SPE 53860 Use of Capillary Pressure Data and Log Calculated Water Saturation for the Characterization of Dual

Porosity, Dual Permeability Systems

C. Ozgen, SPE, Nitec LLC, H. Araujo, SPE, Andina S.A., and T. Firincioglu, SPE, Nitec LLC

Copyright 1999, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 1999 SPE Latin American and Caribbean Petroleum Engineering Conference held in Caracas, Venezuela, 2123 April 1999. 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.

default. The approach and the results of the numerical models are presented in the body of this paper. Introduction Existing commercial numerical simulation packages that can be employed for naturally fractured reservoirs use two media (fracture and matrix) for the storage and flow of fluids. Consequently, when the reservoir characterization is carried out for simulation purposes, the total pore volume has to be divided into two parts. The characterization of the fractures requires integration of the information from multiple disciplines. Through geophysical analysis it may be possible to determine the expected areal distribution of the density of the fractures. Visual logs can provide valuable information on the orientation, the length, the aperture, and the density of the fractures. Transient well test analysis can provide the ratio of the fracture and matrix pore volume, and through interference tests can provide the anisotropy. When available, outcrop studies can be used as analogs for the fracture network. Yet, the detailed quantification of the fracture pore volume at the scale required for numerical simulation is still very difficult. Just as in any other reservoir description parameter, the fracture porosity and its areal and vertical variation has to be accurately described for a reliable reservoir simulation study. A simple technique that addresses the characterization of fracture pore volume for the two media represented in numerical simulation models is presented in this paper. The technique relies on the knowledge of the capillary pressure relationship for the matrix and the total porosity and water saturation calculations from log analysis. The technique is most useful when the pore volumes of the two media are relatively significant. When the pore volume of one medium is very small compared to the other, the results can be misleading. In this paper we provide two example applications of the technique developed for the characterization of fracture porosity for the dual porosity, dual permeability systems. The first example is characterized as a reservoir that contains small (micro) fractures. The second example is a non-fractured reservoir that contains multiple rock types that cannot be distinguished through available techniques.

Abstract This paper describes a simple method to divide the effective porosity values that can be calculated from log analysis into two flow media (fracture and matrix) pore volumes for use in numerical simulation models. The division of the porosity is based on the assumption that the total porosity of the system is made up from two contributing rock types or lithofacies that have distinct pore size and/or pore size distribution such that they can be represented by different capillary pressure functions. This assumption is applicable to the non-fractured as well as to the fractured systems. When the method is used for the fractured systems, the fracture is typically represented by a zero capillary pressure value (gravity segregation). The matrix has its own laboratory measured capillary pressure curve. Based on a predetermined water-hydrocarbon contact and using the difference between the density gradients of the fluids, the capillary pressures of the two media are converted to theoretical water saturation profiles as a function of TVD. Next, at each depth point, the two extreme theoretical water saturation values are compared to the log calculated water saturation measurements. Based on a simple weighting algorithm, the contributions of the fracture and matrix media to the total porosity are calculated. The results can be upscaled to the numerical simulation grid block size or be used in geostatistical approaches. Using the developed technique, we have built simulation models for two different reservoirs. One of the models represents a fractured system, while the other represents a non-fractured system with two distinct rock types. For the non-fractured reservoir, the dual porosity formulation enabled us to accurately upscale the capillary pressure functions by

C. OZGEN, H. ARAUJO, T. FIRINCIOGLU

SPE 53860

Fracture Characterization Search of the petroleum literature yields numerous technical publications on the representation of fractures in the numerical simulators. Publications1-5 can be found on the characterization of fractures with various methods as well. Unfortunately, there are very few publications that cover the subject from an integrated perspective, resulting in a characterization that could be used for numerical simulation, specifically, history matching and prediction of the future performance. Agarwal and Allen6 present such an integrated application where informations from different disciplines were correlated successfully, verified by historical performance, and presumed to be used for prediction of future performace. Thomas, et.al.7 presented another integrated study, where the characterization of the fractures could not explain the observations from the field. With the better understanding of the integration of the data from different disciplines, we expect more significant publications in the near future. The ultimate objective of fracture characterization is to quantify its benefits and/or harms. The most sophisticated method of quantification is the use of numerical simulators. Thus, the characterization should be oriented towards the identification of the reservoir description parameters that affect the performance of the reservoir simulator. From a simplistic point of view, the parameters common to various commercially available dual porosity simulators are; porosity and permeability of the fractures, matrix-fracture exchange term, and matrix porosity. Among these parameters, matrix porosity is typically the best known quantity. Matrix-fracture exchange term is a function of matrix permeability and fracture spacing. Fracture spacing has been the subject of many studies, and some excellent techniques have been developed for its characterization6,7. Fracture permeability can only be characterized through transient well test analysis, but the scale of the results are much different than the scale required for the numerical simulator, thus its use is limited at best. On the other hand, interference tests are critcal in the identification and quantification of anisotropy due to the communication of the fractures. Fracture porosity is the most oversimplified parameter, and it has not received the same level of attention as the previous three. In general, fracture porosity is assumed to be a constant value (typically 0.1% of the bulk volume) and is neglected. In reality, depending on how the fracture is defined in the numerical simulator (i.e. zero capillary pressure, gravity segregation) the fracture pore volume can be significant. For example, vugs that contact the fracture network can be significant volume and should be included in the fracture porosity. The density of the fractures (and the variation thereof) has significant effect on how the fracture porosity is distributed in the numerical simulation grid blocks. Afterall, the fracture porosity in the simulator represents the total fracture pore volume divided by the bulk volume of the grid block. Where

the fractures are denser, the grid block fracture porosity should be higher. In numerical simulators, the variation of fracture porosity has one more profound effect. Since the transmissibility among the dual porosity grid blocks is proportional to effective permeability (which is the product of fracture permeability and fracture porosity), the fracture porosity also controls the communication among these blocks. Thus, the accurate characterization of fracture pore volume reduces the adoptation of unreasonable values for the fracture permeability to obtain a history match. In this paper, we propose a technique that will enable us to characterize the fracture pore volume at a scale that is the same as matrix porosity. Thus, the results of the characterization can be upscaled using the similar techniques that are used for the matrix porosity. Data Requirements for the Proposed Technique The data required for our technique can be obtained from core analysis and log analysis. The core analysis data needed are the capillary pressure curves of the matrix. We recommend the analysis of all the core data for the pore throat size distribution parameter (), the threshold (entry) pressure, and their relationship to the core porosity. The basis of this characterization can be found in earlier publications8-10, and the concept is probably best summarized by Standing11. From log analysis it is possible to obtain total porosity and water saturation as a function of true vertical depth. Other parameters determined from log analysis, such as shale volume can also be useful to filter out the non-reservoir data. Other pertinent information for our technique include the density gradient of the reservoir fluids and the depth of the free water table. The density gradient of the fluids can be obtained from PVT analysis, but if they are not available, they can be reliably estimated. Free water table is defined as the depth that the capillary pressure of the water and the hydrocarbon phase is equal to zero. For single porosity reservoirs, the estimation of free water table requires the knowledge of threshold pressure, the density gradient of the fluids, and the depth of 100% water saturation level in the reservoir. For fractured reservoirs, since we assume that the capillary pressure in the fracture is negligible, the free water table is defined as the lowest known depth of hydrocarbon saturation. Characterization of Fracture Pore Volume Our technique relies on the comparison of the log analysis results (Sw) to the expected water saturation from the capillary pressures (Swf and Swm). We will start with a very simple equation that states that the average water saturation is a combination of fracture and matrix values.
Sw * = Swf * f + Swm * m

Dividing both sides by the system porosity, and after some simplification we get;
Sw = Swf * Ff + Swm * (1 Ff )

SPE 53860 USE OF CAPILLARY PRESSURE DATA AND LOG CALCULATED WATER SATURATION FOR THE CHARACTERIZATION OF DUAL POROSITY, DUAL PERMEABILITY SYSTEMS 3

Ff is the fraction of the total pore volume that corresponds to fracture. We can further simplify the equation so that we can solve for Ff.
Ff = (S wm Sw ) /(Swm S wf )

An example is depicted in Figure 1. In this figure, we can see the depth trace of log analysis calculated water saturation values that represent the average for the whole system (Sw). For practical purposes, this trace should be filtered for a porosity range. On the same figure we placed two more traces, identified as fracture (Swf) and matrix water saturation (Swm). The matrix water saturation values at any depth can be calculated from the capillary pressure data, the density gradient of the fluids, and the free water level. Considering a particular depth, we can identify three points on Figure 1. Point A represents the water saturation of the fracture, point B is the water saturation value determined from the well log analysis, and point C is the water saturation that represents the matrix. For simplicity purposes, if we assume that the following data represents points A, B and C, the Ff can be calculated as; Point A (Swf) = 0% Point B (Sw) = 50% Point C (Swm) = 75% Ff = (0.75 0.50) / (0.75 0.0) = 0.33 Thus, for this sample well, at that particular depth, one third of the pore volume can be attributed to the fracture medium, and the rest to the matrix. Our approach provides results that are in the same scale as porosity and water saturation that are obtained from well log analysis. Thus, we can expect the upscaling and interpolation techniques that apply to porosity can also be applied to Ff. It should be noted that Ff is a function the average porosity of the fractures and the density of the fractures within any control volume of the reservoir. Sara Member of the El Carmen Formation, Vibora Field The Vibora Field is located in the Boomerang Hills region of Bolivia (Figure 2). It contains several producing horizons with one being identified as the Sara Reservoir. Sara is a clean sandstone reservoir with practically no shales and no differentiation in rock types. The reservoir has an original gas cap and a weak aquifer. A few wells have visual logs, and their interpretation indicate presence of small fractures with apertures up to 100 microns. The technically interesting information for the Sara Reservoir comes from the results of the well log analysis. In most cases the wells penetrate the whole formation, and it is possible to pick the water-oil-contact. Unfortunately, when the water-oil-contact is mapped, we obtain an uneven distribution. Various reasons for the variation of the water-

oil-contact were investigated, but all possible explanations were disqualified. Some of the explanations included the possibility of having an active aquifer, the presence of sealing faults and/or compartmentalization, and the uplifting of some compartments to name a few. Since the reservoir sandstone is very uniform, having multiple lithofacies was not an explanation either. For two of the wells in this reservoir, Figure 3 shows the results of the water saturation calculations obtained from log analysis for porosity values that are greater than 9%. As can be seen in this figure, well VBR-20 has deeper water-oilcontact when compared to well VBR-10. Normally, since VBR-20 was drilled a few years later, considering aquifer influx, one would expect it to have shallower contact. If there were no aquifer influx, we would expect it to be as deep as VBR-10. Figure 4 shows similar results for wells VBR-14 and VBR-17D, where both are TVD corrected. Here we can also see that from stratigraphy point of view, the water saturation values are significantly lower at the highest strata. Most of the wells that were studied show this behavior. One possible explanation for the entire behavior that is described above is the possibility of variation of fracture density stratigraphically, vertically, and/or areally. It was postulated that if the fracture density increased at higher strata, since the fractures naturally have lower water saturation, it could result in lower water saturation calculations in the log analysis. Similarly, if the fracture density is higher at the VBR-20 well, it could explain why we could observe a deeper water-oil-contact, when compared to VBR-10 well. To quantify the fracture density, and to be able to analyze its variation in all directions we developed the technique outlined in this paper. Since there were over twenty wells that had to be processed, a Fortran program was developed. The program reads the well log results from flat ASCII files and other pertinent information (such as Pc curves) from a control file. Each data point in the log results file contains TVD, porosity, Vshale, and water saturation. Based on the porosity of the particular point, a capillary pressure curve is selected and converted to depth equivalent. Next, using the formula presented above, the fraction of the pore volume that corresponds to fracture is calculated. In order to analyze the results for the Sara Reservoir, we upscaled the Ff values to six stratigraphic layers, and represented it as the variation of rock type. This means, where we have high fracture volume fraction (Ff), there is more of rock type 1, and where we have low fracture volume fraction (Ff), there is more of rock type 2. Here, it is important to note that because we assigned a non-zero irreducible water saturation to the fracture, the results are more biased towards rock type 1. This bias does not affect the variation in the rock type assignment, which is the point of interest. When the rock type assignments are plotted for all the wells stratigraphically, we observe that more of fracture can be found at higher strata. This can be seen in Figure 5. The result is consistent with our

C. OZGEN, H. ARAUJO, T. FIRINCIOGLU

SPE 53860

earlier observation, and explains the reason behind observing lower average water saturation at the top strata. Figure 6 shows similar information for eight structurally defined layers, where layer 1 is the shallowest, and layer 8 is the deepest section. Here, we see that the scatter of data reduces with increasing depth, but the relationship is not clear. We believe that the reason behind the lack of good relationships is due to the fault planes contained within the structure. Figure 7 shows the structure of the top of Sara reservoir with major faults defined in thicker lines. At of the time of this publication, the processing of the 3D seismic data had not been completed, thus the map presented here relies on well picks, 2D seismic interpretation and 3D geologic modeling of fault planes. To analyze the areal distribution of the Ff value per strata, we converted the information to equivalent fracture porosity values (f), and used standard contouring of available commercial packages. Our purpose was to visualize relationships rather than to quantify, thus, other alternatives such as krigging were not considered. Figures 8 and 9 show the contoured areal distribution of f for two of the twelve layers. These figures show that the fractures are more frequent closer to the major faults. This conclusion also supports the explanation behind the uneven water-oil-contact. The consistency of the results of the new technique with our other expectations regarding 3D distribution of the fault pore volume has encouraged us to develop a detailed 3D reservoir simulation model. The initialization of the dualporosity model yielded the uneven water-oil-contact that was sought, and the history matching exercise required minor modifications to the fracture characterization, mainly at the areas where the well control is inadequate. We expect the improved fault and fracture characterization through the newly acquired 3D seismic data to further improve the results. Yantata Formation, Vibora Field The Yantata Formation is several hundred meters shallower than the Sara Reservoir, and to our knowledge, it is not fractured. Yet, this formation presents another interesting problem that is also common to many other reservoirs throughout the world. The Yantata Formation is divided into three members that are proven to be in communication. The Lower Member of the formation is clean eolian sand with good porosity and permeability. The Middle Member has shallow marine sands with good porosity, but very low permeability and relatively higher water saturation. The Upper Member is a mixture of sands from the two environments, and reflects the properties of both. From top to bottom, the formation contains gas condensate, oil (in the low permeability rock), then gas condensate, and finally a small oil rim. The existing suites of well logs are not adequate to differentiate the two sands, except for the resulting water saturation calculations. As mentioned earlier, the calculated porosity of the two sands are very similar. While it is possible

to use a water saturation filter to eliminate the low permeability sands, these sands contain significant amount of hydrocarbons. Even if the recovery factor of the hydrocarbons from the low permeability sands is very small, through expansion it can provide additional energy for the rest of the reservoir. The quantification of the low permeability sands was accomplished using the same method that was developed for the characterization of the fracture pore volume. One handicap in this application was the lack of core samples from the low permeability rock for the assessment of the capillary pressure. Thus, the capillary pressure curves of the low permeability rock were obtained through the analysis of the log analysis results. Figure 10 shows the plot of water saturation versus depth for the rock within the porosity range of 14% to 17%, for all the wells that penetrate the formation. Using this figure, for the two distinct rock types most appropriate capillary pressure functions were developed. Same exercise was carried through for the remaining porosity bins. The results were upscaled to thicker layers that could be used for analysis. As expected, the Middle Yantata Member showed the most amount of low permeability rock. For each layer, we contoured the percent of low permeability rock. Figure 11 shows our results for one of these layers. Here, we can see that the low permeability, shallow marine rock thins out towards the south of the reservoir. This observation is consistent with where seashore is expected to be at the time of deposition. As the consequence of the characterization of the two rock types, we were able to mimic a non-fractured, dual permeability system by constructing a dual-porosity full filed model. For the simulation model, we specified the high permeability rock as the primary flow medium (fracture), and the low porosity rock as the secondary medium (matrix). In this particular application, regular capillary pressure and relative permeability curves were assigned to the fracture medium, rather than the straight lines that assume gravity segregation. Since porosity is an intensive parameter, for the porosity-permeability transformations of both media we used total porosity (matrix plus fracture). The initialization of the dual-porosity simulation model with distinct capillary pressure curves for the matrix and the fracture resulted in obtaining high water saturation and liquid hyrdrocarbons in the middle strata, while the upper and lower strata contained mostly gas condensate. A successful full field history match was achieved through the further refinement of the method of allocation of the pore volumes to the fracture and matrix media. Here we provide results from two wells that show different GOR profiles, even though both wells are completed at the same vertical depth. Figure 12 shows the behavior of well VBR-8, which is completed in a region with mostly high permeability sands (fracture). As can be seen in this figure, the GOR of the well reflects the condenssate yield of the gas. Figure 13 shows another well (VBR-22A) that is completed in a region with mostly low permeability sands (matrix). As can be seen in

SPE 53860 USE OF CAPILLARY PRESSURE DATA AND LOG CALCULATED WATER SATURATION FOR THE CHARACTERIZATION OF DUAL POROSITY, DUAL PERMEABILITY SYSTEMS 5

this figure, the GOR of this well is much lower, indicating production of liquid hydrocarbons from the matrix, as well as the condensate recovered from the fracture. The lower GOR behavior is observed despite the fact that the reservoir pressure around VBR-22A is lower than the VBR-8. To date, the most complicated issue has been the tuning of the fracture-matrix exchange terms through the modification of the permeability of the matrix. The petroleum literaure does not include any similar applications where a dualporosity simulator was used for a single porosity, dual rock type reservoir, and does not address the proper method for the upscaling of matrix properties in such systems. Additional work on detailed single well, single porosity simulation is ongoing. We expect the results of this work to verify or update the upscaling techniques we have used in the dual porosity representation of this dual rock type reservoir. Conclusions 1. A simple technique to evaluate the pore volume of the fracture media was developed. 2. The technique provides results at the same scale as other well log interpreted parameters, such as porosity and water saturation, thus it lends itself to the same upscaling procedure and interpolation techniques used for the generation of property arrays in reservoir simulation. 3. When the results of the fracture pore volume characterization are analyzed, they are consistent with our understanding of the behavior of fractured systems. 4. The technique can be used for dual permeability systems as well as dual porosity. Nomenclature Ff = pore volume fraction (of fracture) Sw = water saturation = porosity Subscripts m = matrix f = fracture Acknowledgments The authors would like to thank Andina S.A. for the permission to publish this paper. We also thank Chuck Weinstein, Sheila Smothers, and Melissa Moore for their help in the preperation of figures. References
1. Shanks, R.T., Kwon, B.S., DeVries, M.R., and Wichmann, P.A.: A Review of Fracture Detection with Well Logs, paper SPE ????, presented at the 1976 SPE Annual Technical Conference and Exhibition, New Orleans, October 3-6. Beliveau, D.: Pressure Transients Characterize Fractured Midale Unit, paper SPE 15635, JPT (December 1989) 1354.

3.

Rodriguez, A.: A Comprehensive Study of Cementation and Fracture Porosity Relationships for Reservoir Characterization Pembina and Garrington Fields, Canada Examples, paper SPE 21089, presented at the 1990 SPE Latin American Petroleum Engineering Conference, Rio de Jeneiro, October 14-19. 4. Kubik, W., and Lowry, P.: Fracture Identification and Characterization Using Cores, FMS, CAST, and Borehole Camera: Devonian Shale, Pike County, Kentucy, paper SPE 25897, presented at the 1993 SPE Rocky Mountain Regional, Low Permeability Reservoirs Symposium, Denver, April 1214. 5. Lynn, H.B., Bates, R., Layman, M., and Jones, M.: Natural Fracture Characterization Using P-Wave Reflection Seismic Data, VSP, Borehole Imaging Logs, and the In-Situ Stress Field Determination, paper SPE 29595, presented at the 1995 SPE Rocky Mountain Regional, Low Permeability Reservoirs Symposium, Denver, March 20-22. 6. Agarwal, B., and Allen, L.R.: Ekofisk Field Reservoir Characterization: Mapping Permeability Through Facies and Fracture Intensity, paper SPE 35527, presented at the 1996 SPE European 3-D Reservoir Modelling Conference, Stavanger, April 16-17. 7. Thomas, M.B., Nutakki, R., Baker, R.O., Geppert, E.C.J., and Gould, B.: A New Interpretation of Frcature Distribution in Waterton Sheet III: An Integrated Reservoir Characterization Study, paper SPE 35605, presented at the 1996 SPE Gas Technology Conference, Calgary, April 28May 1. 8. Burdine, N.T.: Relative Permeability Calculations From Pore Size Distribution Data, Trans., AIME (1953) Vol. 198, 71-78. 9. Land, C.S.: Calculation of Imbibition Relative Permeability for Two- and Three-Phase Flow from Rock Properties, paper SPE 1942, presented at the 1967 SPE Annual Fall Meeting, Houston, October 1-4. 10. Land, C.S.: Caomparison of Calculated with Experimental Imbibition Relative Permeability, paper SPE 3360, presented at 1971 SPE Rocky Mountain Regional Meeting, Billings, June 2-4. 11. Standing, M.B.: Notes on Relative Permeability Relationships, 1975, Stanford University.

2.

C. OZGEN, H. ARAUJO, T. FIRINCIOGLU

SPE 53860

De pth vs S w (1 4 % < P hi < 1 6 %) V B R-2 2 A

34 00

34 50

Sw
35 00

De pth

35 50

Swm Swf
P o in t A
36 50

36 00

P o in t B

P o in t C

37 00 0 0.2 0.4
Sw

0.6

0.8

Fig. 2-Map of Bolivia and adjacent areas showing principal geological features and location of the Boomerang Hills. (From Welsink and others, 1995.)

Fig.1-Water saturation vs. depth example

D e p t h v s S w ( 9 % < P h i) V BR-2 0

D e p t h v s S w ( 9 % < P h i) V BR-1 0

3400

3400

3450

3450

3500

3500

De p th

3550

De p th 0 0 .2 0 .4 Sw 0 .6 0 .8 1

3550

3600

3600

3650

3650

3700

3700 0 0 .2 0 .4 Sw 0 .6 0 .8 1

Fig. 3-Sara Reservoir, water saturation from log analysis for wells VBR-10 and VBR-20, with water saturation calculations from a capillary pressure curve as reference.

SPE 53860 USE OF CAPILLARY PRESSURE DATA AND LOG CALCULATED WATER SATURATION FOR THE CHARACTERIZATION OF DUAL POROSITY, DUAL PERMEABILITY SYSTEMS 7

D e p t h v s S w ( 9 % < P h i) V BR-1 4

D e p t h v s S w ( 9 % < P h i) V BR-1 7 D

340 0

340 0

345 0

345 0

350 0

350 0

De p t h

355 0

De p t h 0 0 .2 0 .4 Sw 0 .6 0 .8 1

355 0

360 0

360 0

365 0

365 0

370 0

370 0 0 0 .2 0 .4 Sw 0 .6 0 .8 1

Fig 4-Sara Reservoir, water saturation calculations from log analysis for wells VBR-14 and VBR-17D.

Stratigraphic Variation of Rock Type

Structural Variation of Ff 1.0

1 .9

0.9

1 .8

0.8

1 .7

0.7

1 .6

0.6 Ff (fraction)

Rock Type

1 .5

0.5

1 .4

0.4

1 .3

0.3

0.2
1 .2

0.1
1 .1

0.0
1 1 2 3 4 5 6

Strata

Structural Layer Number

Fig. 5-Rock type variation as a function of strata, Sara Reservoir, all wells.

Fig. 6-Ff variation with depth, Sara Reservoir, all wells.

C. OZGEN, H. ARAUJO, T. FIRINCIOGLU

SPE 53860

Fig. 7-Sara Reservoir structure top contour

Fig 8-Sara Reservoir, rock type 1 porosity for layer number 1

Fig. 9-Sara Reservoir, rock type 1 porosity for layer number 3

SPE 53860 USE OF CAPILLARY PRESSURE DATA AND LOG CALCULATED WATER SATURATION FOR THE CHARACTERIZATION OF DUAL POROSITY, DUAL PERMEABILITY SYSTEMS 9

Depth vs Sw (17% > Phi > 14%)

2550

2600

2650

2700 Depth 2750 2800 2850 2900 0 0.2 0.4 Sw 0.6 0.8 1

Fig. 10-Yantata Reservoir, water saturation from log analysis, for all the wells, porosity bin 14%-17%, and two identifiable capillary pressure trends.

Fig. 11-Yantata Reservoir, low permeability rock porosity distribution for layer 6.

10

C. OZGEN, H. ARAUJO, T. FIRINCIOGLU

SPE 53860

Fig. 12-History match plot of VBR-8L, Yantata Reservoir.

Fig. 13-History match plot of VBR-22A, Yantata Reservoir.