摘要
利用国产GF-1 WFV影像和神经网络建立多光谱遥感影像水质多参数定量反演模型,能实现高效率、大范围、连续空间的水库水质多参数变化监测,探究了国产卫星影像于水质遥感反演的可行性,并为湖长制和湖泊富营养化评估提供技术支持。以广东省佛山市一中型水库为例,基于GF-1 WFV影像,利用神经网络模型建立了东风水库叶绿素a(Chl-a)、透明度(SD)、总磷(TP)、总氮(TN)、高锰酸盐指数(CODMn)5个水质参数与影像数据间的定量反演模型,5个水质参数反演模型预测值与实测值之间的决定系数R 2均达到0.8以上,平均相对误差均在40%以下,研究结果证实了国产卫星影像于水质遥感反演的可行性,研究成果可为水库的水质和富营养化监测提供参考。
This paper establishes a multi-parameter quantitative inversion model of water quality in multispectral remote sensing images by domestic GF-1 WFV image and neural network model to achieve high-efficiency,large-scale,continuous-space and multi-parameter change monitoring of reservoir water quality,explores the application feasibility of domestic satellite image inremote sensing inversion of water quality,and provide technical support for lake chief system and lake eutrophication assessment.Taking a medium-sized reservoir in Foshan City,Guangdong Province as an example,based on the GF-1 WFV image,a quantitative inversion model between 5 water quality parameters of Chl-a,SD,TP,TN and(CODMn)and image data of Dongfeng Reservoiris established with neural network models,the determination coefficient(R^2)between the predicted and measured values of the 5 water quality parameters all reached above 0.8,with the average relative errors of less than 40%.The results confirm the feasibility of domestic satellite image for remote sensing inversion of water quality,which can provide a reference for the water quality and eutrophication monitoring of the reservoir.
作者
郑炎辉
张园波
何艳虎
ZHENG Yanhui;ZHANG Yuanbo;HE Yanhu(Guangzhou Fengzeyuan Water Conservancy Technology Co.,Ltd.,Guangzhou 510663,China;Institute of Environmental&Ecological Engineering,Guangdong University of Technology,Guangzhou 510006,China;Southern Marine Science and Engineering Guangdong Laboratory(Guangzhou),Guangzhou 511458,China)
出处
《人民珠江》
2020年第7期57-62,84,共7页
Pearl River
基金
国家自然科学基金(51979043、51509127)
南方海洋科学与工程广东省实验室(广州)人才团队引进重大专项(GML2019ZD0403)。
关键词
多光谱
遥感影像
神经网络
水质反演
GF-1
WFV
multispectral
remote-sensing image
neural network
water quality inversion
GF-1 WFV
