Three-dimensional galvanic distortion of three-dimensional regional conductivity structures: Comment on “Three-dimensional joint inversion for magnetotelluric resistivity and static shift distributions in complex media” by Yutaka Sasaki and Max A. Meju

Abstract

[1] Distortion of natural time variations of the electric and magnetic fields induced by external magnetic sources by local, near-surface inhomogeneities has been the bane of the magnetotelluric method (MT). These distortions are caused by galvanic charges on conductivity gradients at the boundaries of near-surface inhomogeneities that are below the resolution of the MT experiment. They often dominate over the inductive response of deeper structures, leading to erroneous interpretations if not appropriately considered. Indeed, one can argue that MT has become a generally robust and useful geological mapping tool only with the advent of distortion recognition, appraisal and removal methods. On land, these distortions are usually low in regions of laterally uniform surficial layers, such as sedimentary basin environments, and extreme in highly heterogeneous resistive environments, such as on Precambrian regions. On the seafloor they are usually extreme. They are always present and affect the MT responses derived from the observed fields to a greater or lesser extent, and must be considered in any analyses and subsequent interpretations. [2] Distortion methods for MT data responding to onedimensional (1-D) and two-dimensional (2-D) structures (reviewed below) have been advanced and are usually quite effective, except in the situations of very severe distortion. For data from three-dimensional (3-D) structures however, the problems caused by these galvanic charges becomes far more complex and require innovative analysis and treatment. [3] In their paper, Sasaki and Meju [2006] describe their approach to 3-D inversion of MT data for 3-D structure taking galvanic distortion, which they characterize and simplify as static shifts, into account. Their paper is an update on prior work by Sasaki [2004] with the same approach. The assumption made by the authors is that the magnitudes of the observed off-diagonal elements of the impedance tensor can be described as geometrically shifted versions of their true values, namely in their terms