3. AVO Inversion of a Mobil Data Set

Abstract

The objective of the present work is to obtain estimates of the P-and S-wave velocities and densities of the subsurface. The inversion is carried out on 952 τ -p transformed CMP gathers, each containing 61 p values ranging from 0.05 to 0.35 s/km.The forward modeling is performed by convolving the reflectivity with the wavelet, and it includes water bottom multiples, transmission effects, absorption and array filter effects. The basic assumption is that the subsurface is close to horizontally layered. A damped Gauss-Newton algorithm is used to minimize a least-squares misfit function.Comparison with the two well logs shows good agreement for the P-wave velocity estimates in the areas where the geology is close to horizontally stratified. In deeper and faulted areas the deviation between estimated and measured P-wave velocities is larger. The same trend holds for the S-wave velocity estimates, while the densities are the poorest resolved seismic parameter.