It has been widely discussed whether future climate change will be a linear response to human activities, or whether it will be punctuated by abrupt shifts, often called tipping points. One proposed tipping point is the loss of Arctic winter sea ice, i.e. the transition from a seasonally ice-covered Arctic to an Arctic Ocean that is sea-ice free all year round under increasing atmospheric CO2 levels. We show that in comprehensive climate models, such loss of winter sea-ice area is faster than the preceding loss of summer sea-ice area for the same rate of warming. The large sensitivity of winter sea-ice area in complex models is due to the asymmetry between melting and freezing: An ice-free summer requires the complete melt of even the thickest sea ice, which is why the perennial ice coverage decreases only gradually as more and more of the thinner ice melts away. In the case of seasonal ice, however, sea-ice areal coverage remains high as long as sea ice still forms in winter, and then drops to zero wherever the ocean warms sufficiently to no longer form ice at any time of the year. Although the loss of winter sea ice can therefore be faster than the preceding loss of summer sea ice, it is still a gradual and reversible phenomenon.
We also find that expected trends in variance and autocorrelation of sea-ice area and thickness are not specific to the existence or the mechanism of abrupt ice loss. Using a hierarchy of models, we find a consistent and robust decrease of variance and autocorrelation of sea-ice volume before summer sea ice is lost. We attribute this to the fact that thinner ice can adjust more quickly to perturbations. Thereafter, the autocorrelation increases, mainly because it becomes dominated by the ocean water’s large heat capacity when the ice-free season becomes longer. We show that these changes are robust to the nature and origin of climate variability in the models and do not depend on whether Arctic sea-ice loss occurs abruptly or irreversibly in a model. Based on these results, the prospects of detecting statistical early warning signals before an abrupt sea-ice loss seem very limited. However, the robust relation between state and variability can be useful to build simple stochastic climate models, and to make inferences about past and future sea-ice variability from only short observations or reconstructions.
The talk will be concluded by a short discussion on the implications of the case of Arctic sea ice for the analysis of other potential climate tipping points.