摘要
以东苕溪流域和息县流域为例,运用基于贝叶斯理论的GLUE方法对新安江模型参数不确定性进行分析评价.结果表明,两个流域都存在大量"等效性"参数,不同的参数组能模拟出相同的效果,验证了GLUE方法的重要观点:模型模拟结果的好坏是由参数组合决定的.根据参数与似然值散点分布图,可将研究的7个参数分为不敏感参数(KG,KI,CG,CI)、敏感参数(CS)和区域敏感参数(K,SM)3类,通过计算90%置信度的不确定范围,可实现洪水概率预测的目的.
In this study,the uncertainties of the Xin'anjiang model parameters were analyzed with the GLUE method based on the Bayesian theory,through application to the Dongtiaoxi and Xixian catchments.The study results show that there is a large number of equivalent parameters for both catchments,verifying the argument in the GLUE method that the simulation results are determined by the combination of the parameters.According to scatter plots of the likelihood values of the behavioral parameters,the Xin'anjiang model parameters can be classified into three groups: the non-sensitive parameters(KG,KI,CG,and CI),the sensitive parameter(CS),and the regionally sensitive parameters(K and SM).The probability prediction for floods can be made through the calculation of the uncertainty ranges at a 90% confidence level.
出处
《河海大学学报(自然科学版)》
CAS
CSCD
北大核心
2011年第6期618-622,共5页
Journal of Hohai University(Natural Sciences)
基金
国家自然科学基金(50479017)
教育部博士学科点专项基金(2009009411005)
教育部长江学者和创新团队发展计划(IRT0717)
作者简介
戴健男(1985-),男,江苏徐州人,硕士研究生,主要从事水文模型与水文预报研究.E-mail:nopning85@163.com