Hydrodynamic parameters estimation from self-potential data in a controlled full scale site

- A multi-physical approach has been developed for real porous media characterization.
- Hydrogeophysical experiments have been performed in a controlled full-scale facility.
- Hydraulic head and self-potential variations have been acquired during a pumping test.
- The collected data have been used into the coupled Richards-Poisson inversion model.
- The obtained results remove all the approximations adopted in previous works.

A multi-physical approach developed for the hydrodynamic characterization of porous media using hydrogeophysical information is presented. Several pumping tests were performed in the Hydrogeosite Laboratory, a controlled full-scale site designed and constructed at the CNR-IMAA (Consiglio Nazionale delle Ricerche - Istituto di Metodologia per l'Analisi Ambientale), in Marsico Nuovo (Basilicata Region, Southern Italy), in order to obtain an intermediate stage between laboratory experiments and field survey. The facility consists of a pool, used to study water infiltration processes, to simulate the space and time dynamics of subsurface contamination phenomena, to improve and to find new relationship between geophysical and hydrogeological parameters, to test and to calibrate new geophysical techniques and instruments. Therefore, the Hydrogeosite Laboratory has the advantage of carrying out controlled experiments, like in a flow cell or sandbox, but at field comparable scale. The data collected during the experiments have been used to estimate the saturated hydraulic conductivity ks [ms−1] using a coupled inversion model working in transient conditions, made up of the modified Richards equation describing the water flow in a variably saturated porous medium and the Poisson equation providing the self-potential ϕ [V], which naturally occurs at points of the soil surface owing to the presence of an electric field produced by the motion of underground electrolytic fluids through porous systems. The result obtained by this multi-physical numerical approach, which removes all the approximations adopted in previous works, makes a useful instrument for real heterogeneous aquifer characterization and for predictive analysis of its behavior.