next up [*] print clean
Next: Introduction Up: Table of Contents

A self-adaptive algorithm for choosing reference velocities in the presence of lateral velocity variations

Huazhong Wang and Guojian Shan

wang@sep.stanford.edu, shan@sep.stanford.edu

ABSTRACT

Seismic wave propagation depicted with the perturbation theory has important and wide-spread uses in reflection seismology. As we know, in perturbation theory, wave propagation needs a reference velocity. The closer the reference velocity is to the true velocity, the more accurate the wave propagation is. However, it is not easy to choose reasonable reference velocities in the presence of severe lateral velocity variations. Assigning a reference velocity value at each spatial point is not computationally feasible, because there is a trade-off between the calculation cost and the number of reference velocities. We show that the accuracy of seismic wave propagation can be more easily improved by choosing a set of reasonable reference velocities rather than by optimizing a one-way wave propagator. Therefore, we introduce a self-adaptive approach to choose a set of reference velocities for an extrapolation layer, in the presence of lateral velocity variations. Through sorting the velocity data an array with increasing values, and by setting a threshold average-velocity ratio or velocity- variance ratio, we can choose a set of reasonable reference velocities for wavefield extrapolation. This method can also be used for image edge detecting. It is flexible and computationally cost-effective.



 
next up [*] print clean
Next: Introduction Up: Table of Contents
Stanford Exploration Project
5/3/2005