The primary objective of geoscientists in reservoir characterization is to produce the most reliable lithofacies model from various domain sources such as seismic data and well data. The minimal contrast of acoustic impedance (VP), shear impedance (VS), and density between lithofacies make it difficult for geophysicists and geologists to characterize the prospects and other lithologies. Conventional modeling techniques have many constraints in handling structural complexity, stratigraphical complexity (velocity and density were almost similar for all deposits), and limited resolution seismic data. This study developed an integrated methodology to address complexity in producing the lithofacies model in the reservoir characterization. This integrated study enhances lithofacies characterization by reproducing the continuity of geological facies in the meandering- fluvial sandstone reservoir from North-east India. It involves simultaneous prestack seismic inversion (SPSI), well data cross plot analysis, non-parametric statistical classification, multi-point geostatistical simulation (MPS), and Poisson impedance analysis.
In the first part of this study, simultaneous prestack inversion was conducted to estimate the elastic properties of the study area from prestack seismic data. Next, the non-parametric statistical technique was applied to cross-plot of well logs to estimate probabilistic density functions (PDFs) for pre-classified lithofacies. After that, estimated PDFs were integrated with seismic elastic properties by the approach of Baye’s classification to develop the probability volumes of each pre-defined lithofacies. Finally, the Single Normal equation Simulation of MPS is applied to Hard data (Well log data) and soft data ( lithofacies probability volumes from seismic data) to estimate realizations. These realizations can be averaged to estimate the improved model of lithofacies. The outcomes of this study were compared qualitatively and quantitatively. Qualitative analyses were conducted by comparing the results of this study with the results of conventional techniques. A sensitivity analysis was conducted to quantitatively estimate the present proposed methodology's robustness. Poisson impedance inversion was also applied to the study area to estimate the lithofacies model from prestack data and compare these results with the proposed methodology results. These results were proven to improve the lithofacies characterization and successfully reproduced the missing lithofacies that were not characterized with conventional methods. This study successfully addressed the issues in classifying complex sandstone deposits in Northeast India.