Application of the Netherlands Groundwater Model, LGM, for calculating concentration of nitrate and pesticides at abstraction wells in sandy soil areas of the Netherlands

In a study aimed at assessing the impact of historical and future solute leaching into saturated groundwater, the quasi-three-dimensional RIVM groundwater model, LGM (version 2), was used for calculating pathlines, travel times and concentration breakthrough curves at 165 groundwater abstraction locations in the sandy soil areas of the Netherlands. The future concentration variations were assessed for nitrate and three pesticides (atrazine, bentazone and 1,2-dichloropropane). This study was carried out in the framework of the Dutch government's long-term planning for industrial and drinking-water supply. The assessment was carried out for 76 phreatic and 89 semi-confined abstraction sites on the basis of the 1988 maximum-permitted abstraction rates. Three economic scenarios were used for generating solute leaching into saturated groundwater for the scenario years 2000 and 2020. The nitrate leaching flux was simulated by a combination of methods, including the NLOAD model for agricultural land. The pesticide leaching flux was simulated by means of the GEOPESTRAS model. Subsequently, LGM was applied to calculate the concentrations at the abstraction wells, using the concentrations in the leaching flux as input. LGM is a numerical model (based on the finite-element method), comprising complex geohydrological system components. LGM uses spatially variable (heterogeneous) data for four aquifers, covering the entire surface area of the Netherlands. The concentration breakthrough curves were calculated with the module LGMCAM, based on forward pathline tracking in saturated groundwater. The processes used in LGMCAM were advection and full mixing in well screens. Neither denitrification nor degradation of pesticides in saturated groundwater were taken into account. The results are presented (1) as concentration maps for the 165 abstraction sites for 2020 and 2050 and (2) by means of time-based concentration bar charts (1950-2050) for the totalized abstraction rate for phreatic and semi-confined abstraction sites. For a number of abstraction sites a comparison has also been made between calculated and observed breakthrough concentrations. The calculated concentrations can be concluded to be a worst-case outcome. Finally, recommendations for future improvements of the method are given