PS energy imaging condition for microseismic data — Part 2: Sensitivity analysis in 3D anisotropic media

Abstract

In microseismic monitoring, obtaining reliable information about event properties, such as the location, origin time, and moment-tensor components, is critical for evaluating the success of fluid-injection programs. Elastic wavefield-based migration approaches can robustly image microseismic sources by extrapolating data through an earth model and evaluating an imaging condition. The success of these imaging methods, though, primarily depends on the elastic model’s accuracy. The previously developed extended PS energy imaging condition can provide valuable information about the accuracy of the elastic model parameters including vertical P- and S-wave velocities as well as anisotropy coefficients. Using the SEG advanced modeling Barrett Unconventional model, we have assessed the influence of errors in the anisotropy parameters by conducting a sensitivity analysis in three types of 3D models: transversely isotropic with a vertical symmetry axis, transversely isotropic with a horizontal symmetry axis, and orthorhombic media. Our analysis on zero-lag and extended PS energy images computed with perturbed anisotropy models shows that event images exhibit different moveout patterns of misfocused energy with respect to the distorted Thomsen parameters [Formula: see text] and [Formula: see text]; however, for this model, the [Formula: see text] parameters have almost no influence on images regardless of the applied perturbations, which are reflected in the minimal traveltime differences in the data. The dependence of microseismic source images on these parameters provides essential insights into anisotropic model accuracy, and it suggests that misfocused energy on extended image gathers may be used as a criterion for updating earth models through anisotropic elastic image-domain inversion.