摘要
采用逐步回归法分析典型城市降雨径流管理模型(SWMM)水文参数的全局灵敏度,为模型参数的有效识别提供参考.结果表明,汇水区面积对总产流起决定性作用.在雨强较小(10.5mm)的情况下,透水区参数灵敏度很小,可在参数识别中设为经验值;在较强降雨(52.5mm)情况下,管道曼宁系数是决定峰值流量与峰值发生时间的关键参数.减小汇水区面积的不确定性可提高其他参数的灵敏度,有利于参数的有效识别.
Stepwise regression analysis approach was used to assess the global sensitivity of the hydrological parameters of storm water management model （SWMM） in this study. The catchment area played a dominant role in determining surface runoff. When precipitation was low（10.5mm）, the parameter in pervious zone showed very low sensitivity, indicating that those parameter could be set to empirical values. When precipitation was high（52.5mm）, roughness of conduit was the most sensitive parameter to peak flow and peak time. Reduction of the catchment area could increase the sensitivity of other parameters, providing better condition for parameter calibration.
出处
《中国环境科学》
EI
CAS
CSCD
北大核心
2008年第8期725-729,共5页
China Environmental Science
基金
国家“973”项目(2006CB403407)
国家自然科学基金资助项目(50778098/E080403)
