Quantitative estimates of hydrological state variables using electrical or electromagnetic geophysical methods are systematically biased by overlooked heterogeneity below the spatial scale resolved by the method. We generalize the high-salinity asymptotic limit of electrical conduction in porous media at the continuous (e.g., Darcy) scale, by introducing a new petrophysical parameter, the mixing factor M, which accounts for the effect of fluid conductivity heterogeneity on the equivalent electrical conductivity tensor; it is expressed in terms of the volume-average of the product of mean-removed fluid conductivity and electric fields. We investigate the behavior of M for static and evolving fluid conductivity scenarios. Considering 2-D ergodic log-normal random fields of fluid conductivity, we demonstrate, in absence of surface conductivity, that observing the components of the M-tensor allows univocally determining the variance and anisotropy of the field. Further, time-series of the M-tensor under diffusion-limited mixing allows distinguishing between different characteristic temporal scales of diffusion, which are directly related to the initial integral scales of the salinity field. Under advective-diffusive transport and for a pulse injection, the time-series of M have a strong dependence on the Péclet number. Since M is defined in the absence of surface conductivity, we investigate how to correct measurements for surface conductivity effects. The parameter M provides conceptual understanding about the impact of saline heterogeneity on electrical measurements. Further work will investigate how it can be incorporated into hydrogeophysical inverse formulations and interpretative frameworks.

Open Access funding provided by University of Lausanne. This work has received funding from the Swiss National Science Foundation under Grant 200020-184574. We are grateful for the insightful comments provided by Associate Editor Lee Slater, Olaf Cirpka and one anonymous reviewer. We thank Editor Xavier Sánchez Vila for handling the review process.

Peer reviewed