Finite Difference Migration
We used the same velocity estimated for Kirchhoff migration, 1455 m/s, the time step was set to 10 ms and the max dip was set to 80 deg.. Fig \ref{508952} shows the results of migration after applying these parameters. Notice that the migrated image (right pane) is noisy specially along the legs of the syncline. Compared with the Kirchhoff migrated image(Fig. \ref{268565}), the finite difference migration fails to resolve high dipping angles, as the flanks of the syncline. We tried to improve the image by  halving the time step to 5 ms as shown  in the center pane of Fig. \ref{500139}.There is a mild improvement but the result is of lower quality when compared to the Kirchhoff migration results (Fig. \ref{268565}).   We attempted a time step as low as 1ms but it was not allowed by the software. We also tested the effect of a much higher time step, 50 ms. The right pane of Fig. \ref{500139} shows the corresponding results. Observe that the syncline becomes wider and  has a more gently dipping flanks, however there are artifacts when it meets the flat interface. Finally we also tested a higher , 1000 m/s, and lower velocity, 2000m/s. Results are shown in the center and left pane of Fig. \ref{107994} respectively. As expected, a lower velocity, produces a migrated image which stills present concave downward hyperbolas and a higher velocity produces  concave upward hyperbolas.