Classification, modeling and characterization of marine carbonate paleokarst reservoirs in Tahe Oilfield, Tarim Basin, China.

Deep marine carbonate paleokarst reservoirs have large hydrocarbon reserves in Tarim Basin, NW China. Geological modeling and characterization are critical to the development of such reservoirs. However, reliable 3D modeling of paleokarst is a challenging task due to their multi-scale, multi-type, highly discrete, and complex spatial distribution characteristics. Here we show the detailed workflow of paleokarst classification, modeling, and characterization methods according to multi-disciplinary data and karst genesis. First, the paleokarst is classified into large-scale karst caves, small-scale dissolved pores, and multi-scale fractures, and the large-scale caves are divided into weathered crust caves, subsurface river system, and fault-controlled caves according to their genesis. Then, in order to obtain an accurate representation of the 3D external shape and internal architecture of such reservoirs, a classified modeling approach for converting seismic attributes to geological objects based on geostatistics and karst theory is proposed. Combined seismic wave impedance inversion with frequency-divided attribute can quantitatively characterize large-scale caves. Collaborative sequential simulation method is used for small-scale dissolved pores simulation. Multi-scale discrete fracture networks are characterized by several methods, including manual interpretation, ant tracing, and stochastic simulation. Finally, a fusion principle and technique for different type of caves, pores, and fractures based on the karst genesis is developed. The proposed technology has been applied to the 3D geological modeling and quantitative characterization of TH80 Block in Tahe Oilfield, and the reliability of model is verified by 3 new drilling wells. The results show that the strong heterogeneity of study area is characterized effectively, and the 3D model finely describes the spatial distribution and petro-physical properties of different reservoir groups. The reserve size and composition are calculated and refined based on the model. The remaining oil distribution is clarified after model applies to reservoir numerical simulation, which lays reliable geological foundation for further adjustment of development plan. [ABSTRACT FROM AUTHOR]