All of the previous results were obtained using the exact Marmousi velocity model, but in a real exploration scenario the determined velocity model would generally be less detailed and less accurate.
Figure is an approximate model obtained by defining the gross features of the true model and smoothing the slowness within each of the defining regions. This is the same model used by Nichols and it is intended to to be a realistic representation of what an explorationist might reasonably attain using geological information, velocity estimation, and well information.
Figure is the result of standard Kirchhoff migration with first-arrival traveltimes calculated from the surface using the approximate velocity model. There is not too much difference between this and the same migration using the actual velocity model (Figure ). The result of layer-stripping Kirchhoff migration is presented in Figure . The anticline and target structure is imaged, but not as well as with the actual velocity model (Figure ).
Figure is a comparison of the target zone images generated with the approximate velocity and the actual velocity. Although the image obtained with the exact velocity is superior, the one obtained with the approximate velocity is surprisingly good. All the relevant features of the target are imaged, even though the lateral coherency and structural definition are not as good as when the actual velocity is used. This shows that although the method is sensitive to velocity, it still produces a good image which could be improved by a refinement of the velocity model.