Joint Inversion of Seismic and Geoelectric Sounding Using Genetic Algorithm for Geothermal Prospect Identification

Abstract

Combination of various geophysical data within a single inversion framework improves model resolution of subsurface mapping. Genetic algorithm is a global optimization method that mimic Darwinian evolution is well suited for nonlinear geophysical inversion problems. In this paper we have implemented a new approach of genetic algorithm for integration of seismic and geoelectric data for geothermal subsurface mapping. Ray inversion for Near Surface Estimation method is used for inversion of seismic refraction data. In genetic algorithm a best of fittest models from a population is selected and then applied to operators as crossover and mutation to combine the most successful characteristics of each model. Genetic algorithm is applied to seismic and geoelectric data of Unai region to identify geothermal potential of the region. Results of inversion in this region suggests genetic algorithm is effective in solving problems with reasonably large number of free parameters along with the computation of objective function calculations.