| 引用本文: | 周冠华,杨一鹏,陈军,李京.基于叶绿素荧光峰特征的浑浊水体悬浮物浓度遥感反演.湖泊科学,2009,21(2):272-279. DOI:10.18307/2009.0217 |
| ZHOU Guanhua,YANG Yipeng,CHEN Jun,LI Jing.Inversion of total suspended matter concentration in turbid water based on the characteristic of chlorophyll fluorescence peak. J. Lake Sci.2009,21(2):272-279. DOI:10.18307/2009.0217 |
|
| |
|
|
| 本文已被：浏览 6983次 下载 5388次 |
码上扫一扫！ |
|
|
| 基于叶绿素荧光峰特征的浑浊水体悬浮物浓度遥感反演 |
|
周冠华1,2, 杨一鹏3, 陈军4,5, 李京2
|
|
1.北京师范大学资源学院, 北京 100875;2.民政部/教育部减灾与应急管理研究院, 北京 100875;3.中国环境监测总站, 北京 100029;4.中国地质大学(北京), 北京 100083;5.青岛海洋地质研究所, 青岛 266071
|
|
| 摘要: |
| 内陆水体光学特性复杂,其水质参数遥感反演是当前环境遥感研究的热点与难点.2004年10月在太湖实测了67个站点的遥感反射率与相应站点水质参数浓度,通过对水体反射率光谱的分析发现,秋季太湖悬浮物主导了水体光学特性,叶绿素荧光峰的特征主要体现为悬浮物浓度的变化.据此建立了基于水面实测高光谱遥感反射率数据的叶绿素荧光峰特征与悬浮物浓度之间的拟合关系,发现二者具有很好的响应关系.具体分析了叶绿素荧光峰绝对高度、基线高度、归一化高度(分别归一化到560nm附近最大反射率波段与近红外810nm附近最大反射率波段)及荧光峰积分面积(包括积分总面积、基线以下面积与基线以上面积)等几种光谱特征与悬浮物浓度之间的关系,其相关系数(R2)分别为0.8822、0.7483、0.8901、0.8547、0.8927、0.8877、0.8632,平均相对误差分别为27.25%、41.03%、27.11%、25.75%、24.91%、25.47%、27.54%,总体反演精度较高,其中总积分面积法效果最好,基线高度法效果最差,而叶绿素荧光峰波段的位移与悬浮物浓度之间不存在明显的相关性.研究结果表明叶绿素荧光峰特征在浑浊内陆水体悬浮物浓度信息提取中具有很好的应用前景,该方法可为浑浊的二类水体悬浮物遥感反演提供了一个新思路. |
| 关键词: 水质遥感 荧光峰 悬浮物 太湖 高光谱 |
| DOI：10.18307/2009.0217 |
| 分类号: |
| 基金项目:国家科技支撑计划课题(2008BAC34B03);国家自然科学基金(40871168);中国博士后科学基金(20080440313)联合资助 |
|
| Inversion of total suspended matter concentration in turbid water based on the characteristic of chlorophyll fluorescence peak |
|
ZHOU Guanhua1,2, YANG Yipeng3, CHEN Jun4,5, LI Jing2
|
|
1.College of Resources Sciences and Technology, Beijing Normal University, Beijing 100875, P. R. China;2.Academy of Disaster Reduction and Emergency Management Ministry of Civil Affairs & Ministry of Education, Beijing 100875, P. R. China;3.Environmental Monitoring Center of China, Beijing 100029, P. R. China;4.China University of Geosciences, Beijing 100083, P. R. China;5.Qingdao Marine Geosciences Institute, Qingdao 266071, P. R. China
|
| Abstract: |
| The optical property of inland water is very complex, and inversion of water components concentration of inland water body is hotspot and difficulty problem in water quality remote sensing. The remote sensing reflectance and water qualities of 67 stations were acquired in the Lake Taihu in October, 2004. Basing on the characteristic spectral analysis, it was obvious that the water body spectral character in Lake Taihu in autumn was dominated by the total suspended matter (TSM). The properties of the chlorophyll fluorescence peak mainly reflect the variation of the concentration of TSM. Hereby, we established the relation between the properties of the chlorophyll fluorescence peak and the concentration of TSM based on in-situ remote sensing reflectance with 1nm bandwith and water quality measurements. There were good response relationships between them. Making a concrete analysis ofthe correlation relationship between the absolute height of fluoresce peak, the height above the reference baseline, the normalized height referred to the maximum reflectance about 560nm in green band and 810nm in infrared band, the integral area (including the total integral area, the integral area between the spectral curve and the reference baseline, the integral area below the reference baseline and the wavelength axis) and the concentration of TSM. Their correlation coefficients were 0.8822, 0.7483, 0.8901, 0.8547, 0.8927, 0.8877 and 0.8632 respectively, and the average relative error were 27.25%, 41.03%, 27.11%, 25.75%, 24.91%, 25.47% and 27.54% respectively. The inversion accuracy of the total integral area method is obvious better than the reference baseline height method. No outstanding correlation relationship between chlorophyll concentration and fluorescence peak position were found based on reflectance spectra. The results show that chlorophyll fluorescence properties are effective in detecting TSM concentration in turbid inland water body, which provide a new optional sensitive band to retrieval TSM concentration in complex waters. |
| Key words: Water quality remote sensing fluorescence peak suspended matter Lake Taihu Hyperspectral remote sensing |
|
|
|
|
