Multi-Domain Non-Seismic Constrained Velocity Model Construction: A Better Approach to Pre-Stack Depth Migration in Deep Water Frontier Basins

Raterman, Nickolas¹; Tulbah, Basil; Yousuf, Taqi; Snyder, Fred C.; Saragoussi, Emmanuel; Miranda, Francisco; Hili, Iskander; Medina, Elena; McNeice, Gary; Colombo, Daniele; Celis Franco, Valmore; Ramananjaona, Christophe; Scalvinoni, Alberto; Fox, Adam; Sirtori, Andrea; Fainstein, Roberto; Tubbs, Robert; and Zinger, Michael
¹[email protected]

We present an industry first application of the integration of four multi-domain non-seismic measurements during the initial pre-stack depth migration workflow. The complex geology of the Red Sea, a late Oligocene to present rift basin, presents many seismic imaging challenges that limit the effectiveness of conventional marine processing and migration workflows. These challenges include: a shallow unconformity associated with a doubling of seismic velocity less than 500 meters below the sea bottom; over 5,000 meters of mobilized, complexly stratified, and aerially heterogeneous layered evaporite; faulted, laterally variable syn-rift stratigraphy consisting of clastic and carbonate sediments; and significantly extended basement of variable composition, age, and tectonic history. Considering these geologic challenges and the lack of off-set well control, we acquired controlled source electromagnetic, marine magnetotelluric, full tensor gravity gradiometry, and airborne gravity and magnetic measurements immediately before or after the acquisition of wide-azimuth seismic data. Using two different approaches to velocity model construction, we launched parallel depth migration efforts to assess the impact of the non-seismic data on the final seismic image. The first approach relied exclusively on the seismic data using conventional seismic analysis and processing steps such as reflection tomography, velocity floods and scans, and interpretation based masking. The second approach included the same methodologies as the former, but also incorporated forward modeling, single-domain- and simultaneous-joint-inversion of each non-seismic measurement to provide constraints throughout velocity model construction. Analysis of the non-seismic data also revealed significant geologic variations that were not considered during initial interpretation of the regional seismic data and were not readily apparent from analysis of the seismic data alone. Our results show that incorporation of non-seismic data at the onset of seismic processing and depth migration can significantly increase seismic image quality and lower interpretation uncertainty. We suggest that the incorporation of non-seismic data at the onset of seismic processing and depth migration should be an industry best practice in frontier exploration without offset well control.

AAPG Search and Discovery Article #90166©2013 AAPG International Conference & Exhibition, Cartagena, Colombia, 8-11 September 2013