南四湖水体浊度高光谱定量反演模型.doc

文档介绍：南四湖水体浊度高光谱定量反演模型
摘要:根据实测高光谱数据和同步水质检测数据,通过识别反演水体浊度的高光谱敏感波段,分别建立了原始光谱反射率模型、归一化光谱反射率模型、波段比值模型和一阶微分模型,并进行了验证。结果表明,4种模型的决定系数R2均在075以上,其中825 nm处归一化光谱反射率模型和波段比值模型估测浊度精度较高,均方根误差RMSE分别为2879 NTU和2423 NTU。研究结果可以为南四湖水体浊度大面积遥感监测的业务化管理提供技术支持。
关键词:南四湖;遥感;浊度;高光谱;反演
中图分类号:TP791 文献标志码:A 文章编号:16721683(2015)05088305
Quantitative inversion model of hyperspectral for turbidity in the Nansi Lake
CAO Yin1,2,YE Yuntao2,ZHANG Xiaojuan3,SHI Yubo4,JIANG Yunzhong2,ZHAO Hongli2
( of Environmental Science and Engineering,Donghua University,Shanghai 201620,China; of
Water Resources,China Institute of Water Resources and Hydropower Research,Beijing 100038,China; Affairs
Information Management Center,Water Authority,Beijing 100038; of Water Resources Management,
Ministry of Water Resources,Beijing 100053,China)
Abstract:Based on the measured hyperspectral data and synchronous water quality measurement data,through the identification of high hyperspectral bands sensitive to water turbidity,the original spectral reflectance model,normalized spectral reflectance model,band ratio model,and first derivative model were built and verified results showed that the determinative coefficients of the four models are all above ,and the normalized reflectance model in 825 nm and band ratio model of provide a higher precision to estimate water turbidity with the RMSE of NTU and NTU, the technicalsupport forthebusinessoriented management oflargearea water turbidity remote sensing monitoringin the NansiLake.
Key words:Nansi Lake;remote sensing;turbidity;hyperspectral;inversion
南四湖位于淮河流域北部,东、西、北三面承接苏、鲁、豫、皖4省32个县(市)的来水,流域面积317万km2,是我国北方最大的淡水湖,同时也是南水北调东线工程重要的水源地和调蓄湖泊,具有防洪、排涝、灌溉、供水、养殖、通航及旅游等多种功能,其水质好坏直接影响着
“一湖清水向北流”调水目标的实现。2014年以来,山东发生严重旱情,南四湖水位持续下降,甚至从六月份开始先后降至死水位和最低生态水位,水体自净能力下降,水体浑浊度增加,水质恶化,严重影响南四湖附近居民的生活和养殖业。全面监测南四湖水体水质对南水北调东线工程的顺利实施和保障南四湖周边居民的饮用水安全具有重要意义。
浊度是评价湖泊水质的一