Improving Sub-salt Seismic Imaging by Incorporating Magnetotelluric Data in a 3-D Simultaneous Joint Inversion Earth Model Building Workflow — A Case History in Walker Ridge, Gulf of Mexico
The inclusion of non-seismic data, such as marine magnetotelluric (MMT), to augment remote seismic data interpretation is not a new concept. Various 2D and 3D integrated inversion approaches have been discussed and published since the early 1990s. However, effectively integrating multiple single-domain data sources into a sizable 3D solution has only recently become feasible with the advent of algorithm refinements and parallel cluster computational speeds. Indeed, the power of a linked multi-domain approach has long been attractive and lies in the relational as well as the independent nature of the data. Independent in that the magnetotelluric data are sampling a completely unique set of data with their own frequency and spectral bandwidth, while at the same time attempting to resolve exactly the same object as the other domains. And although the magnetotelluric data are inherently lower in resolution, when included as part of a 3D simultaneous joint inversion (SJI) workflow with seismic data, their contribution is effectively enhanced. This proves very useful for improving earth velocity models, particularly in complex salt provinces where the method can refine the differences between salt and sediment resistivity and seismic acoustic impedance.
In 2007-2008 WesternGeco acquired over two thousand marine magnetotelluric receiver sites in the deep water northern Gulf of Mexico. The goal was to invert this data in conjunction with seismic data processing to improve the earth model in challenging salt-imaging areas, and thereby the final seismic image. In 2009, two two-block SJI pilot tests were performed, one around the Walker Ridge Block 52 area encompassing Anadarko’s Shenandoah Lower-Tertiary discovery well. The results were very promising and a larger 1,200 km2 production project surrounding the same area was planned and initiated in 2010. By rigorously integrating wide azimuth seismic (WAZ), SJI processing workflows, careful earth model interpretation, “dirty salt” velocity compensation and RTM migration, a clearer picture of the base salt and sub-salt structures have come forth. In addition, the work is helping to confirm earlier theories as to the nature and extent of several deep para-autochthonous salt feeders, as well as better explain the overall salt tectonic development in the area. In this paper we will discuss the SJI process, review our workflows and model building methods and review some final seismic imaging results.
AAPG Search and Discovery Article #90142 © 2012 AAPG Annual Convention and Exhibition, April 22-25, 2012, Long Beach, California