基于变分理论的AVHRR海表温度反演应用及效果评估

基于变分理论算法实现了METOP-A卫星AVHRR传感器探测数据的海洋表面温度变分反演,进行了连续1个月的海表温度反演试验,并分别从全球、分纬度带和天气系统活跃区域3个方面,将变分反演结果（VAR SST）与利用统计回归方法反演相同卫星得到的海表温度产品（GBL SST）、其他海温融合产品（OISST）及实际浮标观测数据等进行一系列评估。从全球评估指标看出,以OISST为参照,VAR SST要优于GBL SST;以浮标观测为参照,VAR SST略逊于GBL SST,而且VAR SST还改进了GBL SST随时间波动大的缺点;从分纬度带对比看出,在与OISST对比时,VAR SST在低纬度地区和北半球中纬度地区的质量要优于GBL SST,海温反演精度较高。研究还表明,由于变分方法考虑了大气状态的变化,能够更加有效订正卫星遥感过程中大气的削弱作用,从而反演出精度更高的海表温度,尤其在天气系统较为复杂的区域效果明显。

AVHRR SST retrieval using variation algorithm and evaluation

METOP-A/AVHRR sea surface temperature (SST) retrieved by using variation algorithm. SST retrieval was done for one month. In this paper, the result of variation retrieval (VAR SST) was evaluated by the product which retrieved by using regression algorithm of the same sensor and satellite (GBL SST), OISST and the buoy observation. This evaluation was done in the global regional and the weather system active regional. According to the evaluation index, we know that VAR SST is better than GBL SST when they are compared with OISST and worse than it when they are compared with buoy observation. In addition, the VAR SST also improved the GBL SST shortcomings about time fluctuations. When the VAR SST and the GBL SST compared with OISST in different latitude zones, we discovered that the VAR SST, which with higher accuracy, is better than GBL SST in the low latitudes and the middle latitude of Southern Hemisphere. According to the research, variation algorithm can correct the atmospheric attenuation in satellite remote sensing more effectively and get a higher accurate SST, especially in the area of the complex weather system. That''s because the atmospheric temperature and water vapor mixing ratio were considered in this algorithm.