Elastic FWI of multi-component ocean-bottom seismic to update shear-wave velocity models

Elastic full-waveform inversion (FWI) is increasingly recognized as a powerful tool for building compressional velocity (Vp) models, primarily utilizing diving and reflected P-waves. However, updating shear velocity (Vs) models through S-wave inversion in elastic FWI has traditionally been more challenging due to data acquisition limitations and the reduced sensitivity of surface seismic to Vs. This paper introduces a practical methodology for low-wavenumber Vs updates using elastic FWI, driven by converted waves in multi-component ocean-bottom seismic data. The approach involves two key steps: first, building a high-quality Vp model using hydrophone and vertical geophone data, and second, using horizontal geophones to reconstruct the low-wavenumber components of the Vs model from converted wave kinematics. We demonstrate the effectiveness of this methodology in a field application, showcasing improved PS reverse time migration (RTM) imaging and better alignment with PP RTM. Additionally, we compare the elastic FWI approach to a Born-based PS-reflection FWI method, with the elastic FWI model yielding superior PS RTM imaging.