Building a regional hydrogeological model in a data-scarce area and implications for local scale studies: Case of the Tim Mersoï Basin (Northwestern Niger)

Abstract

This study investigates the feasibility of implementing a 3D hydrogeological model in a vast data-scarce area in northwestern Niger where groundwater resources are strongly impacted by climate changes and anthropogenic activities (uranium mining) in northwestern Niger, West Africa. A large scale fully integrated hydrological model was built and calibrated using HydroGeoSphere. The model conceptualization and parametrization were done using various sources of datasets. The model was calibrated using a stepwise approach by increasing the temporal resolution of the model inputs: steady state, dynamic equilibrium, and transient state. The calibration results showed that the model could reproduce the measured groundwater heads with a correlation coefficient of up to 0.79. The model was then validated by comparing total water storage computed from GRACE data over the TMB and the outputs of the model. The results showed the model reproduced quite well the amplitudes and direction of the variations measured by GRACE. Therefore, GRACE data were useful in the qualitative validation of the model. The outputs of the model allowed a quantitative analysis of the groundwater resources of the Tim Mersoï Basin (TMB). The global recharge was estimated at 15.74 mm/year, i.e., an interannual average of 13.58% of the rainfall. Most importantly, this model can be used in a hierarchical model set up to provide initial and boundary conditions for local scales (mining areas for example) studies.