HR: 1330h
AN: GP13B-04    [Abstracts]
TI: A Two-dimensional Version of the Niblett-Bostick Transformation for Magnetotelluric Interpretations
AU: * Esparza, F
EM: fesparz@cicese.mx
AF: CICESE/C. de la Tierra, Km. 107 Carr. Tij.-Eda., Ensenada, B.C 22830 Mexico
AB: An imaging technique for two-dimensional magnetotelluric interpretations is developed following the well known Niblett-Bostick transformation for one-dimensional profiles. The algorithm uses a Hopfield artificial neural network to process series and parallel magnetotelluric impedances along with their analytical influence functions. The adaptive, weighted average approximation preserves part of the nonlinearity of the original problem. No initial model in the usual sense is required for the recovery of a functional model. Rather, the built-in relationship between model and data considers automatically, all at the same time, many half spaces whose electrical conductivities vary according to the data. The use of series and parallel impedances, a self-contained pair of invariants of the impedance tensor, avoids the need to decide on best angles of rotation for TE and TM separations. Field data from a given profile can thus be fed directly into the algorithm without much processing. The solutions offered by the Hopfield neural network correspond to spatial averages computed through rectangular windows that can be chosen at will. Applications of the algorithm to simple synthetic models and to the COPROD2 data set illustrate the performance of the approximation.
DE: 1515 Geomagnetic induction
SC: Geomagnetism and Paleomagnetism [GP]
MN: 2005 Joint Assembly