FORTRAN: G95 (GCC 4.0.3 (g95 0.94!) Jan 17 2013); Grid Analysis and Display System (GrADS) Version 2.0.2
1. For each model at each grid point, the mean annual temperature under the RCP4.5 scenario was calculated.
2. For each model, these data were re-gridded to a common grid.
3. For each model at each grid point, the mean temperature under the RCP4.5 scenario was calculated for two periods: 1976–2005 and 2070–2099.
4. At each grid point, the difference in projected temperature was calculated for 2070–2099 minus 1976–2005.
5. At each grid point, the mean precipitation difference was computed by averaging the following models: For RCP4.5: ACCESS1-0, ACCESS1-3, bcc-csm1-1, BNU-ESM, CanESM2, CCSM4, CESM1-BGC, CMCC-CMS, CNRM-CM5, CSIRO-Mk3-6-0, FGOALS-g2, GFDL-CM3, GFDL-ESM2G, GFDL-ESM2M, GISS-E2-H-p1, GISS-E2-R-p1, HadGEM2-AO, HadGEM2-ES, inmcm4, IPSL-CM5A-LR, MIROC5, MIROC-ESM, MPI-ESM-LR, MPI-ESM-MR, MRI-CGCM3, NorESM1-M. For RCP8.5: ACCESS1-0, ACCESS1-3, bcc-csm1-1, BNU-ESM, CanESM2, CCSM4, CNRM-CM5, CSIRO-Mk3-6-0, FGOALS-g2, GFDL-CM3, GFDL-ESM2G, GFDL-ESM2M, GISS-E2-R-p1, HadGEM2-AO, HadGEM2-ES, inmcm4, IPSL-CM5A-LR, MIROC5, MIROC-ESM, MPI-ESM-LR, MPI-ESM-MR, MRI-CGCM3.
6. For each model at each grid point, the mean annual temperature under the pre-industrial control runs (at least 500 years long) was calculated. The first 100 years of the pre-industrial runs were ignored. Variability was calculated for every grid point as the standard deviation of non-overlapping 20-year means, multiplied by the square root of 2. This is by definition the standard deviation of the difference between two independent 20-year averages having the same variance, and estimates the variation of the difference that would be expected due to unforced internal variability. The median across all models of that quantity is used.
7. The mean temperature difference was plotted for grid points in North America, with model agreement indicated as follows:
• "Whited out" – large changes where multi-model average change is greater than double the standard deviation of the 20-year mean from control runs and less than 90 percent of the models agree on the sign.
• Stippling – large changes where multi-model average change is greater than double the standard deviation of the 20-year mean from control simulations runs, and 90 percent of the models agree on the sign.
• Hatching – No significant change where multi-model average change is less than the standard deviation of the 20-year means from control simulations runs.
• Color only – changes are between one standard deviation and two standard deviations.
cssr-26d1-402b-9740-process
Temp_Projected Changes in Average Annual Temp_V3.png;
access1-0.gs; access1-3.gs; bcc-csm1-1-t.gs; bnu-t.gs; can-t.gs ccsm4-t.gs cnrm-cm5-t.gs csiro-mk3-6-0-en01-t.gs fgoals-g2-t.gs gfdl-cm3-t.gs gfdl-esm2g-t.gs gfdl-esm2m-t.gs giss-e2r-t.gs hadgem2-a0-t.gs hadgem2-es-t.gs inmcm4-t.gs ispl-cm5a-lr-t.gs miroc5-t.gs miroc-esm-chem-t.gs miroc-esm-t.gs mpi-esm-lr-t.gs mpi-esm-mr-t.gs mri-cgcm3-t.gs;
intp.gs intp-1.gs;
mme.gs;
w.f
Red Hat Enterprise Linux Server release 7.3