Three-dimensional magnetotelluric inversion using non-linear conjugate gradients

Abstract

We have formulated a 3-D inverse solution for the magnetotelluric (MT) problem using the non-linear conjugate gradient method. Finite difference methods are used to compute predicted data efficiently and objective functional gradients. Only six forward modelling applications per frequency are typically required to produce the model update at each iteration. This efficiency is achieved by incorporating a simple line search procedure that calls for a sufficient reduction in the objective functional, instead of an exact determination of its minimum along a given descent direction. Additional efficiencies in the scheme are sought by incorporating preconditioning to accelerate solution convergence. Even with these efficiencies, the solution's realism and complexity are still limited by the speed and memory of serial processors. To overcome this barrier, the scheme has been implemented on a parallel computing platform where tens to thousands of processors operate on the problem simultaneously. The inversion scheme is tested by inverting data produced with a forward modelling code algorithmically different from that employed in the inversion algorithm. This check provides independent verification of the scheme since the two forward modelling algorithms are prone to different types of numerical error.

An Investigation of the 3D Electrical Resistivity Structure in the Chingshui Geothermal Area, NE Taiwan

“…Several 3D inversion algorithms have been developed since 1993 (Mackie and Madden 1993;Newman and Alumbaugh 2000;Sasaki 2004). Most of these algorithms require powerful workstations or parallel computers, which make field application difficult (Siripunvaraporn et al 2005a, b;Tuncer et al 2006;Xiao et al 2010).…”

Semiautomatic approaches to account for 3-D distortion of the electric field from local, near-surface structures in 3-D resistivity inversions of 3-D regional magnetotelluric data

“…Assumptions involved in one-dimensional (1-D) and twodimensional (2-D) solutions for 3-D structure or 3-D local distortion produce model artifacts (that is, overestimations or underestimations) of target depths that result in false geologic interpretations of the 1-D or 2-D geophysical model. Researchers attempting to alleviate the ambiguity in 1-D and 2-D interpretations of magnetotelluric data have developed a number of 3-D algorithms (Smith and Booker, 1991;Mackie and Madden, 1993;Newman and Alumbaugh, 2000;Zhdanov and others, 2000;Sasaki, 2001Sasaki, , 2004Zhdanov, 2002;others, 2004, 2005;Sasaki and Meju, 2006;Gribenko and Zhdanov, 2007;Mackie and others, 2007;Han and others, 2008;Avdeev and Avdeeva, 2009;Siripunvaraporn and Egbert, 2009;Gribenko and others, 2010;Siripunvaraporn and Sarakorn, 2011;Egbert and Kelbert, 2012;Kelbert and others, 2014). These efforts have also encouraged the acquisition of an increasing number of 3-D arrays of magnetotelluric data because collection of array data, as opposed to profile data, is well suited to 3-D inversion modeling.…”

3D Magnetotellurics for Mineral Exploration beneath Cover

“…A number of 3D inversion codes are available (Mackie et al 1994, Newman and Alumbaugh, 2000, Sasaki, 2004, Siripunvaraporn et al 2005a, and many model studies have been published (e.g. Siripunvaraporn et al 2005b, Han et al 2009).…”