Application of prestack inversion attributes combined with statistical method to predict lithology: a case study in Western China

摘要

This paper presents a case study of lithology prediction using prestack inversion attributes combined with statistical method in Longwangmiao Formation of Block C in Sichuan Basin, Western China. The main objective of the study was to discriminate gas-bearing dolomite from other lithologies, and to identify gas zones within Longwangmiao Formation. Four lithologies (argillaceous dolomite, tight dolomite, dolomite reservoir and gas-bearing dolomite) were identified on well logs. Rock physics analysis determined that P-impedance and density are sensitive parameters to identify lithology. Probability density functions (PDFs) for each lithology were established based on rock physics analysis. Prestack gathers conditioning was conducted before inversion to get high quality prestack seismic data. Simultaneous inversion was carried out for prestack inversion attributes. The prestack inversion attributes and PDFs were combined to create the lithology volume by Bayesian classification. The predicted lithology volume is highly consistent with geology and has good match with the well drilled later. This prediction method is effective in Block C.