福建省河流枯水径流的自然地理环境影响研究

Multiple Regression Analysis for the Relation between Low Flow and Basin Characteristics for Fujian Province

选取福建省内22个受人类影响和自然条件改变均较小的水文站控制流域,应用3S技术,对各流域的面积、河长、河流平均坡度、形状系数、植被覆盖率、流域平均坡度、平均高程、年均降水量等地理环境要素进行了定量分析;对22个水文站的年最小月流量及多年平均值、变差系数和年极值比进行了统计分析;应用多元回归模型,建立了多年平均年最小月流量与地理环境要素间的关系。研究表明,福建省河流枯水径流量年际变化不大,枯水径流量与流域面积和流域平均坡度显著相关,复相关系数达到0.994;该模型可用于福建省类似条件无资料地区的枯水径流研究。

The study of the low flow characteristics is of great significance to the allocations of water resources and waste loading permits for stream. The low-flow statistics for a stream site can be determined in a variety of ways, depending upon whether the site is coincident with or very close to at least one stream flow-gauging station. For gauged sites, low-flow statistics are determined directly from the statistical frequency analyses. Determining low flows for ungauged streams tends to be a more challenging problem than that for gauged streams. In cases where the site is ungauged, regression analysis is one of the several well known methods that can be used. Regression analysis uses data from gauging stations believed pertinent to the ungauged site. The regression equations often use meteorologic and physiographic variables such as mean annual precipitation and contributing drainage area. Regression equations to estimate low-flow statistics for ungauged site are more commonly used abroad than at home. It is still not presently available for Fujian Province. Twenty-two basins with gauged site in Fujian Province, where the physiographical conditions affected by human activities can be neglected, have been selected. The basin characteristics including channel length and slope, basin area, shape factor, average slope, average elevation, green index and mean annual precipitation have been determined with the 3S technology. For each gauged site, the minimum monthly （30 continuous days） discharges for each year and their mean value, variation coefficients and the ratios of the maximum value to the minimum one have been calculated. The relation between the annual minimum mean discharge and the basin characteristics has been developed using multiple regression analysis. The results showed that the annual variation of the annual minimum mean discharge was small and the regression equation for the low flow included the basin area and average basin slope with a higher R^2 of 0. 994, which is applicable to the estimation of low flow in Fujian Province.