SPE 103284 Continuous Fracture Modeling of a Carbonate Reservoir in West Siberia O. Pinous/Schlumberger, Abdel M Zellou, Gary Robinson, Ted Royer, Prism Seismic; N. Svikhnushin/ Schlumberger, D. Borisenok/Schlumberger, A. Blank/Schlumberger

Copyright 2007, Society of Petroleum Engineers This paper was prepared for presentation at the 2007 International Oil Conference and Exhibition in Mexico held in Veracruz, Mexico, 27–30 June 2007. 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, Texas 75083-3836 U.S.A., fax 01-972-952-9435.

Abstract The field is located in the southeastern part of the West Siberian basin in Novosibirsk oblast (Fig. 1). It was the first field in the basin where commercial oil was produced from the Paleozoic basement. The reservoir consists mostly of limestones and dolomites that are intensively fractured and contain numerous vugs in some zones. The reservoir properties of the matrix are generally negligible, and the production potential of wells is mostly associated with natural fractures and vugs. The presented study was our first project in Russia where a complete integrated approach was implemented to properly characterize a fractured reservoir. The approach included the following tasks: 1) Identification of fractured intervals in wells using a special technique of BKZ logs processing, 2) Spectral imaging and high-resolution inversion of the seismic data, 3) structural analysis of the field, 4) construction of the reservoir properties model, 5) construction of the fracture distribution model using the Continuous Fracture Modeling approach (CFM). A comprehensive description is available on a previous publication1. The final geologic model served as a basis to select the locations for the new wells. The new locations were proposed in the zones with the most intensive development of a network of natural fractures (according to the model). The drilling was associated with significant losses of drilling mud that was an indirect indication of presence of significantly fractured zones. The wellbore image FMS that was recorded in the well, showed a good level of correspondence between the model forecast and the actual result. The well contains interval of numerous fractures and large vugs. Eventually, the well showed a good production results and currently is one of the best producers in the field.

As such, we recommend application of the described integrated approach for modeling complex fractured reservoirs in the other fields of Russian Federation. Introduction The field was discovered back in 1974 by the exploration well 2, which was drilled in the southern part of the anticline that was delineated by seismic data. Commercial flow of oil was produced from the carbonate reservoirs of the “M” horizon that represents the uppermost portion of the Paleozoic basement2. The discovery has attracted a significant attention at the time, being a first demonstration of the productive potential of West Siberian basement3. In the next few years a series of medium and small size oilfields with pre-Jurassic reservoirs have been found in the southeastern part of the basin (e.g. Archinskoe, Chkalovskoe, Urmanskoe, Gerasimovskoe, and others). In all of these fields oil was produced from the basement carbonates and weathering crust. Further investigation on Pre-Jurassic reservoir of the SE West Siberia showed that production potential is mostly related to the basement limestones that have been significantly affected with secondary processes such as dolomitization, leaching, and fracturing4. Following the initial discovery, 19 wells have been drilled in the field, and 8 of them produced commercial oil rates. The results of core investigations and well test analyses showed that the productive unit “M” consists of a complex fracturedvuggy-porous type of a reservoir. A presence of opened fractures was determined as a key factor that defines productive potential of wells. General information The basement of the field consists mostly of Paleozoic carbonates that also include some layers of siliciclastic and volcanic rocks. The overall structure of the field represents a elongated anticline of an irregular shape that is located northwest of Mezhov arch. Interpretation of 3D seismic data showed that the basement strata contain numerous nearly vertical faults (Fig. 2). The faults are rarely traceable above the top of Jurassic Tyumen formation. The reservoir quality rocks are concentrated in the upper part of the basement that is capped by the Jurassic shales or tight rocks of the weathering crust. The reservoir lithology includes limestones and dolomites that originally consist mostly of organic grains including forams, coral debris,

2

algaes, etc. All rocks have undergone a significant recrystallization and the original porosity is mostly deteriorated as a result. Visual descriptions of core showed a presence of vuggy intervals, where dimensions of individual vugs ranges from 2-3 mm to 2 cm. It also indicated that all rocks are naturally-fractured. Most fractures are closed (i.e. filled with calcite or shale minerals), but a few opened fractures with apertures up to 2 mm are reported from core descriptions as well. The common problem for characterization of vuggy and fractured reservoirs is that the application of standard log interpretation methods (commonly applied in Russian industry for porous reservoir) does not make it possible to identify reservoir intervals in wells and to perform their quantitative estimations. Population of the secondary properties in the interwell space during construction of geological model appears to be even a more complicated issue. An application of an integrated interdisciplinary approach and new techniques, presented in this study, has enabled to resolve the above issues and build a high quality geological model for the field. Geological modeling To build the geological model in scope of the multidisciplinary approach the following tasks have been completed: 1) tectonic analysis (evaluation of structural style, Fig. 3) 2) log interpretation using special technique of BKZ processing 3) high resolution inversion and spectral imaging of seismic data 4) modeling of matrix properties 5) fracture modeling These steps can be summarized in an integrated workflow displayed on Fig. 4. The focus of this paper is on the fracture modeling part. A comprehensive description of the first four steps is available on a previous publication1. Fracture modeling Methodology To model the distribution of natural fractures, the Continuous Fracture Modeling5-13 (CFM) method was used. This framework was developed with the objectives of working on a geocellular grid by using any type of fracture indicator available at the wells and providing a framework where any geologic or geomechanical indicator, or any field measurement (seismic), can be incorporated quantitatively in the modeling effort. The glue of the method is a set of artificial intelligence tools that give the flexibility needed to deal with complex fractured reservoirs. It is widely known that structure, lithology, bed thickness, porosity, faults and other geologic factors control the intensity of fractures in geological media. These factors controlling natural fracture distribution in interwell space are termed as fracture drivers. On the other hand, direct information on fractures that comes from well data interpretation is called “fracture indicators”. If a functional relationship between fracture indicators and drivers exists, it can be used to populate fractures between wells. This approach is the core of

[Paper Number]

the CFM approach. The methodology consists of three steps that are given below. The first step is the ranking of all existing geologic drivers. A fuzzy neural network is used to evaluate the hierarchical effect of each geologic driver on the fractures (Fig. 5). As a result, the geologist or reservoir engineer will be able to identify the key geologic drivers affecting fractures. The second step is to create a set of stochastic models using a neural network, which attempts to quantify the underlying complex relationship that may exist between key geologic drivers and fracture intensity. The training and testing of the neural network is accomplished using existing data. The third step is to perform uncertainty analysis by examining the fracture cumulative distribution function resulting from the large number of stochastic models. Using these three steps, the software will be able to predict the 3D distribution fractures and their underlying uncertainty at undrilled locations. Model construction To construct the model of fracture distribution the following data set was used: 1) Fractured intervals in wells from log interpretation, 2) Structural model (horizons and faults), 3) models of porosity and lithological content, and 4) seismic attributes. The fractured intervals in wells came from special BKZ processing and represent continuous “flag” curves without quantification of fracture intensity, azimuth, etc. The intervals were lumped to the model cells and the resulting curves were treated as a relative fracture index (CRI). The CRI curves served as fracture indicators, whereas the main fracture drivers included curvature in 4 directions, deformation volume, distance to faults, porosity, dolomite content, and others. Several iterations were performed to select the most relevant drivers during the ranking process. Out of the thirty nine (39) initial drivers, twenty seven (27) are being used in the final ranking. Once the neural network finds the relationship between the drivers and the fracture intensity, it was used to predict fracture intensity in every cell of the model (Fig. 6). This approach uses a stochastic framework, where many models are generated by using various sets of training and testing data. 50 stochastic realizations were calculated and three out them were selected as the base-, downside and upside cases respectively. Once the fracture distribution is obtained, a connectivity analysis is performed revealing the areas of high potential (Fig. 7) Analysis of the results The resulting model of fracture intensity was used to recommend a location of the new well. Despite getting the new geological information, the new well was aimed to encounter good quality reservoirs and become a good producer. The well was planned in the southern part of the field and became the first one after almost twenty years of non-drilling period. Catastrophic mud losses that started to occur after a penetration of the basement top suggesting that zones of intense fracturing or leaching have been encountered. After

[Paper Number]

the drilling was completed about 90% of core was acquired and logging with FMS microimaging tool completed. The results of core analysis and FMS interpretation showed a presence of two systems of fractures with different orientation (the first one with low angle and the second nearly vertical). While the first one is represented mostly by closed fractures filled with shale minerals, the second one contains numerous opened fractures. The FMS also showed a large cavern (about 1 m) in the middle part of the logged interval that is mostly filled with clay (shale) material. Comparison of the fracture interpretation from the FMS with the model forecast showed their good correspondence. In fact, the intervals with high, medium and low fracture intensity are present approximately at the same levels (Fig. 8). Finally, the production results proved good reservoir properties for the well as it has become one of the best producers in the field. Thus, the application of the complex multidisciplinary approach to characterize and model fractured reservoirs has led to a good confirmation of the model forecast by drilling results. This demonstrates its practical utility for application in similar fields. Conclusions Based on the extensive collaborative work done by a large multidisicplinary team from September 2004 to May 2005 on the field the following conclusions can be drawn: 1. The reality of a well proven multidisciplinary workflow to characterize complex fractured reservoirs is demonstrated and confirmed by drilling results. 2. In absence of direct fracture count from image logs, the BKZ log provides reliable fracture count information that needs to be calibrated. 3. The derived seismic attributes are used quantitatively in subsequent geologic and fracture modeling steps and ensure their robustness to create meaningful models. 4. The efficiency in the geological modeling package in rapidly generating various geologic models constrained by different soft information is demonstrated. 5. The flexibility of the fracture modeling package in generating fracture models using all the geophysical, geologic and production data is demonstrated.

3

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

Acknowledgements The authors would like to thank Mr. Sokolov, Mr. Bahir and the management of Russneft for allowing the presentation of this case study.

12.

References 1.

Pinous, O., Sokolov, E.P., Bahir, S.Y., Zellou, A.M., Robinson, G., Royer, T., Svikhnushin, N., Borisenok, D., Blank, A: “ Application of an Integrated Approach for the Characterization of a Naturally Fractured Reservoir in the West Siberian Basement (Example of Maloichskoe Field)” paper SPE 102562 presented at

13.

the 2006 SPE Russian Oil and Gas Technical Conference and Exhibition, Moscow, Russia, Oct.2-6. Zapivalov, N.P. Abrosimova, O.O., Popov, V.V., Geological and geophysical model of Maloichskoe field in Palezoic intervals of the West Siberian basement and features of its development. Geologiya nefti I gaza, no. 2, 1997. (in russian) Zapivalov, N.P. Beneath the lower edge of the Mesozoic. Sovetskaya Sibir 18.06.2004, (www.sovsibir.ru) Kontorovich, V.A., Berdnikova, S.A., Antipenko, S.V. Geological structure and hydrocarbon prospects of the contact zone between paelozoic and mesozoic strata of the southern part of Vasyugan productive region. Geologiya nefti I gaza, no. 2, 2004. (in russian) Ouenes, A., Robinson, G., Zellou, A:”Impact of PreStack and Post-Stack Seismic on Integrated Naturally Fractured Reservoirs,” paper SPE 87007 presented at the 2004 SPE Asia Pacific Conference on Integrated Modelling for Asset Management, KL Ouenes, A. Zellou, A, Robinson, G, Balogh, D, Araktingi U. “Improved Reservoir Simulation with Seismically Derived Fracture Models,” paper SPE 90822 presented at the 2004 SPE Annual Technical Conference and Exhibition, Houston. Christensen, S., Ebbe Dalgaard, T., Rosendal, A., Christensen, W., Robinson, G., Zellou, A.M., Royer, T.:“Seismically Driven Reservoir Characterization Using an Innovative Integrated Approach: Syd Arne Field,“ paper SPE 103282 presented at the 2006 SPE Annual Technical Conference and Exhibition, San Antonio. Zellou, A, Hartley, L.J, Hoogerduijn-Strating, E.H, Al Dhahab, S.H.H, Boom, W., Hadrami, F.: “Integrated Workflow Applied to the Characterization of a Carbonate Fractured Reservoir: Qarn Alam Field,” paper SPE 81579 presented at 2003 Middle East Oil Show & Conference, Bahrain. Zellou, A., Ouenes, A: “Integrated Fractured Reservoir Characterization Using Neural Networks and Fuzzy Logic: Three case studies “,Soft Computing and Intelligent Data Analysis in Oil Exploration, Elsevier, Amsterdam Laribi, M., Boubaker, H, Beck, B., Chen, H.K, AmiriGarroussi, K, Rassas, S, Rourou, A, Boufares, T, Douik, H., Saidi, N., and Ouenes, A.: “Integrated Fractured Reservoir Characterization and Simulation: Application to Sidi El Kilani Field, Tunisia.”, paper SPE 84455 presented at the 2003 SPE Annual Technical Conference and Exhibition, Denver. Gauthier, B., Zellou, A., Toublanc, A., Garcia, and J.M Daniel:”Integrated Fractured Reservoir Characterization: a Case Study in a North Africa Field,” paper SPE 65118 presented at the 2000 European Petroleum Conference, Paris, Oct.24-25 Boerner, S., Gray, D., Todorovic-Marinic, D., Zellou, A., Schnerk, G: “Employing Neural Networks to Integrate Seismic and Other Data for the Prediction of Fracture Intensity,” paper SPE 84453 at the 2003 SPE Annual Technical Conference and Exhibition, Denver Kouider El Ouahed, A., Tiab, D., Mazouzi, A., Jokhio, S.: “Application of Artificial Intelligence to Characterize Naturally Fractured Reservoirs”, paper SPE 84870 presented at the 2003 Asia Pacific SPE International Improved Oil Recovery Conference, KL, Malayisa, 20-21 October 2003.

4

[Paper Number]

Regional shear direction Compression direction

Fold axis

Figure 3: Interpretation of the tectonic mechanism of the structure development (on the time map of the basement top)

Figure 1: Location of the field in West Siberia

Figure 2: Top basement depth map

Figure 4: Integrated workflow for the characterization of the field.

Porosity

Скв.

Curvature N-S

Скв.

Dol. content

5

Distance to fault

[Paper Number]

Скв.

Attenuation frequency

Seismic impedance

Скв.

Figure 5: Comparison of the fracture indicators (in wells) with various fracture drivers and seismic attributes.

Figure 7: Visualization of the connectivity analysis from the CFM model displayed on the previous figure.

High fracturing

Low fracturing

Middle fracturing

Forecast Figure 6: Continuous Fracture Model (CFM). Fracture distribution based on CRI

FMS

Figure 8: Comparison of the model forecast with the actual drilling results (fracture interpretation is based on the FMS data)