High-resolution, bias-corrected climate change projections

WP3 has delivered a first set of climate change projection data for 1991-2060 to be used in crop modelling. The underlying climate models are all forced by RCP8.5, which is the highest of the standard greenhouse gas scenarios currently recommended by IPCC. Climate model data for the MODEXTREME European case-study areas have been selected from the Euro-CORDEX archive of regional climate models (RCMs). Of the available RCMs, four have been selected which span as much of the range as possible of the precipitation climate change signal of the full model ensemble: CERFACS-RCA, ECEARTH-RACMO, ECEARTH-HIRHAM and MPIESM-CCLM. As RCM data are inherently subject to bias and crop models are sensitive to the absolute value of input variables, daily temperature (maximum and minimum) and precipitation output of the RCMs have been bias corrected against observations (taken from the E-OBS gridded dataset) using an empirical statistical quantile mapping.
The daily RCM output provided for input to the crop models is supplemented by time series of crop-relevant extreme indices of both temperature and precipitation (including indices tailored for particular phenological stages), for the original and bias corrected RCM data as well as for the observations. These indices include the number of frost days, summer days, tropical nights, growing season length, heavy precipitation days, consecutive dry days, consecutive wet days, sowing-emergence critical days and ceiling temperature for development.

Posted in Science briefs.