Reconstruction of 1-D conductivity from dual-loop EM data

Inversions of electromagnetic data from different coil configurations provi de independent information about geological structures. We develop a 1-D in version algorithm that can invert data from the horizontal coplanar (HC), v ertical coplanar, coaxial (CA), and perpendicular coil configurations separ ately or jointly. The inverse problem is solved by minimizing a model objec tive function subject to data constraints. Tests using synthetic data from 1-D models indicate that if data are collected at a sufficient number of fr equencies, then the recovered models from individual inversions of differen t coil systems can be quite similar. However, if only a limited number of f requencies are available, then joint inversion of data from different coils produces a better model than the individual inversions. Tests on 3-D synth etic data sets indicate that 1-D inversions can be used as a fast and appro ximate tool to locate anomalies in the subsurface. Also for the test exampl e presented here, the joint inversion of HC and CA data over a 3-D conducti vity provided a better model than that produced by the individual inversion of the data sets.