Deep Updates - Challenges and Solutions for FWI

Abstract

Conventional velocity model building (VMB) in complex regimes, such as intra and subsalt data, requires time-consuming manual intervention. It is a process that can produce unreliable models, leading to an increase in uncertainty for subsalt lead evaluation. We demonstrate an application of Full Waveform Inversion (FWI) to refine legacy velocity models generated by conventional VMB. We present our solution on a simultaneous long-offsets (SLO) dataset from the Gulf of Mexico, acquired with dual-sensor streamers, which provided low-frequency rich data. The SLO configuration recorded data with 16 km of offset, enabling both refractions and reflections to update the deeper parts of the velocity model. We employ an FWI velocity gradient that eliminates the migration isochrones. This provides support for the intra and subsalt model updates by removing the reflectivity imprint from the updated models. The FWI application successfully refined the geometry of the salt bodies including the base salt and the intrasalt enclosures. RTM images show a marked uplift, particularly for both the salt flanks and subsalt reflectors.