Next: PS common-azimuth downward-continuation Up: PS common-azimuth migration Previous: Summary

Introduction

The cost of 3-D prestack wave-equation migration is proportional to the size of the dataset. Common-azimuth migration is an efficient and robust technique to obtain accurate single-mode 3-D seismic images. This technique takes advantage of the reduced dimensionality of the computational domain. The operator assumes a dataset with zero crossline offset (HY=0); that is, all the traces in the data share the same azimuth Biondi and Palacharla (1996). Common-azimuth data are the result of collection of actual physical experiments or the result of a processing technique to reduce the dimensionality of the data space.

The common-azimuth downward-continuation operator is based on the stationary-phase approximation to the full 3-D double-square root equation. The extension of this downward-continuation operator to converted-wave data is as simple as using two propagation velocities, a P-velocity model and an S-velocity model. This chapter presents a geometric interpretation for the stationary-phase approximation applied to the converted waves experiment. This geometric derivation result validates the stationary path for converted waves. This process results in the PS common-azimuth downward-continuation operator (PS-CAM). The PS-CAM operator also has the same advantage as the common-azimuth operator, that is a decrease in the computational cost for 3-D prestack data.

The previous chapter presented a method to process the PS section of 3-D prestack OBS data, the result of this process is a regularized four-dimensional cube. This result is suitable for the PS common-azimuth migration operator. I obtain two 3-D images for the Alba oil field. The firs image is the PS common-azimuth migration of the regularized dataset using Normal Moveout plus stacking, I will refer to this result as PS-NoMoRe. The second image is using the PS-CAM operator on the result of using the PS-AMO operator to regularize the five-dimensional data cube, I will refer to this process as PS-AMORe. The final results show that the PS-AMORe method produces images with more coherent reflectors along the reservoir level. Also, geological features (i.e. channels, faults) are now clearer and their interpretation is more feasible after PS-AMORe.

Next: PS common-azimuth downward-continuation Up: PS common-azimuth migration Previous: Summary
Stanford Exploration Project
12/14/2006