Understanding the Role of Positive Feedbacks in Abrupt Winter Arctic Sea Ice Loss using CMIP and a Conceptual Model

Abstract:

The melting of Arctic sea ice under future global warming projections has a myriad of social and environmental implications, from threatening high-latitude ecosystems, to contributing to Arctic Amplification and thus influencing global climate. While previous work generally concludes that summer sea ice is expected to decline smoothly until its complete disappearance, studies exploring the decline of seasonal winter Arctic sea ice in the long-term disagree on whether such loss is expected to be gradual or abrupt, and if it is reversible with decreasing greenhouse gas emissions. In particular, six Global Climate Models (GCMs) run for the IPCC’s Coupled Model Intercomparison Project 5 (CMIP5) under the Extended RCP8.5 warming scenario show markedly different behavior of winter sea ice, ranging from gradual decline to abrupt disappearance that resembles a “tipping point” behavior. In this study we propose a year-to-year positive feedback mechanism that can accelerate winter sea ice loss, in which warm, open oceans at the start of winter allow for the moistening and warming of the lower atmosphere, which in turn increases the downwards clear-sky longwave radiation at the surface and suppresses ocean freezing. This leads to delayed and diminished winter sea-ice growth, and allows for increased shortwave absorption due to lowered surface albedo during springtime. Finally, the ocean stores this additional heat throughout the summer and fall seasons, setting up even warmer ocean conditions that lead to further winter sea-ice reduction. We show that the strength of this feedback mechanism is correlated with the magnitude of abrupt sea ice loss seen across the six GCMs, suggesting that it plays a significant role in driving the inter-model variability of winter sea ice behavior. Next, we use a simple, zero-dimensional toy model with coupled sea ice, atmospheric, and oceanic components to investigate which aspects could be responsible for the spread in feedback strengths across GCMs, and whether abrupt winter sea ice loss can occur in physically realistic parameter regimes. Further insight into the likelihood of abrupt winter Arctic sea ice loss will greatly inform social and economic planning for a year-round ice-free Arctic in the future.