Elastic Time-lag full-waveform inversion using OBN data in shallow water environments

In shallow water environments the presence of a hard water-bottom and near-surface geological complexities can give rise to strong guided-waves and other elastic effects in the recorded data. These effects can pose challenges for conventional full-waveform inversion (FWI) based on acoustic approximations, leading to dedicated workflows involving more robust cost functions. Nevertheless, accurate modelling of shallow heterogeneities and strong contrasts using an elastic engine is expected to yield improved inversion results. In this paper, we propose a FWI workflow utilizing ocean bottom node data with a time-lag cost function to overcome many of the modelling challenges and demonstrate it on a shallow-water data set from the North Sea. We generate a 40 Hz acoustic FWI result that heals many of the deeper distortions observed in conventional imaging caused by the shallow complexities. In addition, we highlight that elastic FWI further improves the conventional image, relative to that from the acoustic FWI, especially in the deeper chalk package where elastic effects are expected. Finally, we form 40 Hz FWI Images from acoustic and elastic FWI and find both outperform conventional imaging, with the elastic result having the best event continuity and signal-to-noise ratio from the shallow to the deep.