The prediction of seismic reservoirs in marine carbonate areas in the Sichuan Basin, southwestern China, is very challenging because the target zone is deeply buried (more than 6 km), with multiphase tectonic movements, complex diagenesis, and low porosity, and the incident angle of the seismic data is finite. We developed reliable hydrocarbon indicators of a marine carbonate deposit based on prestack elastic impedance (EI) and well observations. Although the hydrocarbon indicators can be calculated from elastic parameters, the inversion for EI-driven elastic attributes is usually unstable. To constrain the inversion process, we discovered a new strategy to recover the elastic properties from EIs within a Bayesian framework (called Bayesian elastic parameter inversion from elastic impedance). We applied the strategy to a carbonate reef identified at the center of a study line based on the geologic context and the seismic reflection patterns. We then used rock-physics analyses to classify the lithologies and the reservoir at a well location. Rock-physics modeling quantified the hydrocarbon sensitivity of the elastic attributes. Fluid substitution was used to investigate the effects of pore fluids on the elastic properties. A comparison of two synthetic amplitude-versus-angle responses (for gas and brine saturation) with real seismic data showed that the reservoir was gas charged. Using well-based crossplot analyses, reliable direct hydrocarbon indicators can be constructed for a deeply buried gas reservoir and were effective for interpretation in an area of marine carbonates in the Sichuan Basin.

You do not currently have access to this article.