Multi-physics joint inversion of field FWI-GPR and ER surface acquired data

We use ground penetrating radar (GPR) and electrical resistivity (ER) surface acquired data on an alluvial aquifer in a multi-physics joint inversion that recovers electrical permittivity and conductivity. Our algorithm is a simultaneous 2.5D joint inversion that coherently integrates the ER data and the full-waveform response of GPR without invoking petrophysical relationships. We join full-waveform inversion (FWI) GPR with adjoint derived ER sensitivities, without the need to interpolate the GPR and ER computational domains. We compare our recovered parameters to individual FWI-GPR and ER recovered parameters. We use capacitive conductivity and neutron-derived porosity borehole measurements to validate our results. Our joint recovered conductivity resembles both control data better than the control data resembles each other, and better than individual FWI-GPR and ER results. Our joint recovered permittivity exhibits subsurface features with higher resolution than that of the FWI-GPR solution.