Overlap! OTC 2000 and the 9th International Workshop on Seismic Anisotropy—16 years in the making and a cast of thousands

Abstract

After 20 years of working with seismic anisotropy and multicomponent reflection data (and attending 16 years of workshops on seismic anisotropy), I am convinced that 2000 is the watershed year! Now, many geophysicists know that: 1. depth migration places reflectors too deep if the focusing velocity is used for the depth scale and that layer anisotropy (“transverse isotropy with a vertical axis”) is the culprit; 2. vertically aligned fractures and/or unequal horizontal stresses split an arbitrarily polarized shear wave into two differently polarized shear waves, which travel at different velocities; 3. P -waves propagate at different velocities parallel versus perpendicular to aligned fractures and/or in the presence of unequal horizontal stresses. Various and sundry companies have claimed over the years that they have taken advantage of these phenomena and that their “correct” use of anisotropy and/or multicomponent data has improved the efficiency of their exploration and development. But this is the first year that a major oil company, Elf, supported its claim by releasing reserves estimated with the use of anisotropy and the change in reserves estimates resulting from the use of anisotropy when compared to isotropic processing and interpretation. This watershed event occurred on 3 May 2000 at the Offshore Technology Conference. To set this event in context, let's look first at the 9th International Workshop on Seismic Anisotropy, held 26–31 March 2000 near Houston, Texas, U.S. At that meeting, four Elf geoscientists (Berthet, Williamson, Sexton, and Mispel) presented their methods to effectively estimate, map, and use layer-anisotropy (transverse isotropy with a vertical axis, or TIV) parameters that affect prestack depth migration in the talk “Anisotropic prestack depth migration: An offshore Africa case.” A sand reservoir was encased in thick shales whose strong layer anisotropy caused nonhyperbolic moveout on P-P far offsets. The variation from hyperbolic traveltimes has been characterized in terms …