Effect of galvanic distortions on the series and parallel magnetotelluric impedances and comparison with other responses
Main Article Content
Abstract
The series and parallel impedances of the magnetotelluric tensor are appraised in relation to their relative immunity to galvanic electric distortions. The distorted responses are modeled using the Groom-Bailey decomposition of the tensor in terms of twist, shear, statics and strike direction. These four parameters and the undistorted responses are normally considered as unknowns, and are obtained from field data through the solution of an inverse problem. In the present work we use the decomposition as a forward model to simulate distorted sounding curves. Starting with undistorted 2-D TE and TM responses, the tensor is distorted by assuming arbitrary values of twist, shear, static and strike direction. By default, both series and parallel responses are immune to the strike direction because they are invariants under rotation. In addition, series responses are immune to twist and shear and parallel responses only to twist. The dependence of the latter on shear is in the form of a real factor that shifts downwards the amplitude curves. On the other hand, the effect of statics on both series and parallel responses is more complicated than that on the impedance tensor because it cannot be accounted for by a simple shift of the curves. On the whole, there is a positive balance on the part of the series and parallel impedances over the TE and TM responses because some of the distortions are filtered out by the invariants. It is shown that invariance is not sufficient to be immune to any of the distortions. The example chosen is Eggers' eigenvalues, which are immune only to the by-the-fault strike direction. Invariance is not necessary either, as evidenced by the phase tensor, whose elements depend on strike but are immune to all distortions. The derivations are illustrated using soundings from the synthetic COPROD2S1 and field-recorded COPROD2 and BC87 data sets.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.