Possible Linkage Between Winter Extreme Low Temperature Over Western-Central China and Autumn Sea Ice Loss

Based on reanalysis datasets and sea-ice sensitivity experiments, this study has pointed out that the autumn sea ice loss in East Siberian-Chukchi-Beaufort (EsCB) Seas significantly increases the frequency of winter extreme low temperature over western-central China. Autumn sea ice loss warms the troposphere and generates anticyclonic anomaly over the Arctic region 1 month later. Under the effects of synoptic eddy-mean flow interaction and anomalous upward propagated planetary wave 2, the Arctic anticyclonic anomaly strengthens and develops toward Greenland-Northern Europe, accompanied by a weakened stratospheric polar vortex. In winter, following intra-seasonal downward propagation of stratospheric anomalies, the Northern European positive geopotential height anomalies enhance and expand downstream within 7 days, favoring Arctic cold air east of Novaya Zemlya southward (the hyperpolar path) accumulating in Siberia around Lake Baikal. In the subsequent 2-3 days, these cold anomalies rapidly intrude western-central China and induce abrupt sharp cooling, thus more frequent extreme low temperature there. In contrast, the role of autumn Barents-Kara (BK) sea ice loss exhibits quite different features, mainly favoring northwestern, northeastern and eastern China to experience increased frequency of extreme low temperature in winter.

Automated summary

Based on reanalysis datasets and sea-ice sensitivity experiments, this study found that autumn sea ice loss in the East Siberian-Chukchi-Beaufort Seas leads to more frequent extreme low temperatures in western-central China during winter. The loss of autumn sea ice warms the troposphere and creates an anticyclonic anomaly in the Arctic region one month later. This anomaly interacts with synoptic eddy-mean flow and anomalous upward propagated planetary wave 2, strengthening and developing towards Greenland-Northern Europe, while weakening the stratospheric polar vortex.