No abstract

Summary We used an integrated solution by combining "direct" and "inverse" approaches to fracture network characterization in a stochastic numerical model. Static geological data obtained from cores and well logs were used together with dynamic data such as well-test responses to build 3D discrete fracture-network (DFN) models. We used the data obtained from the fractured carbonate Midale field in Canada. The fractured-reservoir model was constructed from static and dynamic (drawdown and pulse-interference tests) data. Matrix and several fracture parameters including fracture length, density/ spacing, aperture, connectivity, and orientation were evaluated in a quantitative sensitivity study to determine which characteristics have a higher influence on the accurate match to well-test response. We used experimental design to optimize the number of simulations needed for a sensitivity study and history match. The sensitivity analysis revealed a strong influence of matrix quality on the pressure response, suggesting that the history match can be specific to the simulated process and not necessarily unique. The results emphasize the contribution of matrix in the Midale reservoir and the need to simulate a broader range of processes for an accurate description of the fracture/matrix system dynamics. In a general sense, the approach used in this study proved to be useful in integrating fracture data from different sources and assessing its reliability and relative importance.

Enhanced oil recovery from challenging/complex fields requires extensive analysis of reservoir structure and good understanding of the effect of this structure on the dynamics of the process. Naturally fractured reservoirs are good examples of this kind and their fracture network characterization is still a bigchallenge. In this study, we analyzed the fracture network system of a portion of the Midale Field, a naturally fractured carbonate reservoir in the Williston Basin of southeastern Saskatchewan, Canada. Our study aims at an extensive characterization of fracture and fracture network properties and construction of a reliable fracture network model for further use in assessing the oil recovery by CO2 injection and CO2 sequestration potential. We integrated static data such as cores, logs and well tests to build 3D discrete fracture network models. Stochastic numerical approach was applied using a commercial software package. A fracture network constructed from static data was calibrated using well test data. Several parameters were evaluated in sensitivity studies to determine those characteristics of the network which have higher influence on the reservoir performance. Simulated well test response was checked against previously published well test data. This study allowed us to recognize uncertainties in critical parameters and propose some measures to manage those uncertainties. Introduction The Midale is a carbonate field located in southeastern Saskatchewan. Following the discovery in 1953, the field was developed on 32 ha spacing and proved to bear 81.9 ? 106 Sm3 reserves of 28.7 °API oil. The field belongs to the Mississippian oil trend located along the northern margin of Williston Basin. Subsequent to primary production up to 1962, the field was subjected to waterflooding on 83 inverted nine-spot patterns. To maintain the production which declined after 1964, an intensive program of vertical and horizontal infill drilling was undertaken. As of the end of 2006, approximately 1,000 wells exist in the field and more than 25% of OOIP was produced with an average watercut of 92%(1). The Midale Field is currently being subjected to tertiary recovery by miscible CO2 flooding. This field-scale CO2 injection was preceded by a 1.78 ha pilot project in 1984 ? 1989 which paved the way for a larger demonstration project. The CO2 Flood Demonstration Project encompassed 10% of the Midale Unit and paved the way for the field-scale application, which is expected to end up with an incremental recovery of 15% of OOIP(2). The Midale Field does not meet typical screening criteria for CO2 flooding. Nevertheless, extensive research and field applications proved that a proper design based on the analyses of the special combination of petrophysical, lithological and fracturing data can result in a successful carbon dioxide flood. The 24 m thick Midale Reservoir section consists of two main layers: dolomite-dominated "Marly" and vugular limestone, called "Vuggy". Both strata contain systematic fractures, though the degree of fracturing varies. Numerous studies conducted on the field revealed some characteristics of the natural fracture network (NFN) by both inverse methods, such as waterflood and carbon dioxide flood performance analyses, and well test analyses and direct methods such as core and log analyses.

As the maturation of conventional oil reserves pushes the industry to explore challenging reserves, state-of-the-art reservoir characterization becomes an integral part of any exploration and production venture. Naturally fractured reservoirs are good examples of such challenging fields. Oil recovery performance estimation from such reservoirs requires a good understanding of reservoir structure and its effect on the dynamics of the process. Addressed in this work is one of the critical issues for fractured reservoirs—that is characterization and 3-D modeling of a fracture network. In this study, we employed an integrated solution by combining "direct" and "inverse" approaches to fracture network characterization in a stochastic numerical model. Static geological data obtained from cores and well logs were used together with dynamic data such as well test response to build 3-D discrete fracture network models. We utilized the data obtained from the fractured carbonate Midale field in Canada. The on-going CO2 injection project requires a reliable description of the fracture system and matrix characteristics in the field for reliable performance analysis. Fracture network constructed from static data was calibrated and validated using well test (interference drawdown and pulse) data. Matrix and several fracture parameters including fracture length, density/spacing, aperture, connectivity, and orientation were evaluated in sensitivity studies to determine which characteristics have a higher influence on the accurate match to well test response. We utilized the factorial experimental design to optimize the number of simulations needed for a sensitivity study and history match. The sensitivity analysis revealed a strong influence of matrix quality on the pressure response. Geological conditions and fracture properties specific to this field explained such distribution of matrix and fracture influence. Through this analysis we were able to clarify the role of fractures in the overall field performance. Matrix/fracture interaction was suggested to be a factor deserving attention. In a general sense, the approach used in this study proved to be useful to integrate fracture data from different sources, as well as to assess its reliability and relative importance.