A Novel Structural Geostatistical Constrained ERT Approach for Hydrogeophysical Characterization in Glacial Sedimentary settings

Effective groundwater exploration is fundamental to the identification and assessment of subsurface water resources, particularly in challenging geological conditions. In such cases, conventional drilling-based approaches are costly and provide only limited one-dimensional information for characterizing the three-dimensional subsurface. Hydrogeophysical techniques offer efficient, non-invasive means of imaging the subsurface and can significantly reduce the need for extensive drilling campaigns. Among these, Electrical Resistivity Tomography (ERT) is one of the most widely used geophysical methods for investigating groundwater systems. However, conventional smoothness-constrained ERT inversion often fails to resolve sharp stratigraphic boundaries or represent internal heterogeneity, limiting its effectiveness in complex geological settings. Structural and geostatistical constraints have each been proposed as enhancements, but when applied separately, they often struggle to simultaneously achieve structural accuracy and inversion stability.

We introduce a novel inversion strategy that integrates seismic-derived structural horizons with unit-specific geostatistical constraints within the open-source PyGIMLi framework. This approach balances structural alignment and internal heterogeneity by enforcing sharp boundaries and applying region-specific spatial continuity models.

This hybrid strategy is evaluated at two glacially influenced groundwater study sites in northern Germany, where complex Quaternary deposits include interbedded sands, tills, and clays. Results demonstrate enhanced delineation of aquifers and aquitards, improved agreement with borehole resistivity logs, and a reduction in inversion artifacts such as over-smoothing or artificial layering. Compared to conventional and single-constraint inversions, the integrated method more effectively resolves thin confining units, anthropogenic disturbances, and laterally variable aquifer geometries with enhanced structural clarity.

This framework offers a transferable solution for hydrogeophysical characterization in heterogeneous environments, particularly where seismic or borehole data are available to guide inversion.