Extreme precipitation events: Comparative evaluation of high resolution regional climate models in European Alpine region

In this study we analyze the ability and sensitivity of four state of the art regional climate models (RCMs) in simulating extreme precipitation events. The results presented are based on analysis of a high resolution (10 km × 10 km) multi-model mixed-physics ensemble of 62 members made with four RCMs namely CCLM, REMO, MM5 and WRF. Various configurations by changing physical parameterizations, domain sizes, nudging as well as nesting methods are tested and its consequent impact on model performance is analyzed. The European Center for Medium-Range Weather Forecasts ERA-40 re-analysis dataset is used to provide the initial and boundary conditions for all simulations. Our analysis focuses on two heavy precipitation events that occurred in the year 1999 in the Alpine region. In general the RCMs have shown remarkable skill in simulating the extreme precipitation events. The model performance is measured on the basis of spatial correlation coefficient. It is found that RCMs are very sensitive to aforementioned changes. Depending on selected event and configurations the correlation coefficient, mean bias ranges between 0.08 and 0.89 and -2.58 mm/event to -24.07 mm/event respectively. The variability is systematically underestimated for most experiments. However, a careful selection of convection, microphysics and vertical resolution can help improve the underestimation of variability. Our results suggest that nesting methods, domain sizes and nudging techniques have dominant impact on model performance.