Présentations Colloques

Oral Presentation
8.10
Session 8.10: Recent developments in groundwater modeling and mathematical tools in Hydrogeology
Lachassagne Patrick
Reverse modeling - a simple and robust method for modeling and forecasting the piezometric level discharge of a pumping well. A case study in a Hard Rock Aquifer
Forecasting the piezometric variations at a well as regards both its discharge, and the characteristics of the pumped aquifer, including its replenishment, is of high practical interest. Such a medium term approach is not straightforward, particularly in complex small size aquifers (a few square kilometers) as in hard rocks. We thus present a simple but original approach that uses the time series of discharge piezometric level (D PL) available at the well, and the piezometric level available at the “present date”. This data set is used to “reversely” compute the piezometric level from the present date towards the beginning of the time series. The parameters of the model are calibrated to accurately reproduce the piezometric chronicle. Then, as with any model, the calibrated model is used to forecast the future. This approach was developed in a weathered-fractured metamorphic aquifer pumped by a unique well. The diagnosis of the early months of the D PL chronicle allowed to characterize its hydrodynamic parameters (T, S), but also the distance to the 4 orthogonal impervious boundaries limiting the aquifer. The model then computes, with a MSExcel sheet the water budget of this aquifer with the efficient porosity and the well losses revealed by the pumping test. A recharge flux was introduced to ensure the calibration. No vertical variation of the specific storage with depth was necessary, nor variation of the recharge rate with the drawdown. The forecasting was then performed on the basis of the four seasonal recharge fluxes defined during the calibration. This approach is very efficient and cost effective, particularly in such a context where water budgeting of the aquifer finally appears to be quite easy. The model is not accurate to compute short term discharge variations but its improvement would complexify it without a great added value for longer term computations.
France