We performed two non-linear iteration of tomography. Figure 9 shows our velocity model. Note that our velocity still follows structural dip. This is especially obvious around the fault at 10 km and above the top of the salt. If we examine the same CRP gathers, Figure 11, we see that we have done a better job flattening the reflectors and have higher frequency events. If we look at the semblance along the reflectors, Figure 10, we see that have significantly less moveout.
Figure 9 Final velocity
The true test is the final migrated image. If we look at the final migrated image we can see better continuity of the reflectors and generally higher frequency image. The final migration image (Figure 12) is quite similar to our initial image (Figure 4). If we take a closer look at the top of the salt body (Figure 13) we can begin to see some differences. At location `A' and `C' we have done a better job defining the salt boundary. At `B' the reflector is sharper and has a more realistic shape. If we look at the side of the saltdome, Figure 14, we can see more improvements. The reflectors are more continuous `A' and extend closer to the salt boundary `B'. We also have done a better job recovering the bottom salt reflector, `C'.