夏季琼东上升流的高频变化特征研究*

以往对上升流的研究更多的是关注其年际或季节变化, 高时间分辨率遥感产品的出现使得研究上升流的高频特征成为可能。本文基于融合的逐日海表温度数据, 结合多尺度分割方法, 提出了一种探测上升流冷信号异质性的新算法, 通过上升流面积和强度指数展示了一个完整的琼东上升流过程。分析SST(Sea Surface Temperature)图像的结果表明, 夏季琼东涌升到海面的上升流存在间断期, 平均间隔为6d。每年夏季(6—9月)平均有98d在海表面识别到上升流, 其平均面积为7698km², 平均强度为1.0℃, 两者存在较高的相关性。琼东上升流发生频率与离岸距离成反比, 琼东北海域为高发区。不同上升流的影响因子可能不同, 离岸风、风应力旋度、热带气旋均与上升流短期变化密切相关。

Study on the characteristics of high-frequency variation of Qiongdong upwelling in summer*

Previous studies on upwelling mainly focus on inter-annual or seasonal variation, but the high-frequency observations of upwelling gain more attention with the availability of high-temporal-resolution remote sensing products. In this paper, we propose an algorithm for detecting cold upwelling signal based on daily merged sea surface temperature (SST) data with a multi-scale segmentation method. The upwelling area and intensity variation reflect a complete process of upwelling off eastern Hainan Island. Analyses on SST images in recent decade demonstrate that the upwelling did not exist daily during summer; instead, it occurred with a mean time interval of 6 days. Colder signature of upwelling events appeared for approximately 98 days on average in every summer from June to September. The mean values of cold area and intensity are 7698 km² and 1.0℃, respectively, which are highly correlated. The frequency of upwelling occurrence is inversely proportional to the offshore distance, and the northeast is a high-frequent area. Offshore wind and wind stress curl make different contributions for different upwelling in the studied area. Offshore wind, wind stress curl and tropical cyclones are closely related to the short-term change of the upwelling.