Principal Investigator: Chester J. Weiss (505-284-6347)

Sponsor: Laboratory Directed Research & Development (LDRD) Program

Duration: (FY)2002-2005

Description: In contrast to the more popular methods of geophysical inversion which seek solutions to an unconstrained optimization problem by imposing stablizing constraints in the form of smoothing operators on some enormous set of model parameters (i.e. "overparametrize and regularize"), we investigate an alternative approach whereby sharp jumps in material properties are preserved in the solution by choosing as model parameters a modest set of variables (such as B-splines) which describe an interface between adjacent regions in physical space. While still overparametrized, this choice of model space contains far fewer parameters than before (thus easing the computational burden, in some cases, of the optimization problem) and most importantly, naturally accomodates the abrupt changes in material properties associated with lithologic boundaries as well as the interface between buried cultural artifacts and the surrounding Earth. We have thus been working on a multi-prong approach whereby the optimization framework leverages existing Sandia software packages (Moocho and Trilinos) and the forward solver is provided by earlier efforts (with some modification) in finite difference modeling applied to marine hydrocarbon exploration problems.

Papers, reports and the latest hot results: Stay tuned!