Regional Climate Change Projections from NA-CORDEX and Their Relation to Climate Sensitivity
The climate sensitivity of global climate models strongly influences projected climate change due to increased atmospheric carbon dioxide in simulations. We show how that matters in a dynamically downscaled ensemble made up of multiple regionals model simulations that have downscaled multiple global models.
Prior to this study, there had been little-to-no documented examination of how global model climate sensitivity influences regional model projections in an ensemble of simulations. We show that climate sensitivity is important in producing an array of regional projections that is representative of what a global model ensemble would have.
We show that global climate model (GCM) climate sensitivity is a very important source of spread in the regional climate model (RCM) ensemble provided by the North American Coordinated Regional Climate Downscaling Experiment (NA-CORDEX), particularly for temperature from a local-to-regional-to-continental scale. The range of temperature projections sampled would be reduced if only low, mid, or high climate sensitivity simulations were considered, or if only the ensemble mean were considered. Precipitation projections correlate with climate sensitivity, but only at a continental scale during the cold season, due to the increasing influence of other processes at finer scales. Additionally, under RCP8.5, the NA-CORDEX ensemble generally spans the full range of temperature and precipitation projections provided by CMIP5.