FUTURE WORK

There are several areas of future work on this project. First, more non-linear iterations need to be done for both the (x,z) and ($\tau$,$\xi$) approaches. A layer stripping approach is also warranted. The w pattern on the third and fourth reflector lead a doublet in our velocity estimate. By resolving the velocity above the second reflector before attempting the global problem we might be able to avoid the doublet. In addition the procedure needs to be tested on a real 2-D data set and then extended and tested in 3-D.

 

tau
Figure 10 Velocity estimate by performing tomography in ($\tau$,$\xi$) space and then mapping back into (z,x) space compared to our velocity estimate in depth, both after two non-linear iterations.