Abstract :
[en] Alluvial aquifers are generally composed of several facies with complex architectures and interconnections depending on the fluvial system. In this context, electrical resistivity tomography (ERT) may provide important information on the spatial distribution of hydrogeological parameters. However, ERT inversion introduces some bias in the resulting resistivity distribution due to regularization and resolution issues. In this study, we refine ERT inversions by incorporating prior information in order to improve the identification of facies through a probabilistic relationship derived from collocated measurements. We then analyze with synthetic cases the effect of spatially varying sensitivity on the probabilistic relationship. As expected, when sensitivity decreases, the distributions of resistivity for the different facies tend to be superimposed. A mean distribution thus overestimates the ability of surface ERT to discriminate hydrofacies in depth.
Scopus citations®
without self-citations
0