本文旨在探寻反气旋性涡旋（AE)、气旋性涡旋（CE）调制下海表特征的相关性分布及其与涡旋外部（OE）区域的差异，为进一步厘清中尺度涡对海气界面的调制机制，改进海洋数值模拟提供理论依据。本文利用2010-2019年长时序遥感观测数据以及皮尔逊积距相关系数、涡旋识别等方法，系统分析了全球中尺度涡调制下海表温度异常、海面高度异常、叶绿素浓度异常、风速异常四种参数相关性的时空分布特征及其差异。结果表明，各参数之间相关性分布赤道太平洋地区相关性最强，印度洋地区以及北美洲西南岸相关性较为显著，其他地区相关性偏弱，且各参数分布呈现出较强的涡旋信号。此外，各参数的相关系数在OE中相比于CE以及AE内部降低0.2左右，相关系数分布的平滑程度约为AE与CE中相关系数分布的20倍，且相关性相反两区域之间存在明显的过渡区域。可以得出结论中尺度涡调制下各参数的相关性分布表现很强的地域特征，其差异主要表现在相关程度以及整体平滑性方面。

Mesoscale eddies are broadly distributed in the global ocean and have significant effects on the sea surface temperature (SST), sea surface height (SSH), chlorophyll (Chl), wind speed (WS), and other ocean parameters. Therefore, the coupling analysis of mesoscale eddies and ocean key parameters is an important part of ocean research. With the realization of individual eddy identification technology, the anti-cyclonic eddy (AE), cyclonic eddy (CE), and outside eddy (OE) can be better distinguished. This enables us to comprehensively study the distribution and difference of correlation of sea surface features by comparing OE with AE and CE, which provides a theoretical basis for further clarifying the modulation mechanism of the mesoscale eddy on the air-sea interface and improving ocean numerical simulation. On the basis of distinguishing AE, CE, and OE, this study calculated the Pearson correlation coefficient using sea surface temperature anomaly (SSTA), sea level anomaly (SLA), chlorophyll anomaly (CHLA), and wind speed abnormal (WSA) data from 2010 to 2019, and compared the smoothness of the correlation coefficient. The results show that the correlation distribution among the parameters has significant regional characteristics. In CE and AE, the correlation is ±0.5 in most areas of the ocean, and ±0.7 in the Northern Indian Ocean and the Equatorial Pacific. In OE, the correlation is ±0.2 in most regions, and ±0.4 in the Northern Indian Ocean and the Equatorial Pacific. In addition, when a positive (negative) correlation occurs in OE, it generally shows a large range of positive (negative) correlation, and there is a noticeable transition region between the two. However, under the influence of eddy, the extreme correlation regions are smaller and mainly present as scattered points, and the transition regions between positive and negative regions are very narrow. In terms of smoothness, the smoothness coefficient of the correlation coefficient of each parameter in OE is about 15, while the smoothness coefficient of AE and CE is about 450, which is much lower than that in OE area. We conclude that the influence of eddy on the correlation of each parameter is mainly reflected in the value, distribution and smoothness of the correlation coefficient. The correlation coefficient of each parameter in OE is about 0.2 lower than that in CE and AE, and the smoothness of the correlation coefficient in OE is about 30 times of that in AE and CE. The distribution of the correlation also has a strong point feature in the original distribution mode due to the modulation of the eddy.