下垫面要素变化对洪水影响的多元统计分析

为研究下垫面变化对流域洪峰模数、径流深等的影响程度，选取大清河水系的紫荆关、阜平、漫水河、城头会和张坊五个典型子流域为研究对象,利用卫星遥感资料提取五个子流域1980、1990、2000年的土地利用情况，分别对洪水按重现期分为大于10年一遇和小于10年一遇两个量级下流域的洪峰模数、径流深及其影响要素进行多元统计分析,利用得出的回归模型分析下垫面要素变化对洪水的影响程度。结果表明，对大于10年一遇的洪水，径流深和洪峰模数主要受林地变化的影响，影响贡献比例分别为0%~65%和0%~40%；而对小于10年一遇的洪水，径流深主要受草地变化的影响，影响贡献比例为10%~120%，洪峰模数则受草地和林地变化的共同影响，影响贡献比例分别为0%~0.8%和0%~2.5%。

Multivariate Statistical Analysis of Influence of Underlying Surface Change on Flood

In order to study the influence of underlying surface change on peak flood modulus and runoff depth, five typical sub-basins (Zijingguan, Fuping, Manshuihe, Chengtouhui and Zhangfang) of Daqing river system are taken as a study case. The land utilization of five typical sub-basins in 1980, 1990 and 2000 are extracted by using satellite remote sensing data. According to the flood return period (larger than 10 years and less than 10 years), multivariate statistical analysis is made on flood peak modulus, runoff depth and its influencing factors. Regression model is used to analyze the influence of underlying surface change on flood. The results show that for larger than 10 years return period flood, the runoff and the flood peak modulus are mainly influenced by change of forest area, and the influence proportion is 0%-65% and 0%-40%; for less than 10 years return period flood, runoff depth is mainly influenced by change of grass area, and the influence proportion is 10%-120%; while the flood peak modulus is influenced by change of forest and grass area, and the influence proportion is 0%-0.8% and 0%-2.5%.