Hip — Climate change

The maps provide a nationwide overview of the expected climate-produced median change in depth to ground water in the periods 2041-2070 and 2071-2100 in a resolution of 100x100 metres, based on a downscaling of the model calculations in 500x500 m (except Bornholm, where model calculations are available in 100x100m). The climate projected changes are available for the emission scenarios RCP 4.5 and RCP 8.5. The downscaled maps can be used to screen the magnitude of a change expected in depth to ground water and for any given statistical sizes in the future. Although this product in 100x100 m does not directly show results of a hydrological model like the equivalent product in 500x500m, but is a result of a machine learning downscaling algorithm, the method is suitable for showing more local details in the change of depth to the groundwater level. This is confirmed by the validation of the downscaled 100x100 m maps and is due to the fact that the change in the groundwater level, among other things, is controlled by variations in topography and/or hydraulic conductivity, which are available in high resolution of at least 100 m (i.e. in a finer resolution than the 500x500 m model). It is ensured that the general patterns in downscaled products are in good accordance with the results of the 500x500 m model, as the downscaling algorithm always uses the corresponding 500x500 m model results as input. When using the downscaled climate projected changes, it is recommended to focus on the uncertainties available for similar climate projected changes in 500x500 metres (standard deviation across climate models). The actual downscaling of the Random Forest algorithm adds additional uncertainty, however, as mentioned, the validation shows that the level of detail increases in the downscaling. Modelling can be used when making climate-resilient and long-term solutions that take into account the expected changes in depth to the ground water in a given area due to climate change. There are no model calculations for Læsø, Endelave, Tunø, Samsø, Anholt, Æbleø, Vejrø and Christiansø.