Satellite Gravity Anomaly Recovery Using Multigrid Methods

Abstract

Dedicated SST- or SGG missions like GRACE and GOCE will provide gravity field information of new quality. But the computation of gravity models from these missions involves the solution of large, dense and ill-conditioned normal equation systems, and therefore fast solvers are needed. One way to deal with the problem is to make use of a hierarchy of lower-dimensional approximations to the unknown gravity anomalies. Such techniques, in numerical analysis well-known as multigrid methods, give rise for fast iterative solvers. We investigate the implementation of multigrid methods to satellite data analysis, including regularized problems. Multigrid algorithms are considered as stand-alone solvers as well as for the construction of preconditioners in the conjugate gradient technique, and numerical results from simulated missions are given. Moreover, an option to accelerate the choice of regularization parameters is shown.