Final result

If we perform another iteration of tomography we see further improvement. Figure 12 shows the velocity model. The velocity has been further increased in the the chalk layer on both sides of the salt (`A' and `B'). On the right side we also see a break in the high velocity zone developing (`C'). In the Liassic where the chalk velocity was increased in the first iteration it has been decreased by the second iteration, especially on the right side of the salt dome (`D'). This decrease continues all the way to the salt edge.

 

vel-final Figure 12 Final velocity. The velocity has been further increased in the chalk layer on both sides of the salt (`A' and `B'). On the right side we also see a break in the high velocity zone developing (`C'). In the Liassic where the chalk velocity was increased in the first iteration it has been decreased by the second iteration, especially on the right side of the salt dome (`D'). This decrease continues all the way to the salt edge.

The CRP gathers (Figure 14) are generally flatter and more coherent. The semblance along the reflectors (Figure 13) is more continuous and closer to zero curvature.

 

elf-sem-ref.vel.best.steer
Figure 13 Semblance panels from ten of the reflectors used in tomography.

 

moveout-vel.best.steer
Figure 14 CRP gathers after the second iteration of tomography.

The migration (Figure 15) using the new velocity (Figure 12) has also improved. The chalk layer boundary (`D') is sharper. The top of salt has fewer artifacts (`B') and the bottom of salt is more continuous (`C'). The upper reflectors on the left side of the image are more continuous. The greatest improvement is seen below the salt edge (`A'). The reflectors are more continuous and of consistent amplitude. If we look at the lower portion of the image in more detail, Figure 16, the differences become even more obvious. The salt reflector is flatter and continuous (`A'). The reflections to the right of the salt (`C') body are of higher frequency and we are beginning see some structure that was not obvious in the initial migration. Below the salt edge (`B') we see dramatic improvement. In the initial migration, reflector continuity was lost under the salt edge. In the final migrated image we are able to continue the reflectors much further.

 

mig-final
Figure 15 Final migrated image. Locations A-D show improvement compared to Figure 3.

 

bot.compare
Figure 16 Result of migrating with the initial velocity, top panel, and the final velocity, bottom panel. We see a more coherent and believable bottom salt reflection (`A'). The reflectors bounding the salt on the left are more continuous (`B') and of higher frequency (`C').