Abstract:  Cap rocks are  of great significance for effective reservoir characterization, reservoir quality assessment, and oil and gas reservoir development. The existing marine seismic prediction methods mainly focus on reservoirs, lacking research and analysis on the spatial distribution of cap rocks. It is urgent to establish a pre-stack seismic inversion prediction method for marine oil and gas cap rocks to provide technical support for the seismic prediction of marine oil and gas cap rocks. Therefore, this paper proposes a pre-stack seismic prediction method for marine cap rock facies and physical property parameters based on rock physics constraints, and conducts an in-depth study on the seismic description of  marine cap rock facies and physical property parameters. First, through the rock physics analysis of marine strata, the marine oil and gas cap rock facies and physical property sensitive elastic parameters are optimized. Secondly, based on the rock physics model, the seismic reflection coefficient equation and elastic impedance equation directly characterized by the lithofacies sensitive parameters and physical property parameters are derived respectively, providing technical support for the seismic characterization of the spatial distribution of the marine cap rock facies and the seismic prediction of the physical property parameters. In order to solve the problem of unstable inversion results caused by the correlation between model parameters, the singular value decomposition method is used to realize the decorrelation of model parameters in the seismic prediction process, and the inversion objective functions of sensitive elastic parameters and physical parameters of   marine cap rocks  are derived respectively under the framework of Bayesian inversion theory. This method is applied to actual data to verify the rationality and effectiveness of the pre-stack seismic inversion prediction method for marine oil and gas cap rocks. The results show that the Poisson impedance prediction accuracy can reach 95%, and the porosity prediction accuracy can reach 85%.

Key words: marine oil and gas, caprocks, porosity, Poisson impedance, pre-stack inversion, rock physics model