Comparative study of the COSMO and ICON models performances over Italy: Key insights from the impactful year 2023

作者：Leo Luca Loprieno, Mario Raffa, Angelo Campanale, Marianna Adinolfi, Paola Mercogliano · 发表于：Atmospheric Research · 年份：2025 · DOI：10.1016/j.atmosres.2025.108339 · 被引用次数：2 · 研究领域：Climate variability and models、Meteorological Phenomena and Simulations、Atmospheric Ozone and Climate

This study presents a comparative analysis of the regional climate models COSMO-CLM (CCLM) and ICON-CLM (ICLM) over Italy, focusing on the year 2023. Given the widespread use of the high-resolution dataset VHR-REA_IT, produced by downscaling ERA5 reanalysis with CCLM at ~2.2 km over Italy, and the transition from CCLM to ICLM within the scientific communities (as the Consortium for Small-Scale Modelling COSMO and Climate Limited-area Modelling CLM), this work aims to assess the added value of using ICLM at convection-permitting scale over Italy during the recent impactful year 2023. Evaluation results and objective metrics, highlight a reduction in the warm bias of ICLM compared to CCLM with respect to the E -OBS gridded observation dataset, especially during the warmer months. Although CCLM shows a general good performance, ICLM improves the representation of daily accumulated precipitation in spring over territories with a complex orography and reduces the bias of wind speeds . Also, the diurnal cycles are generally improved. The ICLM added value is further proven by the analyses focusing on impactful events (two flooding events, a heatwave and a windstorm) hitting the Emilia-Romagna region during the year 2023. The evaluation indicates a general good agreement of ICLM outputs with gridded observations, supporting its adoption for future high-resolution downscaling reanalysis over Italy.