Methodology

Residual migration is a very useful tool for velocity analysis. Even though it is based on a constant velocity media, the residual migration depends only on velocity ratios. Therefore, it is approximately correct and applicable for variable velocity models, at least with depth migration.

Since I handle Stolt residual migration for converted waves using three parameters, it is important to simplify the problem. The more parameters we have to search for, the more complicated the problem is. Using the PP data in order to update the P velocity model the best possible way, will lead to only a two-parameter estimation ($\rho_s$ and $\gamma$).