INDEPENDENT PUSHES HIGH-TECH FRONTIER THROUGH PARTNERSHIP

Bob Tippee Managing Editor - Economics and Exploration State-of-the-art geophysical technology, requiring access to the biggest and best computer systems in the world, is not beyond reach of independent oil and gas producers. Through effective partnerships, independents are even taking part in the development of technology requiring resources that few companies of any size could afford acting solo.

Nov. 20, 1995

2 min read

DEPTH MIGRATION ESSENTIALLY COMPARES A SET OF seismic data to a depth model and calculates corrections to remove differences in a series of steps called iterations.

The Oryx-Sandia-UTD Crada PSDM went through three iterations for a full 3D seismic data volume. Oryx used Paradigm Geophysical Corp.'s Geodepth software with some supplemental programming to build and update the velocity model.

It developed an initial velocity field using what Senior Geophysicist Marcelo Solano calls a "modified conventional" layer-stripping approach for shallow layers, assuming a constant velocity gradient in each layer. Velocities came from analysis of common midpoint (CMP) stacks of shot lines.

Data from a check shot survey in the Garden Banks discovery well constrained velocity ranges for interpretation of four shallow reflection events and inversion of interval velocities. Migration of these data produced the initial, horizon-based was assumed for the area beneath the interpreted events.

The Oryx PSDM used a Kirchoff integral algorithm, a standard depth migration method that allows for direct, target-oriented mapping of time-domain reflection events to their correct space or depth positions from travel-time tables. The program yielded target-oriented image gathers and a volume of depth-migrated stack traces.

The image gathers provided the basis for updates to the velocity field for each iteration. Interpreters calculated velocity flattening events - or aligning reflections in time - on the gathers.

The approach thus used the initial, horizon-based model only to produce a first set of migrated gathers and used image-based modeling for the iterations. Often in depth migration, seismic data are adjusted to fit a velocity-based geologic model throughout the iterations.

For the first and second iterations, Oryx used a grid of 0.62 by 0.5 miles (20th inline and 60th common reflection point, or CRP). The third iteration had extra control points in the cen- ter of the survey, with a final grid of 0.31 by 0.25 miles (10th inline and 30th CRP).

The image-driven approach in velocity modeling makes no assumptions about velocity. Its only purpose is to produce the best possible image, but it also leads to depth misties.

The output data set included 169 bin lines with 371 traces each, one for every other common midpoint. It covered an area of 5.22 miles by 6.13 miles.