贻贝养殖区悬浮颗粒物浓度的卫星遥感反演研究  

Remote sensing estimation of total suspended matter concentration in the mussel culture area

作　　者：刘彬宇 孟家羽 王胜强 成洺萱 孙德勇 张秀梅 许永久 LIU Binyu;MENG Jiayu;WANG Shengqiang;CHENG Mingxuan;SUN Deyong;ZHANG Xiumei;XU Yongjiu(School of Marine Sciences,Nanjing University of Information Science and Technology,Nanjing 210044,China;Key Laboratory of Coastal Zone Exploitation and Protection,Ministry of Natural Resources,Nanjing 210024,China;School of Fishery,Zhejiang Ocean University,Zhoushan 316022,China)

机构地区：[1]南京信息工程大学海洋科学学院,江苏南京210044 [2]自然资源部海岸带开发与保护重点实验室,江苏南京210024 [3]浙江海洋大学水产学院,浙江舟山316022

出　　处：《海洋环境科学》2024年第1期152-160,共9页Marine Environmental Science

基　　金：国家自然科学基金项目(42176181,42176179,42106176);江苏省基础研究计划(自然科学基金)项目(BK20211289,BK20210667);遥感科学国家重点实验室开放基金项目(OFSLRSS202103);自然资源部海岸带开发与保护重点实验室开放基金项目(2021CZEPK02);浙江省自然科学基金资助项目(LY20C030004);大学生创新创业训练计划项目(202210300021Z)。

摘　　要：悬浮颗粒物(total suspended matter,TSM)是重要的水环境参数,影响着海水的透明度和初级生产力,因此总悬浮颗粒物的监测对于海洋牧场环境的评价具有重要意义。卫星遥感技术具有显著的时空观测优势,但目前尚无专门针对海洋牧场小尺度海域的TSM遥感产品。本研究以浙江嵊泗枸杞岛贻贝养殖区为研究对象,基于春、夏、秋、冬4个季节的调查实测数据,建立了面向Landsat-8卫星遥感影像的TSM浓度定量反演模型。验证结果表明,反演模型具有良好的估算精度,决定系数R2为0.72,均方根误差为6.59 g/m^(3),绝对偏差为0.72 g/m^(3),平均绝对百分误差为29.8%;进一步将其用于2021-2022年春、夏、秋、冬4个季节的Landsat-8影像,反演了贻贝养殖区及毗邻海域的TSM浓度遥感产品,分析了其时空变化特征。Total suspended matter(TSM)is an important water constituent,influencing seawater’s transparency and primary productivity.Therefore,monitoring TSM concentration is of great significance for evaluating marine ranch environments.Satellite remote sensing has significant advantages in terms of spatial and temporal observations.However,the current TSM product of ocean color remote sensing is for large-scale regions,lacking specific products for small-scale regions,like the marine ranch.In this study,we focused on the mussel culture area around Gouqi island in Shengsi,Zhejiang province,and conducted four observation cruises during different seasons.Based on the in situ data,we proposed a remote sensing model for deriving TSM concentration from the Landsat-8 satellite image with a high spatial resolution(30 m).The model validation showed a good performance,with determination coefficient,root-mean-square error,bias,and mean absolute percentage error of 0.72,6.59 g/m3,0.72 g/m3 and 29.8%,respectively.Subsequently,the proposed model was applied to generate TSM products in the mussel culture area using Landsat-8 satellite images in different seasons,and the spatial and temporal variations of TSM were further studied.

关 键 词：贻贝养殖区 悬浮颗粒物浓度 遥感反演模型 Landsat-8 OLI卫星影像 

分 类 号：X87[环境科学与工程—环境工程]