Detecting Anthropogenic Greenhouse Gas and Aerosol Influences on the Observed Changes in Extreme Temperatures during 1951-2015

We carried out an updated detection and attribution analysis of extreme temperature changes for 1951-2015. Annual maximum daily maximum/minimum temperatures (warm extremes) and annual minimum daily maximum/minimum temperatures (cold extremes) were considered at global, continental (6 domains), and subcontinental (33 domains) scales. By applying an optimal fingerprinting technique, HadEX3 observations were compared with CMIP6 multi-model simulated response patterns (fingerprints) to anthropogenic (ANT), greenhouse gas (GHG), anthropogenic aerosol (AA), and natural (NAT) forcings. To estimate the internal climate variability, pre-industrial control simulations (CTL) were also used. Results from two-signal detection analysis (where the observations are simultaneously regressed onto ANT and NAT fingerprints) show that ANT signals are detected robustly separated from NAT in global and most continental regions across all extreme temperatures. At subcontinental scales, ANT signals are mainly detected in warm extremes (more than 60% of regions). Results from three-signal detection analysis (where observations are regressed onto three signals including GHG, AA, and NAT) indicate that GHG signals are detected frequently and separated from other external forcings over global, most continental, and many subcontinental (more than 60%) domains in warm extremes. In addition, AA signals are jointly detected in warm extremes over global, Europe and Asia. Further analysis of attributable trends reveal that most of observed warming is explained by GHG signals while AA signals offset GHG-induced warming over globe, Europe, and Asia. Overall, our results well support previous findings with improved detection at some continental scales due to extended period and/or better data coverage.