Quantifying projected effects of climate change on low-flows in the Mur River, Austria: Adapting weather generator outputs to reflect downscaled EURO-CORDEX datasets

In the context of ongoing climate change, the use of climate model predictions plays a crucial role in preparing adaptation strategies that could reflect the expected changes in low-flow conditions within water systems. These predictions are often non-stationary and thus not very suitable for a comprehensive statistical analysis of extremes, mainly due to very short periods of quasi-stationary conditions. In this study, we address these problems by adapting a 10,000-year long dataset of synthetic meteorological outputs, representing the current climate conditions and produced by a large multi-site stochastic weather generator, to account for both spatial and temporal variability of the changing climate through a delta-change approach. We adapted this dataset to depict the mean climate conditions of the final three decades of the 21st century, as defined by 26 different scenarios (13 GCM-RCM combinations for 2 different RCPs) incorporated into the high-resolution Austrian National Climate dataset (ÖKS15) derived from the EURO-CORDEX data. We employed an HBV-distributed hydrological model to simulate synthetic river discharges for both the current and 26 future climate conditions in the Mur River, Austria. The subsequent analysis of the low-flows during the vegetation period was focused on the changes in low-flow frequencies and the number, duration and magnitude of dry periods. The results indicate that the adopted delta-change approach reflects the changes in rainfall amounts but does not address the changes in rainfall frequencies. Future research should, therefore, prioritize the development of methods that can accommodate qualitative changes in rainfall distribution.