Multi-Model Assessment of Regional Surface Temperature Trends: CMIP3 and CMIP5 20 th Century Simulations

Regional surface temperature trends from the CMIP3 and CMIP5 20 th century runs are compared with observations -at spatial scales ranging from global averages to individual grid points -using simulated intrinsic climate variability from pre-industrial control runs to assess whether observed trends are detectable and/or consistent with the models’ historical run trends. The CMIP5 models are also used to detect anthropogenic components of the observed trends, by assessing alternative hypotheses based on scenarios driven with either anthropogenic plus natural forcings combined, or with natural forcings only. Modeled variability is assessed via inspection of control run time series, standard deviation maps, spectral analyses, and low-frequency variance consistency tests. The models are found to provide plausible representations of internal climate variability, though there is room for improvement. The influence of observational uncertainty on the trends is assessed, and found to be generally small compared to intrinsic climate variability. Observed temperature trends over 1901-2010 are found to contain detectable anthropogenic warming components over a large fraction (about 80%) of the analyzed global area. In several regions, the observed warming is significantly underestimated by the models, including parts of the southern Ocean, south Atlantic, far eastern Atlantic, and far west Pacific. Regions without detectable warming signals include the high latitude North Atlantic, the eastern U.S., and parts of the eastern Pacific. For 1981-2010, the observed warming trends over about 45% of the globe are found to contain a detectable anthropogenic warming; this includes much of the globe within about 40-45 degrees of the equator, except for the eastern Pacific.