Characterization of Glutenite Reservoirs Using Bayesian Adaptive Impedance Inversion and Rock Physics

Summary The hydrocarbon resources in glutenite reservoirs are abundant in China. However, due to the near-source sedimentation and rapid facies transition mechanism, the glutenite reservoirs have strong heterogeneity with low porosity and complex pore structures, which makes it difficult to be predicted from seismic data. In this paper, we combine the Bayesian adaptive impedance inversion with rock physics analysis to characterize the lateral variation of glutenite reservoirs. The deduced prior stabilizer can be automatically adjusted based on the seismic noise level and obtain the best compromise between resolution and stability of the inversion result. In addition, the trace-by-trace inversion strategy makes use of the correlation advantages of adjacent seismic traces, making the inversion results have good lateral variations and geological features. Application in Mahu oil field of west China shows that the prediction result of glutenite reservoir has clearly lateral variation and high vertical resolution and matches well with the drilled wells and the sedimentary trend, which demonstrates the effectiveness and feasibility of this method.