A Stable Iterative Scheme for Inversion of Direct Current Resistivity Sounding Data

This paper deals with a scheme for inversion of d.c. resistivity data wherein illconditioning of the system matrix is reduced by making initial guess in terms of reflection coefficients instead of resistivities and ridge regression inversion is performed by composing the system matrix wherein differential weights (ωr and ωt are provided to the set of columns pertaining to reflection coefficient and layer thickness respectively. Four synthetic data sets are chosen to test the inversion scheme and the solutions are compared with those obtained by classical scheme. It is found that the former solution is better than the latter in terms of convergence. One field data set is inverted to obtain roughly same solution while using three different initial guesses.