ADVANCES IN HYDROGEOPHYSICAL JOINT INVERSION

The inversion framework iTOUGH2 provides inverse modeling capabilities for TOUGH2, a general-purpose simulator for multiphase, multicomponent, nonisothermal flows in multidimensional fractured-porous media. We have developed a parallel version of iTOUGH2, MPiTOUGH2, which realizes a hierarchically parallel architecture using the Message Passing Interface. This architecture allows employing large numbers of parallel processes for running many-parameter inverse problems on large meshes, such as those occurring in pixel-based parameterizations. We have further combined the parallel hydrological inversion method with a geophysical modeling and inversion framework, called Electromagnetic Geological Mapper (EMGeo). The combined inversion method provides a tool for investigating coupled hydrogeophysical processes in the context of, for example, geologic CO2 storage, geothermal system characterization, and environmental remediation and monitoring applications. In this work, we focus on the introduction of parallel computing paradigms to iTOUGH2, and carry out basic scalability tests. Further, exploiting the forward simulation capabilities of EMGeo, we present synthetic data inversions, demonstrating the potential model resolution enhancement obtained through joint inversion of hydrological and geophysical measurements.