Impact of future climate changes on hydrology, N-reduction and N-load in a Danish groundwater-dominated catchment

Anne Lausten Hansen1, Christen Duus Børgesen2, Jørgen E. Olesen2 and Jens Christian Refsgaard1

1Department of Hydrology, Geological Survey of Denmark and Greenland, Øster Voldgade 10, 1350 Copenhagen K, Denmark

2Department of Agroecology, Aarhus University, Blichers Allé 20, 8830 Tjele, Denmark

Denmark must further decrease the N-load to coastal waters from agriculture to comply with the Baltic Sea Action Plan and the EU Water Framework Directive. However, the expected future climate changes will affect groundwater and thereby also the transport and reduction of nitrate in a catchment, and the needed N abatements may therefore change with time.

In this study we have analysed how the expected climate changes for the period 2040-2060 will possibly affect the hydrology, N-reduction and N-load in the groundwater-dominated Norsminde catchment in Denmark. We have used an ensemble of four climate models (CM1-4) selected to encapsulate the uncertainty of climate model projections. The projected future climates were downscaled and bias-corrected, and then used as input to the N-leaching model NLES and the groundwater-surface water model MIKE SHE.

The results show that three of the climate models (CM1-3) give an increase in average yearly precipitation in Norsminde catchment of 5% (CM1), 19% (CM2) and 17% (CM3). One climate model (CM4) gives a slight decrease in yearly precipitation of -2%. The monthly distribution of precipitation over the year is found to differ between the four climate models, with CM2 having the largest precipiatation amounts in the late summer/early fall and again in late winter and CM1 in the fall. The future evapotranspiration changes only slightly in future for all four climate models (-1% to 5%). The drain flow component is found to increase for all three climates models (CM1-3) with increased precipitation. However, the drain flow is found to increase more than the increase in precipitation for CM2 and CM3. This could indicate that these two climate models result in more shallow flow paths.

All four climate models project an increase in both N-leaching from root zone and in total N-load at the catchment outlet. But the changes in N-load do not follow the changes in N-leaching in a linear way for all climate models. CM1 projects the highest increase in N-leaching of 61% and CM4 the lowest with 23%. The increase in N-load is 60% and 20% for the two climate models  respectively, and thereby similar to the change in N-leaching. But for CM2 and CM3 the increase in N-leaching is 41% and 26% respectively, whereas the increase in N-load is 55% and 38% and thereby relatively higher than the increae in N-leaching. The reason for this difference in reponse in N-load must be found in the amount of N-reduction in groundwater (GW%). For CM1 and CM4 the change in N-reduction is small or unchanged (-7% and 0% respectively), whereas for CM2 and CM3 the N-reduction decreases with 13%. This lower N-recution in groundwater for CM2 and CM3 is probably caused by the higher drain flow.

Based on the findings from this study we conclude, that the N-load in Norsminde catchment is likely to increase in the future and that stricter abatement targets therefore will be needed. But there is a considerable uncertainty on how much the N-load will increase.