Monitoring freshwater salinization in analog transport models by time-lapse electrical resistivity tomography

Abstract Deep saline aquifers are target formations both for the geological storage of carbon dioxide as well as for geothermal applications. High pressure gradients, resulting from fluid or gas injection processes, provide a potential driving force for the displacement of native formation waters, implicating a potential salinization of shallow freshwater resources. Geoelectrical monitoring techniques are sensitive to compositional changes of groundwater resources, and hence capable to detect salinization processes at an early stage. In this context, numerical simulations and analog modeling can provide a valuable contribution by identifying probable salinization scenarios, and thereby guiding an optimum sensor network layout within the scope of an early warning system. In this study, coupled numerical flow and transport simulations of a laterally uniform salinization scenario were carried out and used to support a subsequent realization in a laboratory sandbox model. During the experiment, electrical resistivity tomography (ERT) was applied in a practical surface–borehole setup in order to determine the spatio-temporal variations of electrical properties influenced by saltwater intrusion. Inversion results of different electrode configurations were evaluated and compared to numerical simulations. With regard to surface–borehole measurements, good results were obtained using crossed bipoles, while regular bipole measurements were more susceptible to noise. Within the scope of a single-hole tomography, the underlying resistivity distribution was best reproduced using the Wenner configuration, which was substantiated by synthetic modeling.