Long-term transfer of diffuse pollution at catchment scale: Respective roles of soil, and the unsaturated and saturated zones (Brévilles, France)
Résumé
Environmental management such as recommended by the European Water Framework Directive requires actions regarding diffuse pollution. However, the efficiency of these actions depends on the reaction time of the hydrosystem. In order to predict future groundwater-quality trends, especially after implementation of environmental measures such as substitution of pesticides or reduction of nitrogen loads, the response time has to be determined. In the case of diffuse pollution, the response time between fertilizer or pesticide application and the subsequent contamination of a spring or well may depend on the distance between the area where the products are spread and the point where water quality is measured. However, some aquifer systems may appear particularly slow, independently of the size of the catchment. At Montreuil-sur-Epte (Val d'Oise, France), no significant change in groundwater quality can be related to the suppression of atrazine application over the small agricultural catchment (4 sq km) of the Brévilles spring, 8 years ago (4 years before the national ban of atrazine). Besides the 8 year monitoring of groundwater quality at the spring and in several observation wells in its catchment, a particular attention was paid to the characterization of water and solute-infiltration rates at different scales. The role of each compartment (soil, and unsaturated and saturated zones) in the slowing down of flow and the transfer of solutes (pesticides and nitrate) from top soil to aquifer, and then to the spring, was determined by a multidisciplinary approach involving mainly soil science, hydrogeology and geochemistry. Detailed investigations aimed at identifying and quantifying the parameters that contribute to the inertia of the system, resulting in a long-term transfer of diffuse pollution. Data covering the monitoring of pesticide transfer in soil, the physico-chemical characteristics of soils, profiles of water contents, tritium activities, and nitrate concentrations measured on rock samples collected by drilling, were compared in order to describe the transfer time in soil and within several meters of unsaturated zone. Tracer tests in groundwater completed the understanding of transport in the saturated zone near the spring. Geochemical measurements and pumping tests confirmed the stratification of the saturated sandy aquifer leading to heterogeneous groundwater contamination. Finally, the inertia of the system results from several factors linked to the hydraulic properties of each underground compartment combined to intrinsic properties of the contaminants, partially explaining the limited improvement of groundwater quality 8 years after a major change in agricultural practice. The observations and results obtained by applying various tools and methods in this catchment led to a reflexion on sampling frequency, methodology and metrology to be adapted to this particular, but not unique, hydrogeological context in order to assess the groundwater-quality evolution and to understand the fate of solutes. The results of this study also raise questions on how efficient and how fast will the positive impact of product substitution or environmental regulations be.