Factors contributing to rapid decline of Arctic sea ice in autumn
at : Jun 11, 2021 16:00:42  (view:299)

Factors contributing to rapid decline of Arctic sea ice in autumn

LI Shuyao1,2, CUI Hongyan1,2,3*, XU Junli1,2, GONG Xiang1,2, YANG Yanzhao1,4, WANG Ping1 & HAN Yuqun1 

1 College of Mathematics and Physics, Qingdao University of Science and Technology, Qingdao 266061, China;

2 Research Institute for Mathematics and Interdisciplinary Sciences, Qingdao University of Science and Technology, Qingdao 266061, China;

3 Laboratory for Regional Oceanography and Numerical Modeling, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266061, China;

4 The First Institute of Oceanography, Ministry of Natural Resources, Qingdao 266061, China

Abstract  Autumn Arctic sea ice has been declining since the beginning of the era of satellite sea ice observations. In this study, we examined the factors contributing to the decline of autumn sea ice concentration. From the Beaufort Sea to the Barents Sea, autumn sea ice concentration has decreased considerably between 1982 and 2020, and the rates of decline were the highest around the Beaufort Sea. We calculated the correlation coefficients between sea ice extent (SIE) anomalies and anomalies of sea surface temperature (SST), surface air temperature (SAT) and specific humidity (SH). Among these coefficients, the largest absolute value was found in the coefficient between SIE and SAT anomalies for August to October, which has a value of −0.9446. The second largest absolute value was found in the coefficient between SIE and SH anomalies for September to November, which has a value of −0.9436. Among the correlation coefficients between SIE and SST anomalies, the largest absolute value was found in the coefficient for August to October, which has a value of −0.9410. We conducted empirical orthogonal function (EOF) analyses of sea ice, SST, SAT, SH, sea level pressure (SLP) and the wind field for the months where the absolute values of the correlation coefficient were the largest. The first EOFs of SST, SAT and SH account for 39.07%, 63.54% and 47.60% of the total variances, respectively, and are mainly concentrated in the area between the Beaufort Sea and the East Siberian Sea. The corresponding principal component time series also indicate positive trends. The first EOF of SLP explains 41.57% of the total variance. It is mostly negative in the central Arctic. Over the Beaufort, Chukchi and East Siberian seas, the zonal wind weakened while the meridional wind strengthened. Results from the correlation and EOF analyses further verified the effects of the ice–temperature, ice–SH and ice–SLP feedback mechanisms in the Arctic. These mechanisms accelerate melting and decrease the rate of formation of sea ice. In addition, stronger meridional winds favor the flow of warm air from lower latitudes towards the polar region, further promoting Arctic sea ice decline.

Keywords  Arctic sea ice, rapid decline, empirical orthogonal function, feedback mechanisms