Fig. 1: Intensification in extreme cold events due to global mean sea level rise.

a The number of sea-level experiments that generate increased cumulative intensity of winter (DJF) extreme cold days at each land model grid. All results are based on the 200-year mean of each experiment. b As in (a), but for frequency. The large dots indicate that in the SL0.15 m/0.3 m, significant changes are simulated in each land model with a confidence level greater than 90% (t-test). The small dots indicate additional regions where significant changes emerge in the SL0.625 m. The stippling indicates that at least one (small dots) and two (large dots) sets of low sea level rise experiments (SL0.15-0.625 m) simulate a significant change with a confidence level greater than 90% (t-test) at each land model. c, d display the cumulative intensity and frequency, regionally averaged in the deep blue regions within the box surrounded by black dash lines in (a) and (b). The solid (hollow) dots indicate the mean change is significant (insignificant) at a 90% confidence level (t-test) compared to the SL0m experiment. Note the cumulative intensity and frequency from our pre-industrial experiment are 17.6 °C and 9 days; the cumulative strength results are taken as absolute values for better demonstration.