Joint inversion of PP- and PSV-wave amplitude data for estimating P- and S-wave moduli and attenuation factor

Huaizhen Chen, Shahpoor Moradi, Kristopher A. Innanen

Beginning with re-expressing P- and S-wave velocities in anelastic media, we first propose frequency-dependent P- and S-wave moduli in terms of P- and S-wave moduli at a reference frequency and P-wave maximum attenuation factor, in which we replace S-wave attenuation factor with P-wave maximum attenuation factor. Based on Zoeppritz equations and their linearized expressions for computing PP and PSV-wave reflection coefficients, we derive frequency-component PP- and PSV-wave reflection coefficients as a function of P-wave maximum attenuation factor, from which anelastic impedances for PP- and PSV waves are expressed. Using the derived reflection coefficients and elastic impedances, we establish a two-step inversion approach, which involves the estimation of attenuative PPand PSV-wave anelastic impedances from frequency-components of partially-stacked seismic data, and the prediction of unknown parameter vector (P- and S-wave moduli, density and P-wave maximum attenuation factor) using the estimated PP- and PSV-wave anelastic impedances. Synthetic tests confirm that the unknown parameters are estimated stably and reliably in the case of seismic data containing a moderate Gaussian noise. Applying the inversion approach to a real data set acquired over an oil-bearing reservoir, we observe reasonable results of P-wave maximum attenuation factor are obtained, which may provide additional proofs for fluid identification.