文章摘要
丹江口水库总氮、氨氮遥感反演及时空变化研究
Spatiotemporal monitoring of total nitrogen and ammonia nitrogen in Danjiangkou reservoir
投稿时间:2021-04-02  
DOI:10.13254/j.jare.2021.0195
中文关键词: 丹江口水库,Sentinel-2,BP神经网络,总氮,氨氮,时空变化
英文关键词: Danjiangkou reservoir, Sentinel-2, BP neural network, total nitrogen, ammonia nitrogen, spatiotemporal change
基金项目:国家自然科学基金(U170420041);中原科技创新领军人才项目(194200510010);河南省科技攻关项目(192102110086);农业农村部耕地利用遥感重点实验室开放课题(2020LCLU002);河南省软科学研究计划项目(192400410076);河南理工大学博士基金项目(B2017-12)
作者单位E-mail
刘轩 河南理工大学资源环境学院, 河南 焦作 454000
河南理工大学测绘与国土信息工程学院, 河南 焦作 454000 
 
赵同谦 河南理工大学资源环境学院, 河南 焦作 454000  
蔡太义 河南理工大学测绘与国土信息工程学院, 河南 焦作 454000 caity2008@hpu.edu.cn 
肖春艳 河南理工大学资源环境学院, 河南 焦作 454000  
陈晓舒 河南理工大学资源环境学院, 河南 焦作 454000  
张文静 河南理工大学资源环境学院, 河南 焦作 454000  
摘要点击次数: 450
全文下载次数: 160
中文摘要:
      为定量反演丹江口水库水质指标含量,明晰水质指标的时空分布特征、迁移转化规律,以南水北调中线工程水源地丹江口水库为研究对象,根据哨兵2号卫星(Sentinel-2)遥感影像不同波段组合的反射率,结合2016年2月的采样点总氮(TN)与氨氮(NH3-N)水质监测数据建立BP神经网络模型,反演2016-2020年TN与NH3-N含量,以此分析库区TN与NH3-N含量的时空变化特征,并探析其变化的影响因素。结果表明:构建的BP神经网络模型中TN和NH3-N的拟合精度均较高,R2分别为0.863和0.877,适用于丹江口水库TN和NH3-N遥感反演研究。2016-2020年丹江口水库水质整体呈向好趋势,NH3-N含量保持Ⅰ类水质标准,而TN含量在Ⅲ类与Ⅳ类水质标准之间。研究表明,利用Sentinel-2影像波段所建立的BP神经网络模型,适用于TN与NH3-N含量的遥感反演,以此分析不同季节适合的反演模型,可以为大型湖泊水生态环境改善及水质监管提供技术支撑。
英文摘要:
      The quantitative inversion of water quality index content could clarify the spatiotemporal distribution characteristics, migration, and transformation laws of water quality indexes. This study focuses on the Danjiangkou reservoir, the source of the middle route of the South-to-North Water Diversion Project. Based on the reflectance of different band combinations of Sentinel-2 remote sensing images, combined with the total nitrogen(TN) and ammonia nitrogen(NH3-N) water quality monitoring data of the sampling points in February 2016, we established a BP neural network model to invert the TN and NH3-N contents from 2016 to 2020 in order to analyze the characteristics of spatiotemporal changes of the TN and NH3-N contents in the reservoir area and to explore the factors affecting the changes. Our results showed that the fitting accuracy of TN and NH3-N in the constructed BP neural network model was relatively high, R2=0.863 and 0.877, respectively, which was suitable for remote sensing inversion research of TN and NH3-N in the Danjiangkou reservoir. The water quality of the Danjiangkou reservoir had shown an overall improving trend from 2016 to 2020. The NH3-N content had been in line with Class Ⅰ water quality standards, while the TN concentration had been between Class Ⅲ and Ⅳ water quality standards. The results show that the BP neural network model based on sentinel-2 MSI image bands is suitable for the remote sensing inversion of the TN and NH3-N concentration. It could provide technical support for the improvement of the water ecological environment and water quality supervision of large lakes.
HTML   查看全文   查看/发表评论  下载PDF阅读器
关闭