Assessment of hydraulic conductivity distribution from ERT-monitored tracer tests avoiding the need for petrophysical relationships

Geophysical methods, such as the Electrical Resistivity Tomography (ERT), have been increasingly used in recent years to analyze the erratic behavior of plumes in natural aquifers. In particular, borehole ERT monitoring of saline tracer tests allows to collect 2D time-lapse electrical data in a control plane, related to spatio-temporal variations of salt concentration within the aquifer. The electrical conductivity (EC) field is reconstructed by means of a geophysical inversion on the basis of raw resistance data, while a petrophysical relationship (e.g., Archie’s law) is usually needed to map EC data into solute concentrations, thus retrieving the plume evolution. The latter, in turn, is often used to evaluate the hydraulic conductivity (K) distribution by inverse modeling. To avoid the need for an in-situ specific calibration of a petrophysical relationship and the previous knowledge of the concentration spatio-temporal evolution, this study proposes a new approach for retrieving the K distribution from an ERT monitored saline tracer test, based on travel time modeling of transport integrated with the ensemble Kalman filter (EnKF). The definition of the solute transport in terms of travel times allows to analyze the sequence of changes in electrical conductivity deduced from an ERT survey without converting the electrical data into concentrations. To do this, a specific travel time procedure is applied: the control plane is subdivided in properly spaced sub-control planes and the cumulative distribution function of the travel time in each of them is independently calculated and then assimilated through EnKF, which allows to update the K distribution. Our approach, initially tested in 3D synthetic aquifers, is here applied to the experimental site of Settolo, Valdobbiadene (TV), where a tracer test monitored by ERT has been carried out. The results show that the suggested method seems to be effective in reproducing the erratic distribution of the hydraulic conductivity at the local scale, which control the non-Fickian evolution of plumes in natural aquifers.