engleski

Near-surface wind climatology over Adriatic and Pannonian regions in an ensemble of RCM simulations

This study explores observed and simulated near- surface (i.e. 10 m) wind over the eastern Adriatic and Pannonian regions. The eastern Adriatic coast is represented by the complex coastline, steep topographic gradients and particular wind regime. The Dinaric Alps separate narrow coastal region from the inland, extending from northwest to southeast. The most famous typical winds along the Adriatic coast are bora (usually blowing in the direction perpendicular to the mountain and experiencing a strong influence of the terrain) and sirocco (usually parallel to the coastline and mostly during the wintertime) and sea/land breezes (dominantly in the warm part of the year) as a part of the regional Mediterranean wind system. In contrast, the Danube river splits Pannonian Basin into two parts bordered by the Alps in the northwest, the Dinaric Alps in the southwest and by the Carpathians in northeastern direction. South of the Danube, which includes the Croatian part of the Pannonian Plain, sporadic mountains have "insular" character with altitudes up to 1000 m above sea level contributing to the area complexity. These particular areas represent a significant challenge for the evaluation of the modeled wind flow and pattern simulated by the latest generation of the regional climate models (RCMs). Surface station observation data from 1996-2008 period have been used due to the length of available observed time series and compared against the daily output obtained from a suite of six RCM simulations from the EURO- CORDEX initiative (CLMcom-CCLM4-8-17, DMI- HIRHAM5, IPSL- INERIS-WRF331F, KNMI-RACMO22E, SMHI-RCA4, DHMZ- RegCM4). All simulations are forced by the ECMWF ERA-Interim reanalysis. Several methodological aspects related to the interpolation techniques when comparing RCMs and observations (or RCMs at two different resolutions) are examined. The evaluation reveals strong sensitivity of the simulated wind flow and wind pattern to the RCM horizontal resolution (12.5 km vs. 50 km). Additionally, different (non)dimensional skill measures discussed (e.g. bias, Brier skill score, Perkins skill score) depend on both seasons and locations analysed in this study. Moreover, RCMs are explored in terms of skill in reproducing specific wind regimes (e.g. bora) where large spread in the RCM ensemble is found.

ensemble; climatology; regional climate models

nije evidentirano