Estimating near-surface anisotropy using azimuthal seismic data without NMO correction

Huaizhen Chen, Kristopher A. Innanen

The presence of near-surface anisotropy causes seismic wave velocity and reflection amplitudes to vary with azimuth, especially at large angles of incidence. In order to improve the accuracy of anisotropy estimates, it is essential to include seismic amplitudes recorded at large angles of incidence. However, seismic waveforms at these angles are usually stretched during the normal moveout (NMO) correction processing and are usually muted. To address this issue, we present a forward modeling approach to generate pre-stack seismic gathers without NMO correction at different azimuths, and we also propose an inversion method to estimate near-surface anisotropy using these gathers. We first re-express the incidence- and azimuth-dependent P- and S-wave velocities using near-surface anisotropy indicator (i.e., layer weakness), and then we derive the P-wave reflection coefficient and azimuthal elastic impedance (AEI), which are functions of P- and S-wave velocities, density, and layer weakness. Based on the derived AEI, we develop a new convolution model that incorporates an operator to generate seismic data of different azimuths withoutNMO correction, and we present a Hamiltonian Monte Carlo (HMC) inversion method of employing azimuthal seismic data without NMO correction to estimate elastic properties and near-surface anisotropy indicator. We validate the robustness of the proposed inversion method using noisy synthetic seismic data, which reveals reliable inversion results areobtained for characterizing near-surface anisotropy.