| 摘要: |
| 利用全球海洋Argo网格数据集、SODA月平均海洋数据集和CCMP风场数据，通过EOF分析，揭示了阿拉伯海5、50、100、200 m层海温全年2次增温、2次降温的双峰变化特征.结果表明，5 m层温度变化双峰信号出现在第一模态，其方差贡献率为75.79%，该信号主要受风场、太阳辐射及风生环流影响；50 m层温度变化双峰信号出现在第三模态，其方差贡献率为11.95%，该信号主要受风生环流影响；100 m层温度变化双峰信号出现在第一模态和第三模态，其中第一模态方差贡献率为52.03%，第三模态方差贡献率为9.55%.由100 m层第一模态可知，100 m层温度变化幅度最大、变化范围最广，是由于100 m层处于海洋温度变化最为剧烈的温跃层中.100 m层海温变化主要受风应力旋度（方向：向上为正）影响，风应力旋度为负时，大气对海洋的强迫导致局地海水辐合，温跃层加深，100 m层部分海域温度升高；风应力旋度为正时，大气对海洋的抽吸导致局地海水辐散，海洋深处的冷水上升，100 m层部分海域温度降低. |
| 关键词: 海洋水文学 上层海温 EOF分析 双峰变化 阿拉伯海 |
| DOI：10.3969/J.ISSN.2095-4972.2018.02.002 |
|
| 基金项目:全球变化与海气相互作用专项资助项目（GASI-IPOVAI-02, GASI-IPOVAI-03）；国家海洋局第三海洋研究所基本科研业务费专项资金资助项目（2017012）；国家重点研发计划资助项目（2017YFC1405100） |
|
| Bimodal character of upper ocean temperature in the Arabian Sea |
| GAO Lu,XU Jin-dian,QIU Fu-wen,LIN Xin-yu |
| (Third Institute of Oceanography ， SOA ， Xiamen 361005 ， China) |
| Abstract: |
| Argo gridded data set， simple ocean data assimilation monthly ocean data set and cross calibrated multiplatform wind field data were used to reveal the bimodal variation character of the twice warming and twice cooling processes at 5m， 50m， 100m and 200m of ocean temperature of the Arabian Sea used empirical orthogonal functions （EOF）.The results indicates that the bimodal pattern of 5m is found in the first mode which explained 75.79% variance contribution rate over the domain. The temperature variation of 5m is mainly influenced by wind， solar radiation and winddriven circulation. The bimodal pattern of 50m is found in the third mode which explained 11.95% variance contribution rate over the domain. In this layer temperature variation is mainly influenced by winddriven circulation. The bimodal pattern of 100m is found in the first and the third mode which explained 52.03% and 9.55% variance contribution rate， respectively.The first mode of 100m shows that the temperature in 100m ranges with maximum amplitude， and 100m of the Arabian Sea is amongst the thermocline of which temperature changes most dramatically in the ocean， therefore the temperature in 100m ranges with maximum amplitude and extension. The distribution of sea temperature at 100m is mainly governed by wind stress curl （positive direction: upwards）. When the wind stress curl is positive， atmospheric forcing the ocean leads to the convergence of the local sea water， and thus thermocline deepened induces the increase of sea temperature at 100m in some regions; in contrast when the wind stress curl is negative， atmospheric pumping the ocean leads to the divergence of the local sea water， and thus deep cold water rise results the decrease in temperature at 100m in some regions. |
| Key words: marine hydrology upper ocean temperature EOF analysis bimodal character Arabian Sea |
