气候学
海冰
海洋学
北极冰盖
北极的
北极
北极海冰下降
地质学
印度洋
北极偶极子异常
环境科学
南极海冰
作者
Xin Cheng,Shangfeng Chen,Wen Chen,Renguang Wu,Shuoyi Ding,Wen Zhou,Lin Wang,Yali Yang,Jinling Piao,Peng Hu
标识
DOI:10.1175/jcli-d-24-0419.1
摘要
Abstract This study shows a close relationship between winter Arctic sea ice concentration (WASIC) anomalies in the Barents-Greenland Seas and the subsequent autumn Indian Ocean Dipole (IOD) based on the observational analysis and numerical simulations. Particularly, more (less) WASIC in the Barents-Greenland Seas tends to lead to a positive (negative) IOD in the following autumn. Above-normal WASIC in the Barents-Greenland Seas results in reduction of the upward turbulent heat flux and induces tropospheric cooling over the Arctic. This tropospheric cooling triggers an atmospheric teleconnection extending from the Eurasian Arctic to the subtropical North Pacific. Numerical experiments with both the linear barotropic model and atmospheric general circulation model can well capture the atmospheric teleconnection associated with the WASIC anomalies. The subtropical atmospheric anomalies generated by the WASIC anomalies then result in subtropical sea surface temperature (SST) warming, which sustains and expands southward to the equatorial central Pacific during the following summer via a wind-evaporation-SST feedback. The resulting equatorial central Pacific SST warming anomalies induce local atmospheric heating and trigger an anomalous Walker circulation with descending motion and low-level anomalous southeasterly winds over the southeastern tropical Indian Ocean. These anomalous southeasterly winds trigger positive air-sea interaction in the tropical Indian Ocean and contribute to the development of the IOD. The close connection of the WASIC anomalies with the subsequent IOD and the underlying physical processes can be reproduced by the coupled climate models participated in the CMIP6. These results indicate that the condition of WASIC is a potential effective precursor of IOD events.
科研通智能强力驱动
Strongly Powered by AbleSci AI