Seismic rock-physics linearized inversion for reservoir-property and pore-type parameters with application to carbonate reservoirs

Abstract

Carbonate reservoirs hold promising potential for hydrocarbon exploration, but they usually exhibit complex pore structures, which lead to difficulties in the seismic/elastic prediction via conventional rock physics modeling, and in turn, hinder an accurate estimation of reservoir properties from seismic observations. This work presents a seismic rock-physics inversion method to jointly estimating reservoir-property (porosity and fluid saturation) and pore-type (pore aspect ratio) parameters in carbonates from seismic/elastic data. The method considers the aspect ratio as a pending parameter, which favors the pore-type variation so as to describe the structure complexity. To achieve the joint estimation, we derive a linearized rock physics model based on the differential effective medium model and the Biot-Gassmann equation that relate porosity, water saturation, and aspect ratio to elastic parameters. Moreover, to realize the estimation with computational efficiency, we derive the analytical solution to the inversion with the linearized model based on the Bayesian inverse theory. We estimate the pore aspect ratio at a well location to provide prior information for the inversion and employ an iterative inversion strategy with piecewise linearization to improve the result. The proposed method is applied to two carbonate reservoirs including 2D and 3D field data sets. Compared with the conventional rock physics inversion scheme with a constant aspect ratio, the application demonstrates that the method achieves better accuracy for reservoir property estimation, which can be a good indicator for the potential target areas of complex reservoirs.