赤道太平洋SMOS海表盐度数据的评估及借助神经网络的订正

Evaluation of SMOS sea surface salinity data in the equatorial Pacific and its correction using neural network

文章对土壤湿度和海洋盐度（soil moisture and ocean salinity, SMOS）卫星遥感所得2011-2012 年赤道太平洋海域海表盐度数据进行了质量控制并首次分析了盐度反演误差的海洋动力过程影响因子, 在此基础上引入神经网络方法对同时期的盐度数据进行了订正.研究发现, 降水及其诱发的表面波会使盐度误差向负方向显著增长; 海面风场导致的海面粗糙度会增大盐度误差, 风速与盐度误差呈微弱正相关; 海表温度变化则对盐度反演无影响.考虑降雨、风速等主要海洋动力过程影响因子, 利用神经网络方法对2011 年12 月赤道太平洋海域的海表盐度数据进行了订正, 其均方根误差由0.3837 降到0.2441.结果发现, 订正后的盐度数据不但消除了因降水等动力过程导致的盐度误差, 亦在赤道太平洋海域揭示了原SMOS 数据无法刻画的高盐舌现象.

In this paper, sea surface salinity data of the equatorial Pacific Ocean in 2011~2012, acquired by the Soil Moisture and Ocean Salinity (SMOS) satellite, was processed for quality control and then analyzed for the first time in terms of dynamic process factors that may impact salinity retrieval. The neural network method was introduced to improve the quality-controlled salinity data of the same time period. It was found that precipitation and surface wave induced by precipitation can increase salinity error substantially in a negative trend. Sea surface roughness caused by wind also increased salinity errors. There was a weak positive correlation between wind speed and salinity error. Changes in sea surface temperature had little effect on salinity retrieval. Considering rainfall, wind speed and other major marine dynamic processes, the neural network method was used to revise sea surface salinity data of the equatorial Pacific in December 2011. The results showed that the RMS (root mean square) of salinity reduced from 0.383 7 to 0.244 1. It was found that not only was the salinity error caused by precipitation and other dynamic processes eliminated, but also the high salinity tongue in the equatorial Pacific was revealed, which the Level-3 data failed to.