Abstract: Seismic Data Processing in Anisotropic Media

Valmore Celis

Conventional processing is designed for isotropic media and cannot handle angle-dependent velocity. However, most crustal rocks are found experimentally to be anisotropic; furthermore, it is known that in sedimentary rock sequences the anisotropy may be caused by preferred orientation of anisotropic grains (such as in a massive shale), preferred orientation of cracks (such as parallel cracks), thin bedding of isotropic or anisotropic layers, etc.

Some authors have found that large position errors arise for steep reflectors when anisotropic effects are ignored in processing, which lead to mispositioning or complete loss of steep structures, fault planes, etc. On the other hand, the influence of anisotropy is also seen during velocity estimation, which is crucial to get a good image of the subsurface. Therefore, it is very important to take into account the anisotropy effect when it is present in the data.

In this work the differences between isotropic and anisotropic processing are shown, both on synthetic and real data when anisotropic media are considered. Here, anisotropy is included in two important steps of the processing sequence, namely, dip-moveout (DMO) and migration. The migration process is treated with explicit extrapolators in anisotropic media and dip-moveout is based on the Fourier transform approach.

Results show how steep reflectors and fault planes are much better imaged when the data is processed with algorithms which take into account anisotropy than with conventional isotropic algorithms.

AAPG Search and Discovery Article #90951©1996 AAPG International Conference and Exhibition, Caracas, Venezuela