Automatic Estimation of Regularization Parameter by Generalized Cross Validation (GCV) Method in 3-D Smooth Inversion of

Abstract

Summary Gravity inverse problem, like other geophysical data inversion, is an ill posed problem that could be solved using Tikhonov regularization. The regularization parameter is one of the effective parameters for obtaining optimal model in inversion of the potential field data similar inversion of other geophysical data. In other word, the regularization parameter has been important parameter for more stability and convergence in inverse modeling. This parameter is experimentally estimated in the most of the inversion method. There are different approaches for automatic estimation of regularization parameter in 3D inversion. In this paper the modified Generalized Cross Validation (GCV) as a new method was used for estimating the regularization parameter in three-dimensional (3-D) inversion of gravity data. However, we used the conjugate gradient (CG) algorithm for computation of model parameters