Nested Regional Climate Model (NRCM) Downscaling in Extreme Temperature: Evaluated by Generalized Extreme Value (GEV) Distribution

It is recognized that changes in frequency and intensity of extreme events are likely to have a larger impact than changes in mean climate. Therefore, generating projections of changes in future extreme events is a significant need. Global models are powerful tools to investigate the climate and climate change on large spatial scales. However, such models do not represent local and mesoscale weather phenomena well due to their coarse horizontal resolution (150–300 km). The fine-scale features in climate projections and potential extreme weather events are better represented in regional climate models.

Here, we use the Nested Regional Climate model (NRCM) to dynamically downscale NCEP-DOE Reanalysis II on a 12-km × 12-km high resolution scale over North America (including Alaska; with 600 × 515 grid cells at longitude and latitude) to simulate the current climate during 1980-2010. The calculated return periods of weekly extreme heat events over different regions of the contiguous United States in the past 30-years are compared with datasets from National Climatic Data Center (NCDC) and NCEP North American Regional Reanalysis (NARR) using generalized extreme value (GEV) distribution modeling technique.

We will also present results from downscaling the Representative Concentration Pathway (RCP) 8.5 and RCP 4.5 from the new Community Earth System Model version 1.0 (CESM v1.0) to predict the GEV distribution of weekly maximum temperature at the end-of-century (2085-2095) and to explore the uncertainties of future maximum temperature induced by different scenarios over stations of CONUS. We have corrected the CO2 atmospheric concentrations in the longwave and shortwave radiation schemes of the NRCM according to the recommended datasets by CMIP5. We have also corrected an inconsistency in skin temperature during the downscaling process by modifying the land/sea mask of CLM 4.0. Keywords: dynamical downscaling; Nested Regional Climate Model (NRCM); generalized extreme value (GEV) distribution; weekly maximum temperature; CMIP5; RCP scenarios