南海北部叶绿素a浓度遥感反演算法评估与COCTS反演算法构建  

Evaluation of Remote Sensing Inversion Algorithm and COCTS Inversion Algorithm for Chlorophyll a Concentration in the North South China Sea

作　　者：李芝凤[1] 韩冰[1] 贾迪 朱建华[1] LI Zhifeng;HAN Bing;JIA Di;ZHU Jianhua(National Ocean Technology Center,Tianjin 300112,China)

机构地区：[1]国家海洋技术中心,天津300112

出　　处：《海洋技术学报》2022年第6期1-7,共7页Journal of Ocean Technology

基　　金：国家重点研发计划资助项目(2016YFC1400903)。

摘　　要：本文利用2018年9—10月和2020年7—8月在南海北部海域获取的叶绿素a浓度和遥感反射率现场实测数据,对3种叶绿素a浓度标准算法(OC2v4、OC3M和OC4v4)及3种区域算法(OC4v4_TP、OC2_X、INOC3)进行了模型精度评估;根据我国自主海洋水色卫星(HY-1C和HY-1D卫星)搭载的水色水温扫描仪(COCTS)的波段设置,构建了适用于我国南海北部海域的两波段、三波段、四波段和五波段比值的叶绿素a浓度反演算法。研究结果表明:标准算法的反演精度优于区域算法;针对HY-1C和HY-1D卫星构建的五波段比值OC5_C算法反演精度较标准算法反演精度有很大提高,叶绿素a浓度反演值与实测值的平均绝对误差为0.08μg/L,平均相对误差为21.36%,可为我国海洋水色卫星遥感数据在南海及同类水体的定量化应用提供算法支撑。Based on the field measurement data of chlorophyll a concentration and remote sensing reflectance collected in the northern South China Sea from September to October,2018 and from July to August 2020,this article evaluates the model accuracy of three standard algorithms(OC2v4,OC3M and OC4v4)and of three regional algorithms(OC4v4_TP,OC2_X and INOC3)for chlorophyll a concentration.A two-band,three-band,four-band and five-band ratio inversion algorithm for chlorophyll a concentration in the northern South China Sea was established,based on the band settings of COCTS offered by China’s independent ocean color satellites,HY-1C and HY-1D.The results show that the inversion accuracy of the standard algorithm is higher than that of the regional algorithm.The inversion accuracy of OC5_C algorithm constructed for HY-1C and HY-1D satellites is greatly improved than that of the standard algorithm.The mean absolute error between the inversion value of chlorophyll a concentration and the measured value is 0.08μg/L,and the mean relative error is 21.36%.This research can provide algorithmic support for quantitative application of ocean color satellite remote sensing data in the South China Sea and similar water bodies.

关 键 词：南海北部 叶绿素A 反演算法 HY-1C HY-1D 

分 类 号：TP701[自动化与计算机技术—检测技术与自动化装置]