Extreme Cold Days and Spells in Northern Europe at 0.5°C–2.0°C Global Warming Levels

Abstract

Projections for cold days and spells in Northern Europe at global warming levels of 0.5°C–2.0°C are derived from bias-corrected output data from 60 runs performed with 25 global climate models. The threshold temperature of a cold day is defined as the 10th percentile of December–February daily mean temperatures at the 0.5°C warming level. A transition from the 0.5°C to the 2.0°C warming level reduces the multi-model mean annual number of cold days by 60%–80%. The total cold day extremity index, consisting of the sum of anomalies below the threshold temperature during the cold season, is reduced even more, by 70%–90%. The largest changes occur in the Kola Peninsula and regions surrounding the White and Baltic Seas. This pattern of change is consistent with the signal-to-noise ratio of the winter mean temperature increase. Nevertheless, inter-model differences in the response are substantial, being largely attributable to the manner in which global warming affects regional winter temperatures. To assess the occurrence of severe cold spells, a 1200-year sample was established by concatenating the bias-corrected output of all 60 model runs. From this extensive sample, it is feasible to robustly determine the return levels of even very extreme cold spells. A severe cold spell occurring once in 10 years on average at the 0.5°C warming level would be experienced approximately once in 30 (70) years at the 1.5°C (2.0°C) level. Correspondingly, a baseline-climate 100-year cold spell would occur at the 2°C warming level less frequently than once in a 1000 years.