Abstract We present a joint inversion of PP and PS seismic data for the estimation of petrophysical properties. PP data represent reflections of compressional (P) waves, while PS data represent reflections of compressional (P) waves converted to shear (S) waves. In the proposed inversion, we integrate a rock physics model to link petrophysical and elastic properties and we propose a solution based on a closed form with analytical derivatives for each step, which makes the inversion very efficient. The inversion framework is deterministic, and it is based on the analytical calculation of the Jacobian matrix of the objective function. The extension of the inversion methodology to PS data improves the characterization of the petrophysical parameters as well as the discrimination of reservoir lithologies and fluid types. To validate the inversion method, we test the proposed approach on a realistic synthetic model. For the geophysical model of the inversion, we assume a convolutional operator with a known wavelet, the Aki-Richards linearized approximation, and a rock physics model based on the soft sand model formulation, that combines Hertz-Mindlin theory, Hashin-Shtrikman bounds, and Gassmann’s equations. To assess the contribution of joint inversion, we perform singular value decomposition of the Jacobian matrix of the objective function for the entire seismic trace. This sensitivity analysis reveals that the Jacobian exhibits higher singular values when the inversion is applied to PP and PS data, leading to an increase in resolution and accuracy of the petrophysical properties estimations. We apply the methodology to a seismic dataset from the Jubarte field, a Brazilian sandstone reservoir. The proposed non-linear optimization provides an efficient and robust method for finding the optimal solution of the seismic and rock physics inversion, with limited computational time and high accuracy, despite the presence of multiple local minima. NOVELTY SUMMARY: The study presents a joint inversion technique that combines PP and PS seismic data to improve petrophysical estimations by leveraging the sensitivity of PS reflection coefficients to various parameters. By utilizing a deterministic inversion framework based on the analytically derived Jacobian matrix, the method demonstrates enhanced resolution and accuracy in differentiating reservoir lithologies and fluid types.
Correa et al. (Thu,) studied this question.