Evaluating our Ability to Predict Future Nitrate Concentrations in Groundwater Based on a Few Key-Measurements

International audienceIn agricultural areas, predicting future nitrate concentrations in groundwater is essential to advise political decisions. However, beside a small minority of highly instrumented, monitored and modelled catchments, most of the sites suffer from a lack of available data. How can we, in these areas where little information is known, infer the evolution of nitrate pollution?This work uses a well-characterized site to identify, among all the natural and anthropic parameters (transit time distribution, reactivity rates, nitrate input chronicle), those that have the strongest impact on future nitrate concentrations. The study site is a 35 km2 agricultural catchment located in Brittany, Western France. We use a three-dimensional groundwater flow model calibrated with gaseous age tracer measurements [1] .The sensitivity of the predictions to each parameter is tested by gradually degrading the information contained in the catchment model. This approach shows that the first two moments of the transit time distribution, the oxygen apparent degradation rate as well as the total mass of nitrate that entered the aquifer are crucial to infer future nitrate trends in groundwater.We then evaluate the possibility of using in-stream measurements to acquire these key-parameters in poorly instrumented sites. This implies a quantification of in-stream groundwater discharge and a characterization of gaseous exchanges occurring between the stream and the atmosphere. Groundwater discharge is quantified using radon, while stream reaeration is studied by performing helium injection coupled to continuous in-situ measurements with membrane inlet mass spectrometry (MIMS).By identifying the key-parameters and developing a simple method to measure them, we aim to help predicting future nitrate trends in poorly instrumented areas.[1] Kolbe, Tamara & Marçais, J & Thomas, Zahra & Abbott, Benjamin & de Dreuzy, Jean-Raynald & Rousseau-Gueutin, Pauline & Aquilina, Luc & Labasque, T & Pinay, Gilles. (2016). Coupling 3D groundwater modeling with CFC-based age dating to classify local groundwater circulation in an unconfined crystalline aquifer. Journal of Hydrology. 10.1016/j.jhydrol.2016.05.020