Precipitation and temperature statistics in high‐resolution regional climate models: Evaluation for the European Alps

Abstract

In this study, high‐resolution climate change data from the regional climate models COSMO‐CLM, HIRHAM, RegCM, and REMO were evaluated in the Greater Alpine Region (GAR; 4°W–19°W and 43°N–49°N) and three additional subareas of 1.5° by 1° in size. Evaluation statistics include mean temperature and precipitation, frequency of days with precipitation over 1 mm and over 15 mm, 90% quantile of the frequency distribution, and maximum number of consecutive dry days. The evaluation for the 1961–1990 period indicates that the models reproduce spatial precipitation patterns and the annual cycle. The mean precipitation domain bias varies between 11% and 40% in winter season and between −14.5% and 11% in summer. Larger errors are found for other statistics and in the various regions. No single best model could be identified comparing modeled precipitation characteristics with observational reference. The study shows that there is still high uncertainty in the expected climate change. Furthermore, future temperature and precipitation changes simulated with different SRES scenarios and calculated by different RCMs overlap. The temperature calculations for the period 2071–2100 related to the period 1961–1990 in the GAR area show an increase in the monthly mean 2m temperature of up to 4.8 K in summer. In the GAR area, a precipitation decrease of up to 29% in summer and precipitation increase of approximately 20% in the winter season is simulated. Summer and autumn temperatures are expected to increase more than winter and spring temperatures. Detailed analysis reveals that the different regional climate model runs based on different regional models, different driving global models and different emission scenarios show similar trends, but differ in the magnitude of the expected climate change signal. All models seem to agree on the increased frequency of high‐precipitation events in the winter season.