Paper Summary
Success in the application of gradient-based full-waveform inversion (FWI) depends on the extraction of the lowest possible wavenumber content provided by the data. In shallow water scenarios with shallow targets, the refractions and diving waves present in the large offsets of streamer data provide an advantageous framework for such success. In deep water scenarios with streamer data containing only reflections, the extraction of the long wavelength features requires either a boost in the lowfrequency content of the data and/or strategies to precondition the data and the model to reduce the nonlinearity of the problem. In this work, we invert reflection events from fully deghosted streamer data that were acquired in deep water offshore of Norway. The deghosting was accomplished by using dual-sensor streamers and a time and depth-distributed source array. The inverted velocity model shows a significant improvement in resolution in the shallow part of the model, which consequently improves the resolution of the migrated image. Results are also validated by the flatness of common image gathers and the waveform fitting between modeled and field shot records.