The equilibrium climate sensitivity (ECS) of global climate models (GCMs) strongly influences projected climate changes. Reasonably, the ECS of a GCM may be expected to affect dynamically downscaled projections. The variance of global climate projections with ECS is well understood. However, there has been little examination of the effect of the ECS of a GCM on regional climate model (RCM) simulations. Therefore, we explore the sensitivity of different RCMs to different ECSs, and establish a corresponding metric of RCM ECS response, or RECS, which distinguishes the magnitude of an individual RCM’s response to ECS by the magnitude of its projection.
This work leverages 50km, 25km, and 12km resolution RCM simulations completed for the North American Coordinated Regional Downscaling Experiment (NA-CORDEX) and the DOE Framework for Assessing Climate’s Energy-Water-Land Nexus using Targeted Simulations (DOE-FACETS) projects. In the end, projections from RegCM version 4 (RegCM4) and the Weather Research and Forecasting (WRF) model correlate well with the ECSs of the driving GCMs. Over North America, both RCMs exhibit an increase in annual and seasonal mean temperature change that strongly correlates with increases in ECS even at a subregional scale. Precipitation change also correlates with increasing ECS, at the continental scale, especially in winter. Generally, WRF has a greater RECS than RegCM4, as WRF produces projections of a greater magnitude for temperature and precipitation, with a difference that is consistent as ECS increases. Since each RCM has a particular response to changes in ECS, a sufficient ensemble of RCMs for evaluating climate change impacts requires variation in RECS. Further, the effect of different RECS in an RCM ensemble may elucidate the range of climate projections as well as their sampled uncertainty.