基于AMSR-E与MODIS数据海表面温度遥感反演研究

Retrieval of sea surface temperature from AMSR-E and MODIS in the Northern Indian Ocean

海洋表面温度是海洋环境的重要参数。遥感技术是进行海表面温度研究的有效手段之一。以印度洋北部海域为研究区域,利用Aqua卫星上的微波数据(AMSR-E)和光学数据(MODIS),进行了海表温度反演研究。首先对AMSR-E L2A数据和MODIS L1B数据进行预处理,然后将AMSR-E的各极化通道亮温数据与实测海表温度进行相关性分析,通过多元线性回归建立AMSR-E海表温度的反演模型,而MODIS海表温度则通过采用线性多通道算法得到,最后以AMSR-E亮温数据为主,MODIS海表温度数据为辅,采用多元线性回归的方法建立了海表温度反演模型。利用该模型反演印度洋北部海域海表温度,反演结果与实测数据相比,其均方根误差为0.323 97℃。

The sea surface temperature （SST） is an important parameter of marine environment. The remote sensing technology is an effective means to retrieve the sea surface temperature. In this paper, we studied the retrieval of sea surface temperature by using the brightness temperature data obtained from the Advanced Microwave Scanning Radiometer-Earth Observing System （AMSR-E） , the infrared data from the Moderate- resolution Imaging Spectroradiometer （MODIS） and in situ SST data from the Global Ocean Data Assimilation Experiment in the Northern Indian Ocean. The original brightness temperature data of the polarization channels from AMSR-E L2A and the original MODIS L1B thermal infrared data were preprocessed firstly, and then the retrieval model of AMSR-E SST was built on the multi-parameters linear regression, based on the correlation among the AMSR-E brightness temperature and the in situ sea surface temperature. The MODIS SST was retrieved by Muhichannel algorithm. Finally, we obtained the SST from the AMSR-E brightness temperature and MODIS SST by an AMSR-E and MODIS SST retrieval model developed by the multi- parameters linear regression. This retrieval model mainly relied on the AMSR-E brightness temperature while making the MODIS surface temperature subsidiary. Compared with the in situ SST, the root mean square error of retrieved results is 0. 323 97℃.