Least squares inversion of one-dimensional magnetotelluric data: An assessment of procedures employed by Brown University |
| |
| Authors: | Jens Pedersen John F Hermance |
| |
| Institution: | (1) Geophysical/Electromagnetics Research Laboratory, Department of Geological Sciences, Brown University, 02912 Providence, RI, USA;(2) Present address: Union Geothermal Division, Union Oil Company of California, 2099 Range Avenue, P.O. Box 6854, 95406 Santa Rosa, CA, USA |
| |
| Abstract: | The pirmary objective of this paper is to review inversion algorithms employed to interpret magnetotelluric data by the geophysical/electromagnetics group at Brown University. Examples are presented from a variety of tectonic regimes — the Rio Grande rift, the Colorado Plateau, and Iceland. Following a brief tutorial in which the basic relations for the inverse prrolem are developed, we discuss the detectability and resolvability of various elements of the models. The correlation between certain model parameters is stressed, a phenomena well-known to interpreters. In this paper we emphasize the importance of identifying the precise range of values for which two or more parameters are coupled — these ranges are defined by what wer term  saturation limits . Generalized inverse theory is employed to arrive at optimal models which are evaluated in terms of resolution and information matrices. The degree to which models fit a given dat base can be assessed using several quantitative approaches. We have found that it is impotant not only to determine the best fitting model, but also to identify those marginal solutions which also fall within the range of uncertainty of the data. The results from our direct invverse method agree favorably with those from Monte Carlo methods. |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
