Concept and background
Increasing CO2 levels cause an intensification of the global hydrological cycle, with an overall net increase in temperature, rainfall and runoff, and will increasingly do so. Rising CO2 levels are also likely to reduce evaporation and there is some evidence that recent increases in river flows globally are due to this effect Although the predictions of future rainfall are fairly uncertain, there are indications, for example, that the Mediterranean region will see reductions of rainfall while more temperate regions, such as Scandinavia, will see increases. The seasonality will also change, causing new, and sometimes unexpected, vulnerabilities.
Denmark has a leading position within hydrological and climate change research. Advanced numerical model codes have been developed including the climate model HIRHAM and the hydrological model MIKE SHE. Additionally, methodologies for comprehensive modelling have been developed. The two modelling systems have recently been used for quantification of the hydrological impacts of future climate change, also referred to as hydrological change.
However, the research on hydrological change is still in its infancy both with respect to model accuracy and uncertainty. Traditionally, the analysis of hydrological climate change impacts has been decoupled from climate research as such. Based on the output of global or regional climate models, hydrological models have been run as stand alone models. This means that the feedbacks to the atmosphere are neglected which has an unknown impact on the predictions of the climate change, particularly at the local scale. Furthermore there is an inherent contradiction in this approach since climate models include their own (very simplified) hydrological model component. For Denmark, the important processes of the groundwater flow and surface runoff through river networks and the interaction between the groundwater and river systems are not represented in current climate models (e.g. HIRHAM). So far, only the scientific framework for coupling of the two model types has been investigated whereas operational models on regional scale has not yet been developed.
Scaling between the climate models and the regional hydrological model, where the model domain and the grid size is reduced in steps, causes significant problems that have not been solved yet. Climate models operate at spatial and temporal scales that are much larger than the scales required to analyse the effects on the hydrological system. Data on climate change scenarios are available at spatial and temporal resolutions (typically 25 km grid and daily values) that are very coarse for direct application in hydrological modelling addressing local scale problems. Remote sensing data of relevance for hydrological studies such as global rainfall have the same problem as data from climate models, namely that they are available at large spatial and temporal scales and require downscaling for most practical application.
Regional hydrological models are increasingly being used as a basis for water resources management, both at national and European scale. The national DK hydrological model will for instance be made publicly available digitally via the internet in the near future. However, in order to make full use of such regional models to address local scale problems, and in this way ensure coherence in the local and national water resources management, a number of downscaling problems need to be solved. These include model technical issues related to grid refinement and issues concerning how to ensure consistency in geological conceptualisation when changing between different scales in complex geological settings.
Geological interpretations are recognised as maybe the primary source of uncertainty in hydrological modelling. No previous studies have evaluated the total uncertainty on hydrological change predictions including both uncertainties on greenhouse gas emission, climate model uncertainty and hydrological model uncertainty.
HYACINTS will develop new methodologies and tools that enable easier and more accurate use of regional scale climate and hydrological models to address local scale water resources problems.