Datapages, Inc.Print this page

COBURN, GARY W., Bell Geospace Inc., Houston, TX

Abstract: Base of Salt Determination Utilizing Full Tensor Gradient Data

The determination of the base of a salt body in some areas of the Gulf of Mexico is extremely difficult utilizing seismic data alone. This is due primarily to subsalt imaging problems. Full Tensor Gradient data was combined with the seismic data in an effort to determine the edges and base of salt in the Green Canyon area. A 2-D TGS regional line that runs from Green Canyon southeast to Walker Ridge was used in this study. A section (approximately 30 mi) of this line runs diagonally across a grid of 3-D gradient data acquired by Bell Geospace, Inc. in 1995. The top of salt was identified and a reasonable interpretation made as to the bases and edges of the salt bodies. These horizons were then imported into a gravity gradient-modeling program. The response of the data measured by Bell Geospace Inc., versus that of the original model indicated significant differences relating to the size, shape, and thickness of the salt bodies within the area covered by the gradient data. The bases and edges of the salt bodies in the model were then altered to match the various gradients and gravity data as closely as possible. The gradient data indicated that various horizons continue through the seismic data "wipe out" zones under the salt bodies. The salt at one point is less than 2,500 ft thick, creating the potential for subsalt objectives.

A comparison of the original salt model, the gradient salt model and a model using only conventional gravity data was made. The gravity data was unable to determine the thickness of the salt as it is composed primarily of long wavelengths which inhibit its ability to image shallow mass density changes. The differences between the three models show that the gradient data is much more sensitive to the salt masses than conventional gravity data. A determination of the base of salt can be accurately made by incorporating Full Tensor Gradient data in the interpretation process. 

AAPG Search and Discovery Article #90924©1999 GCAGS Annual Meeting Lafayette, Louisiana