Assessing the strength of regional changes in near-surface climate associated with a global warming of 2°C

In this study, the strength of the regional changes in near-surface climate associated with a global warming of 2°C with respect to pre-industrial times is assessed, distinguishing between 26 different regions. Also, the strength of these regional climate changes is compared to the strength of the respective changes associated with a markedly stronger global warming of 4.5°C. The magnitude of the regional changes in climate is estimated by means of a normalized regional climate change index, which considers changes in the mean as well as changes in the interannual variability of both near-surface temperature and precipitation. The study is based on two sets of four ensemble simulations with the ECHAM5/MPI-OM coupled climate model, each starting from different initial conditions. In one set of simulations (1860–2200), the greenhouse gas concentrations and sulphate aerosol load have been prescribed according to observations until 2000 and according to the SRES A1B scenario after 2000. In the other set of simulations (2020–2200), the greenhouse gas concentrations and sulphate aerosol load have been prescribed in such a way that the simulated global warming does not exceed 2°C with respect to pre-industrial times. The study reveals the strongest changes in near-surface climate in the same regions for both scenarios, i.e., the Sahara, Northern Australia, Southern Australia and Amazonia. The regions with the weakest changes in near-surface climate, on the other hand, vary somewhat between the two scenarios except for Western North America and Southern South America, where both scenarios show rather weak changes. The comparison between the magnitude of the regional changes in near-surface climate for the two scenarios reveals relatively strong changes in the 2°C-stabilization scenario at high northern latitudes, i.e., Northeastern Europe, Alaska and Greenland, and in Amazonia. Relatively weak regional climate changes in this scenario, on the other hand, are found for Eastern Asia, Central America, Central South America and Southern South America. The ratios between the regional changes in the near-surface climate for the two scenarios vary considerably between different regions. This illustrates a limitation of obtaining regional changes in near-surface climate associated with a particular scenario by means of scaling the regional changes obtained from a widely used “standard” scenario with the ratio of the changes in the global mean temperature projected by these two scenarios.