Initial Report on the Development of a Monte Carlo-Markov Chain Joint Inversion Approach for Geothermal Exploration

Geothermal exploration and subsequent characterization of potential resources typically employ a variety of geophysical, geologic and geochemical techniques. However, since the data collected by each technique provide information directly on only one or a very limited set of the many physical parameters that characterize a geothermal system, no single method can be used to describe the system in its entirety. Presently, the usual approach to analyzing disparate data streams for geothermal applications is to invert (or forward model) each data set separately and then combine or compare the resulting models, for the most part in a more or less ad hoc manner. However, while each inversion may yield a model that fits the individual data set, the models are usually inconsistent with each other to some degree. This reflects uncertainties arising from the inevitable fact that geophysical and other exploration data in general are to some extent noisy, incomplete, and of limited sensitivity and resolution, and so yield non-unique results. The purpose of the project described here is to integrate the different model constraints provided by disparate geophysical, geological and geochemical data in a rigorous and consistent manner by formal joint inversion. The objective is to improve the fidelity of exploration results and reservoir characterization, thus addressing the goal of the DOE Geothermal Program to improve success in exploration for economically viable resources by better defining drilling targets, reducing risk, and improving exploration/drilling success rates