Next-generation resolution through multiparameter FWI imaging

Abstract

Conventional seismic processing and imaging workflows employ a range of techniques such as deghosting, designature, and demultiple to attenuate parts of the wavefield that would otherwise degrade the final image due to the single scattering assumption used in common imaging algorithms. These processing steps can be time-consuming since they are performed in a linear fashion and require significant effort to parameterize and apply. Multi-parameter full waveform inversion (FWI) imaging offers an alternative approach by simultaneously inverting for various parameters including velocity and reflectivity using raw field data as input. This least-squares imaging solution uses primaries, multiples and ghosts to deliver higher-resolution images than traditional imaging algorithms. In this paper, we compare the reflectivity output from an FWI imaging approach against a conventional Kirchhoff 3D pre-stack depth migration (preSDM) and an image-domain least-squares reverse time migration (LS-RTM) at high frequency. We also demonstrate the fidelity of the velocity output from the simultaneous FWI imaging technique.