Precipitation Clustering in European Climate Models and Its Role in Hydrological Forecasting

Heavy precipitation clustering is an important factor for flood risk and hydrological forecasting, but its simulation in climate models is still uncertain. This study examines European climate simulations by comparing global climate models with long-term observations. Consecutive wet days (CWD) and very wet days (R95p) were used as indicators, and clustering behavior was assessed across major regions in Europe. The results show that models capture large-scale extremes with moderate accuracy but underestimate clustering by 20–30% in Mediterranean and Alpine areas. Sensitivity tests also show that clustering depends on temporal resolution, with daily indices showing stronger underestimation than multi-day metrics. These outcomes suggest that current models reproduce intensity more reliably than persistence, which lowers confidence in hydrological forecasts and flood risk analysis. The study concludes that improvements in physical schemes and the use of convection-permitting models are needed to better simulate precipitation clustering in Europe.