Impacts of land use, climate change and hydrological model structure on nitrate fluxes: Magnitudes and uncertainties.

Nitrate pollution and eutrophication are of increasing concern in agriculturally dominated regions, and with projected future climate changes, these issues are expected to worsen for both surface and groundwater. Changes in land use and management have the potential to mitigate some of these concerns. However, to what extent these changes will interact is unknown, and are associated with significant uncertainty. Here, we estimate nitrate fluxes and contributions of major uncertainty sources (variance decomposition analysis) affecting nitrate leaching from the root zone and river load from groundwater sources for an agricultural catchment in Denmark under future changes (2080-2099) in climate (four climate models) and land use (four land use scenarios). To investigate the uncertainty from impact model choice, two different agro-hydrological models (SWAT and DAISY-MIKE SHE) both traditionally used for nitrate impact assessments are used for projecting these effects. On average, nitrate leaching from the root zone increased by 55%-123% due to different climate models, while the impact of land use scenarios showed changes between -9% and 88%, with similar projections for river loads, while the worst-case combination of the three factors yielded a fivefold increase in nitrate transport. Thus, in the future, major land use changes will be necessary to mitigate nitrate pollution likely in combination with other measures such as advanced management and farming technologies and differentiated regulation. The two agro-hydrological models showed substantially different reaction patterns and magnitude of nitrate fluxes, and while the largest uncertainty source was the land use scenarios for both models, DAISY-MIKE SHE was to a higher degree affected by climate model choice. The dominating uncertainty source was found to be the agro-hydrological model; however, both uncertainties related to land use scenario and climate model were important, thus highlighting the need to include all influential factors in future nitrate flux impact studies.