Skill and Inter-Model Comparison of Regional and Global Climate Models in Simulating Wind Speed over South Asian Domain

Abstract

Global Climate Models (GCMs) and Regional Climate Models (RCMs) have been widely used in understanding the impact of climate change on wind-driven processes without explicit evaluation of their skill. This study is oriented towards assessing the skill of 28 GCMs and 16 RCMs, and more importantly to assess the ability of RCMs relative to parent GCMs in simulating near-surface wind speed (WS) in diverse climate variable scales (daily, monthly, seasonal and annual) over the ocean and land region of the South Asian (SA) domain (11° S–30° N and 26° E–107° E). Our results reveal that the climate models’ competence varies among climate variable scales and regions. However, after rigorous examination of all climate models’ skill, it is recommended to use the mean ensemble of MPI-ESM-MR, CSIRO-Mk3.6.0 and GFDL-ESM2G GCMs for understanding future changes in wave climate, coastal sediment transport and offshore wind energy potential, and REMO2009 RCM driven by MPI-M-MPI-ESM-LR for future onshore wind energy potential assessment and air pollution modelling. All parent GCMs outperform the RCMs (except CCCma-CanESM2(RCA4)) over the ocean. In contrast, most RCMs show significant added value over the land region of the SA domain. Further, it is strongly discouraged to use the RCM WS simulations in modelling wind-driven processes based on their parent GCM’s skill over the ocean.