The lack of agreement between magnetotelluric field measurements and the calculations based on essentially two‐dimensional models with either anisotropy or lateral inhomogeneity necessitates a more complex model of the earth than has been previously considered. The Galerkin finite‐element method is applied to a two‐dimensional structure with a tensor conductivity. The importance of considering conductivity as a tensor is illustrated by a model consisting of an anisotropic, conducting dike embedded in an anisotropic half‐space. This model can be distinguished from an isotropic model by the nonvanishing diagonal elements of the impedance tensor, the ellipticity indices, and the skew.

No abstract

Reddy, I. K., D. Rankin, 1971, Magnetotelluric Effect of Dipping Anisotropies, Geophysical Prospecting 19, 84–97. A solution is obtained for the magnetotelluric effect in the case of multi‐layered earth, each layer containing dipping anisotropy, with the strike of anisotropy oriented at an arbitrary angle to the measuring directions. A technique to compute tensor impedances and tensor apparent resistivities is presented and the importance of anisotropies in magnetotelluric interpretation is discussed.

Tensor apparent resistivity values obtained from the magnetotelluric field measurements, in the plains of Alberta indicate a major regional anisotropy. The direction of the major axis of the anisotropy is parallel to the strike of the Rocky Mountains and the sedimentary basin of Western Canada. Results obtained by numerical methods for a two-dimensional geoelectric model support an interpretation of the observed magnetotelluric results, depending on regional structures. The observed polarization parameters of the magnetotelluric fields are also in good agreement with those computed for the two-dimensional model. This study shows that the conclusions based on some earlier magnetotelluric results, which postulated an upper mantle conductivity layer must be treated with strong reservations.

A feasibility study is being made to monitor possible electrical resistivity changes preceding the earthquakes. Numerical calculations, based on dilatancy hypothesis, indicate that the resistivity changes as much as 60 percent can be possible at some of the Southern California sites. Preliminary magnetotelluric measurements in the eight month period indicate that changes in the resistivities greater than 10 percent can be monitored by magnetotelluric measurements. The observed scatter in the magnetotelluric measurements are within the random errors of the estimates, and no major earthquake activity is reported in the vicinity of these sites, during this period.