Abstract
We use the International Centre for Theoretical Physics' (ICTP) regional climate model (RegCM4) nested in the European Community−Hamburg atmospheric model (ECHAM5) and the European Centre for Medium-Range Weather Forecasts' (ECMWF) 40-yr reanalysis (ERA40) data to perform and analyze multi-decadal present-day climate simulation over the Arabian Peninsula. Mean climatology, annual cycle, and interannual variability of rainfall and temperature are compared to observational data. The results show that, while ECHAM5 and ERA40 exhibit drier and warmer biases, RegCM4 simulates higher rainfall intensity and lower warm temperatures, providing a more realistic spatial distribution of these variables over the region. This originates from a more reasonable simulation of specific humidity, wind fields, and geopotential heights compared to the driving field. Annual cycles over the northern Arabian Peninsula (NAP) are better captured by RegCM4 than by ECHAM5, while, over the south (SAP), annual cycles mostly follow the driving field. This suggests that the annual cycle over the NAP is driven by regional processes represented by the internal regional climate model dynamics and physics, and, over the SAP sub-region, the influences of large-scale conditions induced by boundary forcing are dominant. Both RegCM4 and ECHAM5 exhibit similar performance in terms of variance over the low-variability region (SAP); however, the regional climate model outperforms the driving global climate model for the high-variability region (NAP). Therefore, although dynamical downscaling offers some improvements regarding climatology and the annual cycle, it has mixed benefits in terms of interannual variability.