鄱阳湖表层悬沙粒度遥感反演研究

发布时间：2018-03-29 03:33

本文选题：鄱阳湖　切入点：悬浮泥沙　出处：《江西理工大学》2015年硕士论文

【摘要】：鄱阳湖作为我国第一大淡水湖,在长江流域的生态调节中占有举足轻重的作用,同时作为国际上重要的湿地,对鄱阳湖的生态保护及监测有着重要的现实意义。悬浮泥沙作为重要的水质参数之一,是衡量水质好坏的标准之一。鄱阳湖多处水域受附近民生及采砂影响,导致水体异常浑浊,同时也对鄱阳湖的鱼类生存环境造成了破坏。遥感技术作为一种宏观的监测手段,将大面积及高时效的监测成为了可能。本文基于此,以鄱阳湖水域为研究区,通过三个时期的实测调查采样和实验室测量获得悬沙的粒度和浓度数据,进行了鄱阳湖地区光谱特征的研究以及悬沙粒度反演模型的建立。根据实测的粒度数据,对三种分布模式下的中值粒径、平均粒径和众数粒径进行了分析,中值粒径与平均粒径之间有很好的相关性;粒径大小在一定程度上受到浓度的影响,在高浓度水域,粒径与浓度呈负相关,但在低浓度区域相关性较复杂,没有一定的规律性。鄱阳湖实测光谱反射率随着悬沙浓度的升高而变大;随着悬沙浓度的增大,第一反射峰向长波方向移动,峰值移动范围575nm-605nm,且775nm附近的第二反射峰逐渐升高,趋近于第一反射峰。浑浊区反射率随粒径变大而升高,清水区反射率随粒径变大而降低,粒径与反射率的关系在不同泥沙浓度水域呈现出不同的趋势。本文对不同分布模式下的各粒度参数与遥感反射率的相关性进行了探讨分析,发现在浑浊区悬沙粒度与反射率呈负相关,清水区呈正相关;结合环境卫星HJ-1A/1B星CCD影像数据波段反射率的相关分析,浑浊区基于体积百分比的众数粒径与反射率相关性较好,相关系数-0.6左右,清水区基于数量百分比的中值粒径与反射率很强的正相关,相关系数达0.8;对影像单波段、波段平均、波段比反射率与粒度相关性分析,找到较优的反演模式,建立悬沙粒度反演模型,浑浊区以幂函数、清水区以多项式函数关系模型拟合度最高,决定系数R2分别为0.4774和0.9102。在对研究区域调研准同步遥感影像预处理的基础上,将反演模型进行了应用,得到了悬沙粒度的分布图,并对年内四个季节粒度分布的变化进行了分析。
[Abstract]:Poyang Lake, as the largest freshwater lake in China, plays an important role in the ecological regulation of the Yangtze River Basin and is an important wetland in the world. As one of the important water quality parameters, suspended sediment is one of the criteria to measure the quality of water quality. Many areas of Poyang Lake are affected by people's livelihood and sand mining nearby. As a kind of macroscopical monitoring method, remote sensing technology makes it possible to monitor large area and high time. In the study area of Poyang Lake, the particle size and concentration data of suspended sediment were obtained by sampling and laboratory measurements in three periods. The spectral characteristics of Poyang Lake region and the establishment of inversion model of suspended sediment particle size are studied. According to the measured particle size data, the median particle size, mean particle size and mode particle size under three distribution modes are analyzed. There is a good correlation between the median particle size and the average particle size, the particle size is affected by the concentration to some extent, in the high concentration area, the particle size is negatively correlated with the concentration, but the correlation is more complex in the low concentration area. The measured spectral reflectance of Poyang Lake increases with the increase of suspended sediment concentration, the first reflection peak moves to the direction of long wave with the increase of suspended sediment concentration, and the peak value moves in the range of 575nm-605nm, and the second reflection peak near 775nm increases gradually. The reflectivity of turbidity region increases with the increase of particle size, while the reflectivity of clear water region decreases with the increase of particle size. The relationship between particle size and reflectivity shows different trends in different sediment concentration waters. The correlation between particle size parameters and remote sensing reflectivity under different distribution patterns is discussed in this paper. It was found that there was a negative correlation between suspended sediment granularity and reflectivity in turbid region, and a positive correlation between reflectance in clear water region, and a good correlation between reflectance and bulk particle size based on volume percentage in turbid region, combining with correlation analysis of reflectivity in CCD image of environmental satellite HJ-1A/1B satellite. The correlation coefficient is about -0.6, and the correlation coefficient between the median particle size and reflectivity in clear water region based on the quantity percentage is very strong, and the correlation coefficient is 0.8. For the single band, band average, band specific reflectivity and particle size correlation analysis, a better inversion model is found. The inversion model of suspended sediment particle size is established. The power function is used in the turbid area and the polynomial function is used in the clear water area. The determination coefficient R2 is 0.4774 and 0.9102, respectively. Based on the preprocessing of quasi-synchronous remote sensing image, The inversion model is applied to obtain the distribution map of suspended sand granularity, and the variation of grain size distribution in four seasons of the year is analyzed.
【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：X524;X87

【参考文献】