Co-Operative Inversion of Magnetotelluric and Seismic Reflection Data from the Nevada Gold District

Abstract

Various geophysical methods are sensitive to different physical properties and have different resolution. Consequently, each geophysical dataset usually provides different subsurface images. Moreover, as inversion is non-unique; the inverted models from different methods can cause further ambiguity or even contradict in geological interpretation. Thus, it is more efficient to narrow the domain of solution by using complementary information by explicitly incorporating multiple data into the inversion algorithms/process. We use a challenging data set from a mineral exploration environment to demonstrate a novel method of integrating prior and complementary information from two geophysical methods into a co-operative inversion scheme. The proposed method exploits the advantages of the fuzzy c-means clustering technique to provide a common geostatistical model for inversion. Inclusion of borehole information provides confidence in choosing the number of clusters and defining centre values, which then improve both the MT and seismic inversion processes. Interpretation of the acoustic inversion resulting from our process allowed identification of three prospective target zones in the region of a deep borehole. These targets were verified as prospective by geochemical analysis with elevated levels of pathfinder elements and gold.