Variability of high latitude amplification of anthropogenic warming

Climate models have long predicted that the high latitude response of near-surface air temperatures should be greater than at lower latitudes as a result of snow/ice feedbacks. Here we show that a regression analysis of observed global surface temperatures and anthropogenic and natural forcings could misleadingly suggest that climate models fail to capture the observed zonal mean pattern of response to anthropogenic forcings. A better approach to detecting changes and to determine consistency of climate models and observations is to use multiple features of the response pattern derived from physically-based climate models, as has been done in optimal detection studies. We show that multi-variable fingerprints can more easily detect anthropogenic changes thereby offering the potential to more robustly quantify anthropogenic influence on aspects of the climate system such as the cryosphere and the hydrological cycle. Citation: Stott, P. A., and G. S. Jones (2009), Variability of high latitude amplification of anthropogenic warming, Geophys. Res. Lett., 36, L10701, doi: 10.1029/2009GL037698.