We used least-squares inverse theory with the AMO operator as the regularization term. This method satisfactorily solved for interpolation of a 3D irregular data set.

We implemented a similar approach for regularizing the *PS* section of
the OBC data set. For this problem, an iterative procedure is
needed due to the dependence of the AMO operator on
the value.

In order to obtain better results in the future, we recommend
the use of a higher NMO approximation to obtain coherence among the
traces to be stacked on the *PS* section.
Additionally,
formulating the estimation problem in a least-squares sense
should allow a better constraint for its calculation, creating
better *PS* regularized sections. This is an ongoing project.

11/11/2002