Seismic fracture characterization in tight sand reservoirs: A case study of the Xujiahe Formation, Sichuan Basin, China

Nature fractures developed in tight gas-bearing sand reservoirs significantly increase reservoir porosity and permeability and vary in scale, angle, aperture, density. The detection of effective hydrocarbon fractures are requirements for commercial production from these reservoirs. Based on the equivalent medium theory, we propose an approach to eliminate the influence of low-productivity horizontal bedding fractures and directly extract the orientation and intensity information of effective high-angle fractures in the reservoir where hori-zontal bedding fractures are mainly developed. First, we assume that the complex tight gas-bearing sand reservoir is considered as a weakly orthorhombic medium with a single suite of parallel vertical fractures embedded in a vertical transverse isotropy (VTI) medium. Then, we derive a linearized relationship between the reflection coefficient and the fracture orientation and intensity in an orthorhombic medium. Furthermore, we implement a Bayesian elastic impedance (EI) inversion workflow, in which the seismic data are transformed to EI datasets, and subsequently EI datasets are used to estimate the fracture orientation and intensity. Finally, the proposed approach is demonstrated by tests on synthetic examples and field data sets. The predicted results illustrate the approach is feasible and robust, and agree well with the log data and geological interpretation. Compared with the conventional method, our approach has the greater advantages to accurately predict the fracture characteristics. We conclude that the proposed approach can reliably produce the distributions and properties of vertical fracture systems, which may provide crucial fracture information for tight gas-bearing sand reservoirs characterization.