| Title | Bayesian Data Fusion for Geothermal Exploration: New Algorithms and Results |
|---|---|
| Authors | Simon O\'CALLAGHAN, Alistair REID, Lachlan McCALMAN, Olena VELYCHKO, Simon CARTER, Lars KRIEGER, Michelle SALMON, Graeme BEARDSMORE, Edwin BONILLA, Malcolm SAMBRIDGE, Tim RAWLING, Fabio RAMOS |
| Year | 2013 |
| Conference | Australian Geothermal Energy Conference |
| Keywords | joint inversion, data fusion, Bayesian methods, exploration, risk minimisation |
| Abstract | This project addresses the central question of identifying the best locations for drilling in geothermal target exploration from a probabilistic perspective. We develop algorithms to perform data fusion and joint inversions from a number of geophysical surveys to infer several rock properties such as density, magnetic susceptibility, seismic reflection, resistivity and thermal conductivity. Our methods can seamlessly integrate surveys including gravity, magnetics, seismic reflection, magnetotellurics and borehole geophysics to predict the most relevant properties for geothermal resource categorisation. As part of a probabilistic inference procedure, we also estimate uncertainties for each quantity of interest, providing principled strategies for risk minimisation in drilling projects. This paper presents a number of updates to the project including modelling and inference using parallel tempering Monte Carlo techniques. The probabilistic approach to the inversion problem not only quantifies the risk associated with drilling, but will be able to assist in strategies for further sensing that optimally increase exploration confidence. The Bayesian formulation allows expert knowledge to be added in the form of prior distributions, making the contribution of human judgment to the final result explicit. |