Global Patterned Sea Surface Temperature Forcing Data for CMIP5

A global patterned sea surface temperature (SST) forcing data set was required for CMIP5 Experiment #6.6 to simulate cloud feedbacks and response to imposed change in SST pattern and to isolate the role of atmospheric processes in the response of clouds and precipitation to global warming. The data set is a normalized multi-model ensemble mean of the ocean surface temperature response pattern [i.e., the change in ocean surface temperature (TOS) between years 0-20 and 140-160, the time of CO2 quadrupling in the 1% runs] from thirteen CMIP3 Atmosphere-Ocean General Circulation Models (AOGCMs) (i.e., CCCMA, CNRM, GFDLCM20, GFDLCM21, GISSER, INMCM3, IPSL, MIROC-MEDRES, MIUB, MPI, MRI, NCAR-CCSM3, and NCAR-PCM1). Before computing the multi-model ensemble mean, each model's TOS response was divided by its global mean and multiplied by four (4). This guaranteed that the pattern information from all models was weighted equally and the global mean SST forcing was the same as in the uniform +4K experiment (#6.8 in the CMIP5 experimental design document). Please note that CFMIP SST perturbation experiments (6.5, 6.6, 6.8) do not perturb sea ice, which were kept at control values, as for experiment 3.3.