| In recent decades,Arctic sea ice has continued to decrease.The influence of autumn sea ice changes on winter surface air temperature(SAT)has been controversial,recent studies have shown that the effects of different regional variations of the sea ice are often different or even opposite.Therefore,based on the Hadley Center’s sea ice data and NCEP-NCAR reanalysis data,this paper studies the winter surface air temperature response of sea ice in different sea areas in autumn.To further explore the possible physical mechanism of this response process.The results show that in the past 40 years,the Arctic sea ice has been dominated by the coordinated changes of the Barents Sea-Kara Sea(B-K)and Laptev Sea-Beaufort Sea(L-B),with the centers mainly concentrated near the Barents Sea to Beaufort Sea.The response of winter SAT is Eurasian continental cold anomaly.In contrast,In a few years,sea ice variation also manifested as dipole between B-K and L-B,which was rarely mentioned in previous studies.When the sea ice decreases in the B-K area while increases in the L-B area,anomalous warming in the Arctic and anomalous low temperature in the mid-latitude is more obvious than that in the first mode.In this condition,significantly low SATs appear not only in the central and northeastern Eurasia but also in the North Atlantic and central North America.It shows that the impact of local sea ice changes in different regions on the winter weather and climate is different.When the sea ice in the B-K and L-B sea areas decreases uniformly,there is a circulation response in the upper atmosphere,which is manifested as a weakening of the stratospheric polar vortex.The influence of the sea ice changes in the B-K sea area on the surface air temperature and the upper atmosphere plays a major role,but the L-B sea still has certain effects.When the sea ice in the B-K and L-B sea area change in reverse phase,the response of the stratospheric circulation is weakened.Evaluating the surface temperature response of sea ice changes in different modes of CMIP6,in general,it seems that the surface temperature simulates the polar warm anomaly of the first mode change of sea ice well,but the response to the mid-high latitude cold anomaly is not clear.The model responds ideally to the mid-high latitude cold anomaly caused by the second modal change of sea ice,and can reproduce the local cold anomaly response.The tropospheric circulation response corresponding to the model also has a certain reproducibility,and the anomalies of the surface temperature in each model can correspond well to the changes of the tropospheric circulation,and the model data itself has good variable consistency.In the model data,the B-K sea ice reduced temperature and the response of the strato-tropospheric circulation are not well reproduced,but individual models have a certain ability to reproduce local cooling and circulation changes.Therefore,the model evaluation of CMIP6 still has certain reference value.When the B-K and L-B sea ice change consistently,they will have a significant impact on the temperature in the northern part of Xinjiang.In the autumn of 2009,the sea ice of B-K and L-B decreased,providing a favorable background field for the frequent cooling and snowfall events in northern Xinjiang.In this winter,the polar vortex in the stratosphere was weak,the Ural high is strong,and the surface Siberian high was strengthened,resulting in 5 continuous cooling and snowfall processes.During this period,the upper polar vortex is southerly.The abundant water vapor conditions,combined with the strong and continuous oscillation of Siberia,are the reasons for the frequent occurrence of extreme cooling and snowfall. |