Sentinel-2-MSI和Sentinel-3-OLCI离水反射率产品在黄河口的适用性评估  

Evaluation of applicability of Sentinel-2-MSI and Sentinel-3-OLCI water-leaving reflectance products in Yellow River Estuary

作　　者：罗亚飞 钟小僅[1] 付东洋 严立文 张翼[3] 刘一霖[4] 黄海军 张泽华[2] 祁雅莉[1] 王倩 LUO Yafei;ZHONG Xiaojin;FU Dongyang;YAN Liwen;ZHANG Yi;LIU Yilin;HUANG Haijun;ZHANG Zehua;Qi Yali;WANG Qian(Key Laboratory of Climate,Resources and Environment in Continental Shelf Sea and Deep Sea of Department of Education of Guangdong Province,College of Electronic and Information Engineering,Guangdong Ocean University,Zhanjiang 524088,China;Key Laboratory of Marine Geology and Environment,Institute of Oceanology,Chinese Academy of Sciences,College of Oceanography,University of Chinese Academy of Sciences,Qingdao 266071,China;College of Ocean Science and Engineering,Shandong University of Science and Technology,Qingdao 266590,China;College of Earth Science and Engineering,Shandong University of Science and Technology,Qingdao 266590,China)

机构地区：[1]广东海洋大学电子与信息工程学院,广东省高等学校陆架及深远海气候、资源与环境重点实验室,广东湛江524088 [2]中国科学院海洋研究所海洋地质与环境重点实验室,中国科学院大学海洋学院,山东青岛266071 [3]山东科技大学海洋科学与工程学院,山东青岛266590 [4]山东科技大学地球科学与工程学院,山东青岛266590

出　　处：《大气与环境光学学报》2023年第6期585-601,共17页Journal of Atmospheric and Environmental Optics

基　　金：国家自然科学基金(42006148,42206187,41976166);中国科学院海洋地质与环境重点实验室开放基金项目(MGE2020KG12)。

摘　　要：近岸高浊度二类水体具有复杂的海洋光学特性,不同影像源反射率产品在该类水体的适用性尚待充分论证。以黄河口水体为典型研究对象,以适用于黄河口的ACOLITEDSF算法校正的Landsat离水反射率产品为参考,通过星星匹配,对基于不同大气校正算法的Sentinel-2-MSI(S2-MSI)和Sentinel-3-OLCI(S3-OLCI)离水反射率产品在黄河口的适用性进行了评估。结果表明,在黄河口高浑浊-极度浑浊水体,iCOR算法校正的S2-MSI和S3-OLCI的离水反射率产品与参考产品的一致性高于其他算法,其次为FLAASH和Sen2Cor算法,C2RCC算法表现相对较差。iCOR、FLAASH和Sen2Cor算法除在高浑浊水体的近红外波段平均百分比相对误差EMARD超过34%外,在绿、红波段的EMARD均小于24%,Sen2Cor算法结果整体上与FLAASH的相似。iCOR、FLAASH和Sen2Cor算法随着水体浑浊程度增加,误差越小;而C2RCC算法则随着水体浑浊程度的增加,误差越大,且整体存在低估。研究结果可为高浊度二类水体大气校正的选择提供有效借鉴,并为黄河口悬沙高分辨率动态监测打下基础。The optical properties of highly turbid coastal water bodies are complicated,so the application of reflectance products based on different satellites in such kinds of water bodies is desperately needed to be fully evaluated.Taking the water body of the Yellow River Estuary(YRE)as a typical research object,the applicability of the water-leaving reflectance products corrected by different atmospheric correction algorithms for Sentinel-2-MSI(S2-MSI)and Sentinel-3-OLCI(S3-OLCI)image is evaluated based on the reference of Landsat reflectance products corrected by ACOLITE DSF atmospheric correction algorithm suitable for YRE.The results show that S2-MSI and S3-OLCI data corrected by the image correction for atmospheric effects(iCOR)are in good agreement with Landsat reflectance products for highly-extremely turbid water bodies in YRE,followed by FLAASH(Fast line-of-sight atmospheric analysis of spectral hypercubes)and Sen2Cor(Sentinel-2 correction)corrected products,while C2RCC(Case-2 regional coast colour)performs poorly.iCOR,FLAASH and Sen2Cor corrected S2-MSI and S3-OLCI reflectance products show good results in the green and red bands with the mean absolute relative difference E_(MARD) lower than 24%,except that in the near-infrared band the EMARD is higher than 34%in highly turbid waters.Among them,the FLAASH and Sen2Cor algorithms have similar results.Compared with that in highly turbid waters,the E_(MARD) is even lower for iCOR,FLAASH and Sen2Cor algorithms in extremely turbid waters.On the contrary,C2RCC algorithm performs relatively poor in highly turbid waters and even worse in extremely turbid waters,with overall underestimation of water-leaving reflectance.This research is significant for it can provide useful reference for choosing atmospheric correction methods in highly turbid waters and lay the foundation for dynamic monitoring suspended sediment in YRE with high resolution.

关 键 词：大气校正 黄河口 Sentinel-2-MSI Sentinel-3-OLCI 离水反射率 

分 类 号：P733.3[天文地球—物理海洋学]