For many years, we have used seismic data processing techniques such as digital filter theory to enhance the signal‐to‐noise (S/N) ratio of the observed data to obtain the best possible image of the subsurface. With our increased knowledge and the increasingly sophisticated computers at our disposal, modern methods of data analysis are tending to be inversions through model fitting, i.e., actually finding the best model of the subsurface which would produce the observed data.