Five regional climate model simulations centered over the Iberian Peninsula (IP) have been analyzed in terms of their representation of extreme events. The selected models (PROMES, two versions of WRF, MM5 and REMO) constitute a sample of state-of-the-art regional climate models (RCMs). The climate complexity of the target domain poses a challenge for the models and an opportunity to compare their ability to simulate precipitation and temperature extremes. Unlike in previous similar studies (such as PRUDENCE and ENSEMBLES European projects), these present climate simulations (1990-2007) have a domain centered over IP with high horizontal resolution (25 km x 25 km) and are nested into ERA-Interim. Daily precipitation and maximum and minimum temperature data from RCMs were analyzed over 11 hydrological basins. A new high-resolution (0.2º x 0.2º) database (Spain02) of daily temperature and precipitation interpolated from surface observations over Spain was used to validate the results.
A common statistical analysis (precipitation distribution and temperature percentiles) and a set of precipitation and temperature extreme indices have been used to understand the spatial patterns of extremes over this region and to test the ability of the RCMs to reproduce these features. Some particular extreme events have also been analysed. In general, the models are able to simulate both precipitation and temperature climate extremes and particular extreme events. The results for the northern IP show the smallest di erences between simulated and observed precipitation and temperature indices. On the contrary, over the southeast basins these di erences are larger, probably due to the extreme climate of the region. Regarding the speci c extreme events, the drought su ered in 2000 is better simulated by all models than that in 1994 and the 2003 heat wave. Results di er depending on the selected basin, index and RCM. There is no model outperforming the rest in the representation of the climate extremes of the region. The ensemble mean improves the skill of any individual model.
Tweets by SantanderMeteo
