A decorrelation-based hybrid global search algorithm for inversion of 1D magnetotelluric data

Abstract

An efficient algorithm is presented in this paper to improve the perturbation efficiency of the adaptive downhill simplex simulated annealing (ASSA) method for magnetotelluric inversion with oblique correlated misfit valleys. The correlated model space is rotated to a less dependent space which is defined by the eigenvectors of the parameter covariance updated progressively by the inversion itself. The downhill simplex step and rotation work together to decorrelate the correlated model space. The application to two synthetic cases and real data indicates that ASSA in the rotated space generally has a better convergence and efficient behaviour than ASSA without rotation. In the rotated space, the high rejection rates, which occurred in the unrotated space, are avoided. At low temperature, the estimated covariance can be used to approximate the global covariance. For all cases, ASSA in the rotated space gives better inversion results.