1989—2018年南极布兰斯菲尔德海峡海冰时空变化分析  

Spatial-temporal variation of sea ice in the Bransfield Strait,Antarctica from 1989 to 2018

作　　者：卓梦婷 胡松[1,2] 朱国平[1,3,4,5] ZHUO Mengting;HU Song;ZHU Guoping(College of Marine Sciences,Shanghai Ocean University,Shanghai 201306,China;Experimental Teaching Demonstration Center for Marine Science and Technology,Shanghai Ocean University,Shanghai 201306,China;Center for Polar Research,Shanghai Ocean University,Shanghai 201306,China;Polar Marine Ecosystem Laboratory,Key Laboratory of Sustainable Exploitation of Oceanic Fisheries Resources,Ministry of Education,Shanghai 201306,China;National Engineering Research Center for Oceanic Fisheries,Shanghai 201306,China)

机构地区：[1]上海海洋大学海洋科学学院,上海201306 [2]上海海洋大学海洋科学与技术实验教学示范中心,上海201306 [3]上海海洋大学极地研究中心,上海201306 [4]大洋渔业资源可持续开发教育部重点实验室极地海洋生态系统研究室,上海201306 [5]国家远洋渔业工程技术研究中心,上海201306

出　　处：《上海海洋大学学报》2021年第6期1113-1122,共10页Journal of Shanghai Ocean University

基　　金：国家重点研发计划(2018YFC1406801);国家自然科学基金(41776185)。

摘　　要：基于1989—2018年美国冰雪中心海冰密集度数据,分析近30年南极布兰斯菲尔德海峡及附近海域海冰分布规律及变化趋势。结果表明,布兰斯菲尔德海峡内部海冰属于1年冰,海冰密集度于2月达到最低,8月达到最高,历史上强厄尔尼诺年份与该区域海冰密集度较低的年份存在明显对应关系。对30年冰情线性拟合得到年际变化趋势,海峡内结冰月份普遍延迟,年结冰月数以减少为主。3个代表站点的海冰密集度分析结果显示,海峡内A站点海冰变化与海峡口外北部海域B站点的相关性较强,与海峡口外南部靠近威德尔海区域C站点的相关性弱。循环神经网络模型可用于预测和分析海冰密集度的时序变化,尤其是适合海冰密集度相对较低的时序分析。Based on the sea ice concentration data from the US National Snow and Ice Data Center,the distribution and trend of sea ice in the Bransfield Strait in recent 30 years are analyzed.The results show that the sea ice in Bransfield Strait is dominated by one-year ice,reaches the lowest in February and reaches the highest in August.There is obvious correlation between the years with low sea ice concentration and the strong El Ni1 o years in the historical years.The trend of sea ice concentration in the Bransfield Strait in recent 30 years shows that the freezing month of sea ice was generally delayed and the sea ice concentration was mainly reduced.The analysis on ice concentration in three typical stations shows,the correlation between the sea ice in Bransfield Strait(Station A) and the sea ice in the north of the Strait mouth(Station B) is strong,while the correlation with the sea ice in the south of Bransfield Strait near the Weddell Sea(Station C) is weak.The recurrent neural network(RNN) model can be used to predict and analyze temporal variation in sea ice concentration,particularly in the years with low ice concentration level.

关 键 词：海冰密集度 布兰斯菲尔德海峡 南极 全球变化 循环神经网络 ENSO 

分 类 号：P728.2[天文地球—海洋科学]