Abstract
Separate inversions of the North American Central Plains (NACP) and Thompson Belt (TOBE) anomalies are undertaken using a method in which a 'least-blocked' model is sought. The algorithm first constructs a 3-column starting model by the juxtaposition of 1D 'least-layered' inversions of the TE data at 3 sites covering the region under investigation and then minimizes the least squares fit of the TE and TM apparent resistivity and phase responses by adjusting the positions of the horizontal and vertical boundaries and the resistivities of the resulting blocks in the composite starting model. The procedure is repeated with successively 4, 5, ... columns until the addition of further columns is no longer justified by a significant improvement in the fit. Inversion of the COPROD2 data by this method reveals a conductive surface layer of sediments slightly less than 3 km thick across the whole region, underlain by an NACP anomaly roughly 120 km wide and composed of two adjoining conductive blocks extending to the basement 60 km deep and whose top surfaces are at depths of roughly 12 km and 19 km respectively, and a TOBE anomaly consisting of a single narrow block only about 7 km wide rising from the basement to the base of the surface sedimentary layer. The TOBE anomaly is found to be much more conductive than the NACP anomaly.
