of about 55 ft. Lateral continuity in the overall extension of the HH interval is good. Vertical continuity is complex and generally poor. Overall depositional setting is that of an alluvial meander belt. Net oil sand thickness in the GG interval is quite variable, ranging from less than 20 ft to more than 100 ft, averaging about 40 ft. Lateral continuity is judged to be only fair while vertical continuity is generally poor. A moderate-to-lowenergy alluvial complex dominates the GG interval. Net oil sand thickness of the FF interval averages about 20 ft but is quite variable, ranging from zero to more than 60 ft. Lateral continuity as well as vertical continuity are poor. This interval consists predominantly of low-energy alluvialplain sediments. The remaining intervals, EE through AA, are generally thinner with increasing lateral variability of sand quality towards the lower intervals. The water-oil contact (WOC) cuts across the FF through AA intervals, leaving the

Abstract Three-dimensional thermal compositional simulation studies were conducted to evaluate the performance of three horizontal wells under cyclic steam injection and steamflooding in the Bachaquero-01 heavy oil reservoir, Venezuela. In the steamflooding scheme investigated, the horizontal wells were used as injectors with existing (and new) vertical wells as producers. Simulation results indicate: (i) oil recovery of about 15% of original-oil-in-place with cyclic steam injection compared to about 25% under steamflooding with no new producer, and about 50% under steamflooding with additional producers, (ii) main benefits of steamflooding are in re-pressurization and improved thermal efficiency, and (iii) higher oil recovery with additional wells result from improved areal sweep efficiency. Introduction Located in the eastern coast of Lake Maracaibo, Venezuela (Fig. 1), the Bachaquero-01 heavy oil reservoir lies at about 3000 ft. ss. and contains some 7 billion STB of 11.7 degrees API gravity oil with an in-situ viscosity of 635 cp. Cold production began in 1960, but since 1971 the field was produced under a massive cyclic steam injection system. Todate some 370 cyclic-steam injection wells have produced from the field, yielding an oil recovery of only 5.6% of original-oil-in-place (OOIP). The reservoir pressure has dropped from an initial 1370 psia to its present value of about 700 psia. Oil production peaked at 45 MSTB/D in 1991, and has since declined to its current level of 40 MSTB/D. To arrest production decline, three horizontal cyclic-steam

M.G. RODRIGUEZ, D.D. MAMORA

SPE 59335

uppermost Arena Principal with the largest oil-bearing areal extent and the AA with the smallest areal extent. Holbrook and Moretti1 showed that vertical communication between the intervals is not present everywhere but occurs in a sufficiently widely distributed number of places. For engineering purposes, individual intervals may be considered to be in communication and to act as a reservoir unit. This possibility is taken into consideration in constructing the numerical model. Table 1 summarizes the rock and fluid properties of Bachaquero-01. Simulator Schlumberger-GeoQuests simulator, ECLIPSE 500 version 98A,2 was used in the study, in part because it is PC-based. It is a fully implicit, n-component thermal simulator with both black oil and compositional options. This simulator also models oils with non-volatile components. Three phases are allowed in the simulation: water phase (which only contains water), an oil phase (containing anything but water), and a gas phase that can only contain steam and hydrocarbon components. Recovery processes such as steam drive, steamassisted-gravity drive (SAGD), hot water injection, single cyclic steam injection, infill drilling, horizontal wells, and pattern floods can be simulated using ECLIPSE 500. Overview of Simulation Models Three separate models were constructed to simulate the areas where the three cyclic steam horizontal wells are located. These model areas were named as follows: (i) Area LL125 where the re-entry well LL125 was drilled and completed on, (ii) Area LL3343 which contains horizontal well LL3343, and (iii) Area LL3487 containing horizontal well LL3487. In addition to the horizontal wells, vertical wells were included in the areas modeled for proper history matching and reservoir drainage considerations. Analysis of a sample taken from well LL25933 indicates the oil is composed of 61.54% of heptane-plus and 29.71% of methane. Thus, it was decided to run simulation using the two-component thermal option. The reservoir fluids in the model therefore consist of three phases, namely, vapor, oil, and water. The vapor phase may contain steam and methane while the oil phase contains heptane-plus and methane. Most of the grid dimensions for Area LL125 and Area LL3343 were based on a previous model.4 Local grid refinement was used around the horizontal wells and vertical wells. Model layer thickness was based on inter-well geological sections and logs. Empirical correlations were developed to obtain permeability as a function of porosity, based on regression analysis. Due to scarcity of log data from well LL2318,5 a satisfactory fit was not obtained. It was thus decided to further divide the intervals into sub-intervals to improve the correlations. Since Arena Principal contains 75% of OOIP, attention was focused on this interval. Based on permeability data and deep-induction resistivity log results, Arena Principal was divided into three sub-intervals, AP-1, AP-2 and AP-3.

For each model sub-intervals AP-1, AP-2, AP-3 and intervals HH and GG were identified and incorporated. Rock properties and relative permeability data were based on core analysis. Four sets of relative permeability data were available and were used in each of the sub-intervals or intervals. The relative permeability curve for each layer was selected as a function of the effective permeability measured for the plugs tested. Fluid properties were based on PVT analysis of a sample taken from well LL525.6 In compositional simulation, a number of parameters are required to enable proper calculation of phase composition and partitioning. For each component, these are as follows: heat of vaporization, critical temperature, critical pressure, boiling point temperature, acentric factor, and interaction coefficient. These physical properties were calculated based on the method developed by Whitson.7 These properties corresponded to single component number (SCN) 37, which was selected as a function of the molecular weight of the oil sample from well LL2593. During the study, ECLIPSE module PVTi was not available. With PVTi it would have been possible to directly calculate the physical properties of the heptane-plus given the available PVT/compositional data. Thus, it was necessary to make sensitivity runs to determine which physical properties had the greatest effect on the results. It was found that the molecular weight had a dominant effect on the results. ECLIPSE 500 assumes the steam quality to be that at the sand face, while in the field this parameter is measured on surface. Wellbore heat loss calculations were thus made using Satter's method8 to estimate steam quality at reservoir depth. This procedure consists of dividing the length of the wellbore into several intervals. Using thermal parameters defined for each interval, the heat losses from and temperature drop across each interval then are summed to obtain the heat losses to and temperatures at any depth as a function of time. Well completion intervals were properly taken into consideration by adjusting the kh product in the well model. The skin factor for each well was estimated by historymatching the well flowing bottom hole pressure (FBHP). Area LL125 Model As for wells in other parts of Bachaquero-01, the wells in this model are all completed in Arena Principal which is of fluvialdeltaic origin of good lateral continuity. Area LL125 is composed of horizontal well LL125 and four neighboring vertical wells (LL2296, LL2404, LL2435 and LL3178). The oldest well, LL125, started production in January 1969. A horizontal sidetrack was drilled in 1995. Wells LL2296 has produced since 1980, wells LL2404 and LL2435 since mid1980s, and LL3178 since September 1991. All the wells have produced under cyclic steam injection. The original pressure of this area was about 1370 psia (the original reservoir pressure). However, the reservoir pressure of this area has since dropped to about 700 psia, an average pressure decline of 20 psi/year. This area is probably the most depleted area in Bachaquero-01.

SPE 59335

INCREASED OIL PRODUCTION FROM BACHAQUERO-01 BY STEAMFLOODING USING HORIZONTAL WELLS

Simulation Model. The Cartesian model of Area LL125 has grid dimensions of 11224, the areal dimensions being the same as those used in a previous study. 4 The thickness of each of the four layers was based on the average thickness of subintervals AP-1, AP-2, AP-3, and interval HH. Table 2 shows measured and average properties for each layer. Fig. 3 presents the areal grid dimensions. History Matching. History matching was carried out for the producing period of Area LL125, January 1969 to March 1998, approximately 30 years of history. This period was modeled by specifying the historical oil production and cyclic steam injection rates and allowing the model to calculate water and gas production rates. The history-matching procedure used in this study was as follows. First, the reservoir pressure was matched. This was achieved by adjusting the value of rock compressibility to 12010-6 psi-1, indicating significant compaction drive being present. Figs. 4 and 5 show the results of the satisfactory history match of reservoir pressure and oil production rate, respectively. Second, the water production rate was history-matched by adjusting the water relative permeability curve (krw) and the irreducible water saturation (Swi ) for each layer. To achieve a satisfactory history match krw was reduced considerably from the measured krw. Third, a satisfactory gas production history-match was achieved by setting critical gas saturation (Sgc) to 0.05. Finally, FBHP's were matched by adjusting the skin factor. Area LL3343 Model Model Area LL3343 contains sand of deltaic-origin. The WOC is at 3,062 ft. (near the reservoir base). The nearness of the WOC to completion intervals is one main reason for this area to be less developed. Moreover, this part of the reservoir shows higher pressure and has declined at 18 psi/year. Area LL3343 includes the vertical wells LL2366, LL2610, LL2781 and LL2788, and the horizontal LL3343. The vertical wells are completed as open-holes across intervals Arena Principal, HH and GG and have been producing since 1983. All the wells have produced under cyclic steam injection except well LL2781. Horizontal well LL3343 is completed on the Arena Principal and has produced under cyclic steam injection since August 1995. Simulation Model. An 11275 Cartesian model was used for Area LL3343 (Fig. 6). Areal dimensions were based on a previous work.4 The thickness and reservoir properties of each layer were estimated from log data, following the same procedure used for Area LL125 (Table 3). History Matching. History matching for Area LL3343 was performed from start of production in mid-1980. For history matching, the procedure used was the same as that for Area LL125, except for some differences related to the characteristics of Area LL3343, such as a different WOC (3,062 ft). It was thus necessary to add a bottom layer (layer GG with a thickness of 70 ft). To simulate water coning as

observed in wells LL2781 and LL2610, vertical-to-horizontal permeability ratio was increased by a factor 10 in the lowest block for each well. Based on logs from well LL2610 and LL278, a water-oil transition zone exists for each of these wells, as described by core capillary pressure data. After obtaining an unsatisfactory history match, it was decided to use capillary pressure based on the Leverett function.9 Eventhough these values were higher than those from core analysis, a good match was not obtained. The capillary pressure values were then increased by a factor 10 to yield a better history match. In all the runs radial local grid refinement around these two wells was used. Figs. 7 and 8 show the satisfactory history match for Area LL3343. Area LL3487 Model Located in a poorly developed part of Bachaquero-01, Area LL3487 has produced since 1955. The reservoir pressure in this area has declined at an average of 10 psi/year. Area LL3487 is of deltaic origin with the WOC at 3,370 ft. Area LL3487 includes five vertical wells (LL36, LL112, LL160, LL2527 and LL2849) and horizontal well LL3487. Well LL160 has produced since 1955. The wells are completed as open-holes in the Arena Principal, HH, and GG. Only wells LL2527 and LL2849 have been cyclic steamed. In May 1997 horizontal well LL3487 was drilled in Arena Principal and thereafter produced under cyclic steam injection. Simulation Model. A 12205 Cartesian model was used for Area LL3487 (Fig. 9). Areal dimensions were the same as those used in Area LL125 and Area LL3343. Layer thickness was estimated from logs (Table 4). The estimated drainage shape for Area LL3487 (not a rectangle) required defining certain grid blocks to be inactive blocks. History Matching. History matching for Area LL3487 followed the same procedure used for Area LL125 and Area LL3343. Special mention is made of the difficulty in history matching the rapidly increasing water cut observed in wells LL160 and LL2527. History match was improved by adjusting the water relative permeability curve. Radial local grid refinement was used around these wells. Figs. 10 and 11 show the best history match results. Forecasting Simulation runs were conducted to predict the performance of the three areas under cyclic steam injection and steamflooding. For each of the cases studied, the well FBHP (pwf) was kept constant as well as the skin factor. The prediction runs were made for a period of 20 years. The steam injection rate was based on the most recent injection rate for each well and was kept constant. The steam quality at the reservoir face was kept constant at 75%. Results for each case were compared on the basis of cumulative oil production. Horizontal wells provide greater injectivity than vertical wells. However, inverted vertical well patterns are in most

M.G. RODRIGUEZ, D.D. MAMORA

SPE 59335

cases more suitable for the offshore conditions in Maracaibo Lake. Therefore, for the steamflooding cases considered, the horizontal wells were used as injectors while existing and new vertical wells were used as producers. In general, the following cases were run for each area except where they are not applicable due to the geometry of the area: Case 1: production from existing wells with no further cyclic steam injection. Case 2: continuing cyclic steaming in existing wells until a cumulative 10 cycles for each well. Case 3: steamflooding with the horizontal well as injector and the existing vertical wells as producers. Case 4: steamflooding with the horizontal well as injector and new horizontal producers. Case 5: steamflooding with the horizontal well as injector surrounded by eight new vertical wells. The results of the prediction runs for each simulation area are discussed in the following. Area LL125 FBHPs were set at 375 psia for vertical wells and 500 psia for the horizontal well. These values are based on FBHP surveys taken in the area. These parameters were kept constant during the period under consideration. Table 5 shows the results of the prediction cases considered. The incremental recoveries for Cases 2-5 are compared against that of Case 1. Note that cumulative oil production as April 1, 1998 amounts to 12.2% of OOIP. It can be seen that the oil recovery factor for Case 3 is almost twice that of Case 1, showing the benefits of steamflooding. Moreover, the steam-oil ratio (SOR) of 1.2 for Case 2 indicates that cyclic steam stimulation will no longer be efficient when this value is compared to the overall historical value of 0.2 for this area. The simulation results for Case 3 indicate that there is an increase in oil production eight months after the start of steamflooding. Production rate continues to increase until a peak of 1,400 STB/D in August 2003. Moreover, the reservoir is also repressurized (Fig. 12). Case 4 considers four additional horizontal producers: two parallel to horizontal injector LL125 and two orthogonal to it. Several runs were made to find the optimum configuration (in terms of length and position of the horizontal section) to give the highest oil production. With this optimum pattern, prediction runs were made. Results of simulated reservoir pressure and production profiles show that the reservoir pressure remains almost constant at about 700 psia during the prediction period. Case 5 involves eight vertical producers and the horizontal injector well. The eight vertical wells replace the four horizontal producer wells in Case 4. Several simulation runs were made to find the best distribution of the wells to yield the highest oil production. Wells located at the corners of the area were completed in the bottom layer (HH) to delay steam breakthrough, while wells further away from the injector were completed across the whole reservoir.

The reservoir pressure in this case increases constantly until the end of the prediction period. The maximum oil rate is reached in the year 2007 and is less than that obtained in Case 4. The recovery factors for Case 4 and Case 5 are the same, 53% of OOIP, indicating sweep efficiency by the four horizontal producers and eight vertical producers are similar. Area LL3343 Location of horizontal well LL3343 and existing vertical wells do not permit the addition of a horizontal producer parallel to well LL3343. The following cases were therefore considered for this area. Case 1: production from existing active wells with no further cyclic steam injection. Case 2: continuing cyclic steaming in the existing active wells until a cumulative 10 cycles for each well. Case 3: steamflooding with the horizontal well as injector and the existing active vertical wells as producers. Case 4: work over the inactive well LL2366, drill new vertical producer wells around the horizontal well, and cyclic steam all wells. Case 5: steamflooding with the horizontal well as injector surrounded by vertical existing, new and worked over wells. FBHP's were set to 475 psia for vertical wells and to 600 psia for horizontal wells. Table 6 summarizes the simulation results. In Case 1 the oil recovery factor is 10.9% of OOIP. It can be seen from the results of Case 2 and Case 4 (cyclic steam injection schemes) that there is a substantial increase in the SOR (1.9 and 1.5, respectively) considering that the historical average is 0.42. Also, steamflooding increases the cumulative oil production at an acceptable SOR of 2.1. Fig. 13 presents the results for Case 3. The reservoir pressure increases sharply in the next five years to about 1,360 psia that is thereafter maintained. Furthermore, the production rate increases to a peak of 1,170 STB/D in 2008. Steam breakthrough is not observed in the prediction period because the reservoir pressure is relatively high and therefore the injection rate low. Case 4 involves two new vertical wells and a worked over producer (well LL2366) which has been closed in due to sand production. Case 5 is similar to Case 4 except that the two new vertical wells and well LL2366 are completed in layer HH to avoid early steam breakthrough. Area LL3487 In this area the following cases were studied. Case 1: production from existing active wells with no further cyclic steam injection. Case 2: continuing cyclic steaming in the existing active wells until a cumulative of 10 cycles for each well. Case 3: steamflooding with the horizontal well as injector and the existing active vertical wells as producers. Case 4: cyclic steam all wells including worked over wells that have been closed due to sand problems.

SPE 59335

INCREASED OIL PRODUCTION FROM BACHAQUERO-01 BY STEAMFLOODING USING HORIZONTAL WELLS

Case 5: steamflooding with the horizontal well as injector and all vertical producers. Case 6: steamflooding with the horizontal well as injector and four new horizontal producer wells. Case 7: steamflooding with the horizontal well as injector surrounded by eight new vertical producers. For this area, FBHP's were set at 475 psia for vertical wells and 600 psia for horizontal wells. Table 7 shows the results for all the simulation cases studied. It is observed that steamflooding not only increases the oil recovery factor but also improves the SOR to values that are as good as that for cyclic steam injection. In Case 2, with continued cyclic steaming, the overall SOR increases from 0.3 to 1.8. In Case 3 where only the active vertical wells are considered, there is early steam breakthrough in well LL2527, due to its closer proximity to the horizontal injector. Oil recovery factor is double that for Case 1. Fig. 14 shows the simulated oil production results. In Case 4 (cyclic steaming) and Case 5 (steamflooding), the vertical wells were assumed to be producers, including worked over wells LL36 and LL112 which have been closed in due to sand problems. To delay steam breakthrough in well LL2527, this well was closed in during steamflooding and kept open during cyclic steam injection. The recovery factor for steamflooding (32.1% in Case 4) is about double that for cyclic steam injection (13.9% in Case 5). SOR in both cases are about the same, 2.0 - 2.4. New producers around the horizontal injector are included in Case 6 and Case 7. Case 6 involves four new horizontal producers, two of them parallel to the horizontal injector and two horizontal producers perpendicular to it. Case 7 involves eight new vertical wells around the horizontal injector. In Case 7 the side wells were completed in the bottom layer while the corner wells were completed in all layers to improve oil recovery. In both cases the simulated reservoir pressure remains constant at about 750 psia while the oil recovery is increased to 40% of OOIP (Table 7). Summary and Conclusions A simulation study has been conducted for three areas of the Bachaquero-01 heavy oil reservoir in Venezuela. Each of these areas contain a recently drilled horizontal well which like most vertical wells in the field - have produced under cyclic steam injection. The main objective of the simulation study was to compare the performance of cyclic steam injection against that of steamflooding. Simulation was conducted using the threephase, two-component oil and thermal compositional option in the numerical simulator, ECLIPSE 500. The threedimensional Cartesian models used had grid dimensions of 11224, 11275, and 12205. The following main conclusions can be drawn from the simulation results. (1) For the three areas studied, steamflooding using existing wells increases the oil recovery factor to about twice that under cyclic steam injection at the end of 20 years of

production. The oil recovery factor increases from about 1216% of OOIP with cyclic steam injection to about 23-33% of OOIP with steamflooding. (2) Continuation of cyclic steam injection would not only result in lower oil recovery compared to steamflooding but also result in increasingly less efficient thermal operations with SOR increasing from 0.2-0.4 to 1.2-1.8. (3) Repressurization of Bachaquero-01 reservoir is observed as a result of steamflooding, which contributes to enhancing oil production rates and ultimate recovery. (4) Steamflooding results in an increase of the SOR to 2.2-4.2 which are still excellent values for steamflood projects. (5) When additional vertical wells (new or workovers) or horizontal wells are included as producers, with a horizontal injector in a steamflood scheme, oil recovery factor as high as 53% of OOIP are obtained. This recovery factor is in line with that observed in the largest steamflood project in the world, the Duri field, Indonesia.10,11 The potential gain with steamflooding (53% of OOIP) over cyclic steam injection (about 15% of OOIP) for the Bachaquero-01 is large - about 2.7 BSTB. Recommendations Based on the results of this study, the following recommendations are made. (1) Given the very encouraging simulation results for switching from cyclic steam injection to steamflooding for the Bachaquero-01 reservoir, it is recommended to conduct a steamflood pilot test in Area LL125 using existing wells. Area LL125 has been chosen as there would be minimum capital layout, and because of the rapid increase in oil production expected (less than a year). (2) If the steamflood pilot test in Area LL125 were successful, it is recommended to continue with steamflooding in Area LL125 and to conduct similar pilot tests in the other two areas. (3) Most of the existing 300 vertical well producers on the Bachaquero-01 reservoir have been specifically located in anticipation of future steamflood using seven-spot patterns. Based on the results of this study, it is recommended to conduct a study of steamflooding for the whole Bachaquero01 reservoir particularly in the use of horizontal wells (or sidetracks) as injectors. Acknowledgments The authors wish to thank PDVSA for kind permission to release data for use in the study and for publication of the study results. References 1. Holbrook, P. W. and Moretti, F. J.: Reservoir Geology of the Bachaquero-01 Reservoir, Bolivar Coastal Field, Lake Maracaibo, Venezuela, Reservoir Division EPR.81PS.81 VC.33.81, Houston, Texas (Aug. 1981). 2. E300 Reference Manual 98A, Schlumberger-GeoQuest, Houston, Texas (1998).

M.G. RODRIGUEZ, D.D. MAMORA

SPE 59335

3.

4.

5.

6.

7.

8. 9.

10.

11.

12.

13. 14.

Estudio de Fluido del Yacimiento, Pozo LL2593, Campo Bachaquero, Core Laboratories Int., S.A. Archivo: RFL2701112, Maracaibo, Venezuela (May 1985). Fernandez, B. G. and Zerpa, L. B.: Numerical Simulation of Two Horizontal Wells Drilled in a Heavy Oil Reservoir, Maracaibo Lake, Venezuela, paper SPE 37105 presented at the SPE International Conference on Horizontal Well Technology, Calgary, Canada, Nov. 1996. Burroughts, M. H., Davis, K. A., Lozano, J. A., Smith, P. G. and Wooten, J. M.: Core Analysis Report B-1 Reservoir, Well LL-2318, Reservoir Division, Reservoir Evaluation Division, EPR.123PS.82 VC.50.82, Houston, Texas (Oct. 1982). Reservoir Fluids Analysis Report for Well LL525, Wright Petroleum Laboratories, Tulsa, Oklahoma (Jan. 1967). Whitson, Curtis H.: Characterizing Hydrocarbon Plus Fractions, paper SPE 12233 presented at the 1980 European Offshore Petroleum Conference and Exhibition, London, England, Oct. 21-24. Satter, A.: Heat Loss During Flow of Steam Down a Wellbore, JPT (July 1965) 845-851; Trans., AIME, 234. Amyx, J. W., Bass, D. M. Jr., and Whiting, R. L.: Petroleum Reservoir Engineering-Physical Properties, McGraw-Hill, New York (1960). Pearce, J. C. and Megginson, E. A.: Current Status of the Duri Steamflood Project Sumatra, Indonesia, paper SPE 21527 presented at the International Therma Operations Symposium, Bakersfield, California, Feb. 7-8, 1991. Gael, B. T., Putro, E. S., Masykur, Akmal, and Lederhos, L. J.: Reservoir Management in the Duri Steamflood, paper SPE/DOE 27764 presented at the SPE/DOE Ninth Symposium on Improved Oil Recovery, Tulsa, Oklahoma, 17-20 Apr., 1994. Zarantonello, S. E., and Romero, O.: B-1 Reservoir Engineering Study, Reservoir Division EPR.106PS.81 VC.37.81, Houston, Texas (Dec. 1981). Prats, M.: Thermal Recovery, Monograph Series Volume 7, SPE, Dallas, Texas (1986). Matai, C. C., and Dalton, R. L.: Reservoir Simulation, Monograph Series Volume 13, SPE, Richardson, Texas, (1990).

TABLE 1 - Bachaquero-01 rock and fluid properties Oil gravity Average reservoir depth Original pressure at 3,000 ft Bubble point pressure Permeability Porosity Net oil sand thickness Initial oil viscosity Temperature Gas-oil ratio Oil saturation Irreducible water saturation Reservoir heat capacity API ft psia psia md % ft cp o F scf/STB % % Btu/cu.ft-oF 11.7 3,000 1,370 1,319 2,000 33.5 200 635 128 87 80 20 32.7 26.4 6010-6 7.037

Reservoir thermal conductivity Btu/D-ft-oF Rock compressibility psi-1 OOIP BSTB

TABLE 2 - Area LL125: log-derived properties Top, Thickness, ft ft 2499 2600 2640 2698 2560 2598 2635 2715 2562 2600 2625 2742 2600 2680 2720 2780 2609 2655 2695 2794 101 40 58 75 38 37 80 0 38 25 117 98 80 40 60 0 46 40 99 0 Porosity, fraction 0.22 0.34 0.36 0.16 0.35 0.36 0.31 0.00 0.27 0.35 0.34 0.28 0.32 0.33 0.31 0.00 0.32 0.34 0.27 0.00 Sw, fraction 0.52 0.20 0.12 0.65 0.17 0.12 0.25 0.00 0.41 0.15 0.21 0.37 0.31 0.22 0.32 0.00 0.27 0.18 0.41 0.00

Well

Layer

LL125 AP-1 AP-2 AP-3 HH LL2404 AP-1 AP-2 AP-3 HH LL3178 AP-1 AP-2 AP-3 HH LL2435 AP-1 AP-2 AP-3 HH LL2296 AP-1 AP-2 AP-3 HH

SPE 59335

INCREASED OIL PRODUCTION FROM BACHAQUERO-01 BY STEAMFLOODING USING HORIZONTAL WELLS

TABLE 3 - Area LL3343: log-derived properties Top, ft 2805 2835 2870 2969 2822 2870 2920 2992 2773 2832 2867 2964 2762 2810 2858 2935 Thickness, ft 30 35 99 0 48 50 72 0 59 35 97 87 48 48 77 89 Porosity, fraction 0.18 0.34 0.33 0.00 0.18 0.34 0.33 0.00 0.14 0.36 0.33 0.31 0.10 0.33 0.30 0.24 Sw, Fraction 0.57 0.17 0.26 0.00 0.64 0.19 0.23 0.00 0.64 0.13 0.24 0.30 0.75 0.24 0.39 0.47

TABLE 5 - Area LL125: simulated forecast results Final oil recovery factor, % OOIP Case 1 Case 2 Case 3 Case 4 Case 5 14.4 16.2 32.7 53.0 53.3 Incremental oil recovery, % OOIP 0.0 1.9 18.3 38.6 38.9

Well Layer LL2610 AP-1 AP-2 AP-3 HH LL2781 AP-1 AP-2 AP-3 HH LL2788 AP-1 AP-2 AP-3 HH LL2366 AP-1 AP-2 AP-3 HH

SOR, BCWE/STB --1.2 3.1 1.5 1.5

TABLE 6 - Area LL3343: simulated forecast results Final oil recovery factor, % OOIP Case 1 Case 2 Case 3 Case 4 Case 5 10.9 12.0 26.1 14.9 38.0 Incremental oil recovery, % OOIP 0.0 1.1 15.2 4.0 27.1

TABLE 4 - Area LL3487: log-derived properties Top, ft 2927 2991 3026 3102 0 2962 3026 3061 3133 3216 2935 2999 3034 3124 0 2982 3045 3090 3149 0 2900 2955 2977 3070 3177 Thickness, ft 64 36 76 85 0 64 36 72 83 76 64 36 90 69 0 63 45 59 101 0 55 22 93 107 96 Porosity, fraction 0.16 0.33 0.29 0.29 0.00 0.16 0.33 0.29 0.29 0.18 0.16 0.33 0.29 0.29 0.00 0.08 0.31 0.26 0.26 0.00 0.14 0.37 0.33 0.34 0.19 Sw, Fraction 0.65 0.28 0.41 0.41 0.00 0.65 0.28 0.41 0.41 0.63 0.65 0.28 0.41 0.41 0.00 0.78 0.38 0.53 0.53 0.00 0.63 0.10 0.26 0.25 0.57

SOR, BCWE/STB --1.9 2.2 1.5 2.1

Well Layer LL36 AP-1 AP-2 AP-3 HH GG LL112 AP-1 AP-2 AP-3 HH GG LL160 AP-1 AP-2 AP-3 HH GG LL2527 AP-1 AP-2 AP-3 HH GG LL2849 AP-1 AP-2 AP-3 HH GG

TABLE 7 - Area LL3487: simulated forecast results Final oil recovery factor, % OOIP 11.1 12.9 23.0 13.9 32.1 42.3 40.7 Incremental oil recovery, % OOIP 0.0 1.8 11.9 2.8 21.0 31.2 29.6

Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Case 7

SOR, BCWE/STB --1.8 4.8 2.0 2.4 1.9 1.6

M.G. RODRIGUEZ, D.D. MAMORA

SPE 59335

Grid blocks (x-y plane)

i 1 j 1 2 3 4 5 6 2 3 4 5 6 7 8 9 10 11

LL 125

WORLD

SOUTH AMERICA

7 8 9 10

MARACAIBO CABIMAS

11 12 13 14

LL2404 LL3178

LAGUNILLAS BACHAQUERO
125'

15 16 17

LL2435

VENEZUELA

MARACAIBO LAKE

B ACHAQ UER O-0 1 RES ERV OI R

18 19 20 21 22

LL2296

Fig. 1 - Location of Bachaquero-01 reservoir.

160'

Fig. 3 - Area LL 125: model areal grid.

1300 1200

GAMMA RAY DEPTH

RESISTIVITY
Pressure, psia

1100 1000 900

ARENA PRINCIPAL

800 700 600 500 400 300

LAGUNILLAS FORMATION BACHAQUERO MEMBER

200

measured pressure simulated pressure

HH

100 0

Jan-69

Jan-71

Jan-73

Jan-75

Jan-77

Jan-79

Jan-81

Jan-83

Jan-85

Jan-87

Jan-89

Jan-91

Jan-93

Jan-95

Time, years

GG

Fig. 4 - Area LL125: reservoir pressure history match

FF
Oil rate, STB/D 1800 Cum. oil, MMSTB 5.0 measured oil rate 1600 1400 1200 simulated oil rate measured cum. oil simulated cum. oil 4.0 3.5 3.0 4.5

EE DD CC BB AA

1000 2.5 800 2.0 600 400 200 0 Jan-69 Jan-71 Jan-73 Jan-75 Jan-77 Jan-79 Jan-81 Jan-83 Jan-85 Jan-87 Jan-89 Jan-91 Jan-93 Jan-95 Jan-97 1.5 1.0 0.5 0.0

Time, years

Fig. 2 - Type log for Bachaquero-01.

Fig. 5 - Area LL125: oil rate history match.

Jan-97

SPE 59335

INCREASED OIL PRODUCTION FROM BACHAQUERO-01 BY STEAMFLOODING USING HORIZONTAL WELLS

Grid blocks (x-y plane)

i 1 j 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 2 3 4 5 6 7 8 9 10 11

Oil rate, STB/D 3500 measured oil rate 3000 2500 2000 simulated oil rate measured cum. oil simulated cum. oil

Cum. oil, MMSTB 3.0

LL2781

2.5

2.0

1.5 1500
LL3343 LL2788

1.0 1000 500 0 Aug-80 Aug-81 Aug-82 Aug-83 Aug-84 Aug-85 Aug-86 Aug-87 Aug-88 Aug-89 Aug-90 Aug-91 Aug-92 Aug-93 Aug-94 Aug-95 Aug-96 Aug-97 0.5

0.0 Aug-98
12

LL2610

Time, years

125'

20 21 22 23 24 25 26 27

LL2366

Fig. 8 - Area LL3343: oil production rate history match.

160'

Fig. 6 - Area LL3343: model areal grid.

Grid blocks (x-y plane)

i 1 j 1 2 3 4 5 6 7 2 3 4 5 6 7 8 9 10 11

LL2849

LL 36

1400 1200

8 9 10

LL 160

1000 Pressure, psia 800

11 12 13

LL 112

600 400 measured pressure simulated pressure 0 Aug-80 Aug-81 Aug-82 Aug-83 Aug-84 Aug-85 Aug-86 Aug-87 Aug-88 Aug-89 Aug-90 Aug-91 Aug-92 Aug-93 Aug-94 Aug-95 Aug-96 Aug-97 Aug-98

125'

14 15 16 17

LL2527 LL3487

200

18 19 20

160'

Time, years

Fig. 7 - Area LL3343: reservoir pressure history match.

Fig. 9 - Area LL3487: model areal grid.

10

M.G. RODRIGUEZ, D.D. MAMORA

SPE 59335

1400 1200

Oil rate, STB/D 2500

Cum. oil, MMSTB 12.0

10.0 2000

1000 Pressure, psia

8.0

800

1500 6.0

600
1000

400 measured pressure simulated pressure

0 500

4.0

200 0 Dec-54 Dec-56

simulated oil rate simulated cum-oil

2.0

0.0 Aug-80 Aug-82 Aug-84 Aug-86 Aug-88 Aug-90 Aug-92 Aug-94 Aug-96 Aug-98 Aug-00 Aug-02 Aug-04 Aug-06 Aug-08 Aug-10 Aug-12 Aug-14 Aug-16 Aug-18 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 Dec-16 Dec-18

Dec-58

Dec-60

Dec-62

Dec-64

Dec-66

Dec-68

Dec-70

Dec-72

Dec-74

Dec-76

Dec-78

Dec-80

Dec-82

Dec-84

Dec-86

Dec-88

Dec-90

Dec-92

Dec-94

Dec-96

Dec-98

Time, years

Time, years

Fig. 10 - Area LL3487: reservoir pressure history match.

Fig. 13 - Area LL3343 (Case 3): simulated oil production forecast (steamflooding horizontal well injector, existing vertical producer).

Oil rate, STB/D 1200 measured oil rate 1000 simulated oil rate measured cum. oil 800 simulated cum. oil

Cum. Oil, MMSTB 3.0

Oil rate, STB/D 3000

Cum. oil, MMSTB 10.0

simulated oil rate

2.5
2500

simulated cum. oil

2.0

2000

600

1.5

1500

400

1.0

1000

200

0.5

500

0 Dec-54 Dec-56 Dec-58 Dec-60 Dec-62 Dec-64 Dec-66 Dec-68 Dec-70 Dec-72 Dec-74 Dec-76 Dec-78 Dec-80 Dec-82 Dec-84 Dec-86 Dec-88 Dec-90 Dec-92 Dec-94 Dec-96 Dec-98

0.0

0 Dec-54 Dec-56 Dec-58 Dec-60 Dec-62 Dec-64 Dec-66 Dec-68 Dec-70 Dec-72 Dec-74 Dec-76 Dec-78 Dec-80 Dec-82 Dec-84 Dec-86 Dec-88 Dec-90 Dec-92 Dec-94 Dec-96 Dec-98 Dec-00 Dec-02 Dec-04 Dec-06 Dec-08 Dec-10 Dec-12 Dec-14

Time, years

Time, years

Fig. 11 - Area LL3487: oil rate history match.

Fig. 14 - Area LL3487 (Case 3): simulated oil production forecast (steamflooding horizontal well injector, existing active vertical producer).

Oil rate, STB/D 1800 1600 1400

Cum. oil, MMSTB 14

12

10 1200 1000 800 600 4 400 8

simulated oil rate

200 0 Jan-69 Jan-71 Jan-73 Jan-75 Jan-77 Jan-79 Jan-81 Jan-83 Jan-85 Jan-87 Jan-89 Jan-91 Jan-93 Jan-95 Jan-97 Jan-99 Jan-01 Jan-03 Jan-05 Jan-07 Jan-09 Jan-11 Jan-13 Jan-15 Jan-17 Jan-19

simulated cum.-oil
0

Time, Years

Fig. 12 - Area LL125 (Case 3): simulated oil production forecast (steamflooding horizontal well injector, existing vertical producer)