Modelling pesticide transport in a shallow groundwater catchment using tritium and helium-3 data

Using tritium and helium-3 data for calibration, a 2-D transport model was set up to explain the occurrence of bentazone, dichlorprop, glyphosate, isoproturon, MCPA and metamitron in a small groundwater catchment in southern Sweden. The model was parameterised with site-specific degradation and sorption data to enable transport simulations. Local climatological data and a 21-year record of agricultural pesticide use within the study area were used as boundary conditions. Model output was evaluated against a 7-year long pesticide monitoring data-series from two monitoring wells within the study area. The model successfully predicts observed breakthrough of bentazone, dichlorprop, isoproturon and MCPA. However, it fails to simulate observed occurrences of glyphosate and metamitron. Glyphosate and metamitron exhibit relatively high sorption potential, and their occurrence is suggested to be the result of non-equilibrium preferential flow paths which the model cannot reproduce due the conceptualisation of the system as homogenous and isotropic. The results indicate a promising methodological approach applicable to groundwater contamination risk assessment, and demonstrate the potential for transport model calibration by means of tritium and helium-3 data. Main constraints of the study relate to the relatively simple system conceptualisation, indicating a need for further consideration of physical and chemical heterogeneity.