降水丰枯及下垫面变化对滦河干流径流量的影响

Influence of high and low precipitations and underlying surface changes on runoff in the Luan River mainstream

在收集滦河干流上游流域1961—2011年51年间的月降水和月径流实测资料的基础上,运用降水–径流双累积曲线和Kolmogorov-Smirnov双样本检验,将该流域的降水–径流回归关系划分为1961—1983年、1984—1999年和2000—2011年三个阶段,并分别建立了三个阶段丰水期和枯水期的降水–径流多元回归模型。三个阶段的多元回归模型均通过显著性检验,对实测年径流拟合的平均残差均小于15%,决定系数均大于68%。采用所建模型,定量分析了三道河子水文站控制流域内降水丰枯变化对径流的影响程度和流域下垫面改变对径流的影响程度。结果表明:在1961—1983年的下垫面条件下,相对于1961—1983年间的降水丰枯特征,1984—1999年、2000—2011年间降水的丰枯变化使得流域径流分别增加了8.08%和减少了25.34%;在1961—1983年的降水条件下,相对于1961—1983年的下垫面条件,1984—1999年、2000—2011年间的下垫面的改变使该流域径流分别减少了19.40%和43.16%。这对潘家口水库的入库流量势必产生一定影响。

On the basis of monthly precipitation and runoff observations in the upper reaches of the Luan River basin over the period of 1961-2011, rainfall-runoff regression relationships have been studied by dividing this study period into three smaller periods （1961-1983, 1984-1999 and 2000-2011）. For each period, two rainfall-runoff multivariate regression models have been developed for high and low flow periods, respectively, using a rainfall-runoff double mass plot method and the two-sample Kolmogorov-Smimov test. All of the regression models for the three periods have passed the significance tests, with the average runoff residuals being less than 15% and determination coefficients greater than 68%. Based on these regression models, we have quantitatively analyzed the impact of changes in both precipitation and underlying surface on the runoff in the control catchment of the Sandaohezi hydrological station. The results reveal that under the condition of underlying surface over 1961-1983, the precipitation during 1984-1999 and 2000-2011 can generate 8.08% more and 25.34% less runoff than 1961-1983 respectively; under the rainfall condition of 1961-1983, the change in underlying surface over 1984-1999 and 2000-2011 can generate 19.4% and 43.16% less runoff than 1961-1983 respectively. This would affect the inflow discharge of the Panjiakou reservoir to some extent.