基于光谱混合分解的流域不透水面提取及其动态分析——以于桥水库为例

不透水面不仅是城市非点源污染的主要来源,还是流域生态环境变化的主要因素之一。不透水面的数量、位置、几何形状、分布格局以及透水率与不透水率的比值,均影响着流域的水文环境,因此成为研究热点。本文以天津于桥水库流域为例,综合遥感(RS)与地理信息系统(GIS)技术,从流域尺度上研究1984~2013年间不透水面覆盖度的变化。在ENVI 5.1软件支持下,利用遥感影像获取1984,1994,2004和2013年4个时相的不透水面信息。采用修正后的归一化水体指数剔除水体信息,排除水体对不透水面提取精度的影响。运用线性光谱混合分析法(Linear Spectral Mixture Analysis,LSMA),提取流域不透水面覆盖度。结果表明:流域内不透水面覆盖度大多集中在1~5级,植被覆盖程度较高。近30年间不透水面比例逐年增加,2013年比1984年增加了2.802%,呈线性增长。中等分辨率的遥感影像适合流域尺度的不透水面提取的结果可作为流域水文及规划管理的重要基础性数据。

Dynamic Analysis of Impervious Surface Based on Spectral Mixture:Taking the Yuqiao Watershed as an Example

Imperviousness in watershed is a key index to measure urban development status which exerts an important impact both on eco-hydrological process and spatio-temporal pattern.The magnitude, location, geometry and spatial pattern of impervious sur-faces, and the previous–impervious ratio in a watershed have hydrological impacts.In this study, we adopt the technology of remote sensing(RS)and geographical information system(GIS) to analyze the change of impervious surface coverage from 1984 to 2013 in the Yuqiao watershed.Based on the ENVI5.1software , we extract basic impervious surface information form remote sensing image of four time periods which are 1984,1994,2004 and 2013.We adopt modified normalized difference water index( MNDWI) to exclude water information and eliminate the effect of water on the impervious surface extraction to confirm our accuracy.Linear spectral mixture anal-ysis is used to generate impervious surface area.Results show that, impervious surface area increased and became more connected o-ver time.Impervious surface coverage mostly concentrates in 1-5 classes when the average of ISA shows a linear growth.Medium-resolution remote sensing images for basin-scale extraction of impervious surfaces, the results can provide important basic data for the hydrology and watershed planning and management.