Time-lapse electrical surveys to locate infiltration zones in weathered hard rock tropical areas
Résumé
In West Africa, infiltration and groundwater recharge processes in hard rock areas are depending on climatic, surface and subsurface conditions, and are poorly documented. Part of the reason is that identification, location and monitoring of these processes is still a challenge. Here, we explore the potential for time-lapse electrical surveys to bring additional information on these processes for two different climate situations: a semi-arid Sahelian site (north of Burkina and a humid Sudanian site (north of Benin), respectively focusing on indirect (localized) and direct (diffuse) recharge processes. The methodology is based on surveys in dry season and rainy season on typical pond or gully using Electrical Resistivity Tomography (ERT) and frequency electromagnetic (FEM) apparent conductivity mapping. The results show that in the Sahelian zone an indirect recharge occurs as expected, but infiltration doesn't takes place at the center of the pond to the aquifer, but occurs laterally in the banks. In Sudanian zone, the ERT survey shows a direct recharge process as expected, but also a complicated behavior of groundwater dilution, as well as the role of hardpans for fast infiltration. These processes are ascertained by groundwater monitoring in adjacent observing wells. At last, FEM time lapse mapping is found to be difficult to quantitatively interpreted due to the non-uniqueness of the model, clearly evidenced comparing FEM result to auger holes monitoring. Finally, we found that time-lapse ERT can be an efficient way to track infiltration processes across ponds and gullies in both climatic conditions, the Sahelian setting providing results easier to interpret, due to significant resistivity contrasts between dry and rain seasons. Both methods can be used for efficient implementation of punctual sensors for complementary studies. However, FEM time-lapse mapping remains difficult to practice without external information that renders this method less attractive for quantitative interpretation purposes.