Fast 3-D converted-wave depth-variant common conversion point stacking

Shaowu Wang, Mark C. Lane, Donald C. Lawton

In converted-wave data processing, one of the key steps is common conversion point (CCP) stacking or binning. There are two methods of implementing CCP binning. One uses asymptotic CCP binning and the other uses depth-variant CCP binning. Although asymptotic CCP binning is simple and fast, it is only a first-order approximation of the true conversion point. Conventional depth-variant CCP binning methods are accurate for horizontally layered media. However, because a complicated expression is required and calculations must be performed at each binned time sample, the method is time-consuming.

In this paper, we develop a fast 3-D converted-wave depth-variant common conversion point (CCP) stacking method. In its implementation, source-receiver azimuth is taken into account and samples are mapped into their new CCP binning locations block-by-block, instead of sample-by-sample. We also demonstrate a modification of the algorithm, extending it to depth-variant velocity models with a constant or slowly varying ratio of P-wave to S-wave velocities. Finally, a physical modelling example demonstrates the feasibility of the new method.