The effects of Arctic sea-ice thickness loss and stratospheric variability on mid-latitude cold spells

The Arctic is a key indicator of global climate change. Annual surface temperatures are increasing at more than twice the rate of the global average, and passive microwave satellite observations of Arctic sea-ice extent show a loss of nearly 40% over the last few decades. Despite recent advances in climate models, availability of observations, and statistical analysis, our understanding of ice-ocean-atmosphere interactions, and the teleconnections between the Arctic and other regions, remains incomplete. Assessing the atmospheric response to the rapid changes in Arctic sea ice will help to determine future societal impacts from climate change, including extreme weather events in the densely populated mid-latitude regions.The broad objective of this thesis is to improve our understanding of Arctic climate variability. To be more specific, this dissertation will explore (i) the internal variability of Arctic sea-ice thickness (SIT), (ii) the atmospheric response to thinning sea ice, and (iii) the role of stratospheric and tropospheric pathways in modulating Arctic-mid-latitude teleconnections. SIT has an important effect on the Arctic energy budget, and therefore it is critical to represent it accurately in global climate models. However, limited observations of SIT (satellite-derived and in situ) have prevented a robust analysis of the atmospheric response to its spatial and temporal variability. Here, I use perturbation experiments in a high-top atmospheric global climate model to examine the importance of stratosphere-troposphere coupling and other causal pathways for teleconnections between Arctic sea ice and the mid-latitudes.Internal variability contributes to an uncertainty of 10-20 years in the timing of future mean SIT falling below 0.5 m in a large ensemble of simulations of a fully-coupled global climate model. This loss of SIT is found to reinforce the large-scale tropospheric response to Arctic sea-ice concentration and contributes up to one third of the surface warming response. Using a high-top model, the Quasi-biennial Oscillation (QBO) is found to significantly modulate the response to Arctic sea-ice decline. The stratospheric polar vortex weakens in response to sea-ice forcing during easterly QBO winters. However, there is little-to-no stratospheric response to sea-ice loss during the westerly phase of the QBO. Finally, in a series of coordinated coupled and uncoupled global climate model perturbation experiments, the “warm Arctic, cold Siberia” temperature anomaly pattern is found to be closely related to the strength of the Siberian High and mid-tropospheric warming response. By comparing simulations that are directly forced with warmer temperature in the Arctic region (corresponding to projected changes), to sea-ice forced experiments (where the forcing corresponds to the same future time), it is shown that sea-ice loss alone is insufficient to bring about the entire thermal signature of Arctic amplification