Respective roles of direct GHG radiative forcing and induced Arctic sea ice loss on the Northern Hemisphere atmospheric circulation

Remarkable separability of circulation response to Arctic sea ice loss and greenhouse gas forcing

“…This decrease is apparently not due to differences in the Sum and Full sea ice concentrations that may have arisen due to imperfections in nudging (see the green curves in Figures a and b). The drop in linearity appears to be due in part to a decrease in the contribution of sea ice loss, whose response has virtually no resemblance to the Full pattern in the extended summer months of May through September (Figure c) due to there being less response to ice loss in summer months [see, e.g., Oudar et al , ]. Except for SAT in winter, the Full response is generally dominated by the CO 2 doubling contribution, consistent with the findings of Barnes and Polvani [], but this contribution also exhibits a decrease in resemblance to the Full pattern in spring (mainly April–May), and for thermodynamic variables in spring and fall (mainly October–November; Figure e).…”

“…We found that the responses to both Arctic sea ice loss and to doubled CO 2 are relatively insensitive to mean state and are quite additive in that their sum represents the response to the combined forcing remarkably well. The decrease in SLP in the Aleutian Low, increase in SLP over Eurasia, and decrease over the Arctic and northern North America due to sea ice loss appear in a number of studies using different coupled climate models [ Sun et al , ; Deser et al , ; Blackport and Kushner , ; Oudar et al , ], suggesting that they may be robust features of the winter response to Arctic sea ice loss. Furthermore, Oudar et al [] show a zonal wind response to Arctic sea ice loss that counters that of increased greenhouse gases in the North Atlantic but enhances it in the Pacific in winter, consistent with our findings, and adding to evidence that a lack of robust North Atlantic zonal wind or jet changes in CMIP5 models may be related to Arctic sea ice loss [ Barnes and Polvani , ].…”

“…AOGCM studies indicate that the extent of influence of Arctic sea ice loss is the entire globe, with small but significant tropospheric warming whose pattern resembles the response to increased greenhouse gases [ Deser et al , ; Blackport and Kushner , ], and stronger atmospheric circulation changes compared to the atmospheric general circulation model (AGCM) response [ Deser et al , ]. The results of the AOGCM studies suggest that sea ice can modulate the global warming response, including sometimes reinforcing and sometimes opposing the response to greenhouse gas (GHG) forcing [ Deser et al , ; Oudar et al , ].…”

“…We thus neglect the effects of transient climate responses, tropospheric aerosols, and other GHGs present in realistic climate scenarios used in other work [ Deser et al , ; Blackport and Kushner , ; Oudar et al , ]. We note that our experimental design is very similar to that of Oudar et al [], but whereas they analyze the separate components of circulation change from increased GHGs and Arctic sea ice loss, we focus on the complementary question of whether these separate “forcings” are linearly additive. We show that separate attribution to sea ice and GHGs on temperature and circulation can be made to a degree that is perhaps surprisingly precise, from hemispheric to regional length scales.…”

Global Impacts of Arctic Sea Ice Loss Mediated by the Atlantic Meridional Overturning Circulation

“…Moreover, Arctic sea ice decline can cause large‐scale changes in ocean overturning circulation. Specifically, sea ice retreat generates positive buoyancy anomalies in the Arctic that can spread to the North Atlantic and induce the weakening of the Atlantic meridional overturning circulation (AMOC; e.g., Liu et al, ; Oudar et al, ; Scinocca et al, ; Sévellec et al, ; Sun et al, ). In a different context, the impacts of weakened AMOC have been studied in freshwater hosing experiments, which show a robust southward shift of the ITCZ (e.g., Barreiro et al, ; Broccoli et al, ; Fedorov et al, ; Liu & Hu, ; Stouffer et al, ; Zhang & Delworth, ).…”