--> Application of Full-Waveform Inversion for Shale Diapirs and Carbonate Rafts in the U.S. Central Gulf of Mexico

AAPG/SEG International Conference & Exhibition

Datapages, Inc.Print this page

Application of Full-Waveform Inversion for Shale Diapirs and Carbonate Rafts in the U.S. Central Gulf of Mexico

Abstract

Abstract

Improving subsalt imaging is the key to reducing risks for subsalt exploration and appraisal. In the US central Gulf of Mexico, shale diapirs and carbonate rafts are associated with complex salt tectonics. They have very different velocities compared to surrounding clastic sediments, and are difficult for conventional reflection tomography to invert. They must be properly resolved in the earth model prior to prestack depth migration.

Full-waveform inversion (FWI) is able to use the displacement between observed and predicted data to derive a low-frequency velocity model. By using refracted and diving wave energy, it is more robust to mitigate cycle skipping with an inaccurate initial model. Because FWI is usually run at low frequencies, statistical shot sampling can be used to reduce the cost. For every iteration, FWI randomly selects some shots according to the shot radius, which is related to FWI frequency. Inserting the salt into the initial FWI model will achieve a better match between the observed and modeled data, thereby, reducing the possibility of cycle skipping and/or reducing the number FWI iterations.

We applied FWI to a large reprocessing project in the US central Gulf of Mexico, covering ∼55,000 km2. We found that FWI was able to invert for a better velocity model in shale diapirs and carbonate rafts by full utilization of the abundant refracted and diving wave energy in both the dual-coil data and traditional wide-azimuth data. We will showcase many low-velocity shale diapirs, mainly observed in the Green Canyon and Keathley Canyon areas, as well as high-velocity carbonate rafts or carapace layers, mainly observed in the Keathley Canyon area. Proper resolution of these geological features in our earth model will lead to improved exploration and development successes.