The future of Arctic sea ice

The reduction in sea ice is having, and is expected to continue to have, a dominant impact on local climate and ecology, and to affect extreme weather and global climate by modification of exchanges of heat and momentum in the atmosphere and ocean with lower (European) latitudes. Climate projections, generated using complex climate models, indicate that the Arctic Ocean will become seasonally ice free in the coming decades. However, these models underestimate the strength of the link between polar warming and sea ice loss; they only achieve observed rates of ice loss with unrealistic polar warming. Observations of the Arctic have improved in recent years with new satellites, e.g. IceSat-2 and CryoSat-2, measuring sea ice properties from space, and field experiments such as MOSAiC providing detailed measurements of sea ice physical processes. These show that Arctic sea ice is becoming thinner, less extensive, more fragmented, and more seasonal. Climate models of sea ice physics, built in a time of perennial ice, inadequately represent the seasonal, fragmented nature of the emerging ice cover. We will combine different satellite data products to provide estimates of the local sea ice mass budget. These measurements, among others, will be used to provide an unprecedently stringent test of sea ice models. We will enhance our sea ice models through the incorporation of representations of physical processes observed to be important in the seasonal ice cover physics, such as an evolving floe size distribution and advanced representation of frazil ice, both of which are already seen to play a leading role in the, more seasonal, Southern Ocean sea ice cover. This project will result in a necessary upgrade to model representation of Arctic sea ice. The new sea ice physics will be brought into a full climate model, which will be used to explore their impact on the ice cover of the past few decades, and their impact on decadal predictions. Our analysis of the climate simulations will utilise ideas we have explored in simpler, more idealised models and analysis of previous climate model simulations. Our aim is to produce more realistic simulations of Arctic sea ice trends and variability in the recent past and near future, as we approach a seasonally ice-free Arctic Ocean. This project will: (i) simulate the observed rates of Arctic sea ice loss in combination with the observed rates of Arctic warming; (ii) more tightly constrain when the Arctic Ocean will become seasonally ice free; and (iii) test the hypothesis that climate models’ mismatch between rates of sea ice loss and Arctic warming is a consequence of inadequate physical representation of the modern Arctic sea ice cover.