论文部分内容阅读
水文模型结构本身的缺陷、模型输入输出误差、水文模型参数冗余及其复杂的非线性联系等,导致了流域水文模型的不确定性.基于贝叶斯理论的MCMC方法及GLUE方法近年来被广泛应用于流域水文模型的不确定性研究工作中.为比较上述2种模型不确定性分析方法的分析效果及其优劣,以位于汉江流域的牧马河流域作为研究对象,采用集总式概念性水文模型SMAR模型作为实验模型,推求其模型参数的不确定性及参数的后验分布.采用基于实测流量资料估计的置信区间可靠性作为评判标准,实验结果表明:就SMAR模型而言,MCMC方法能够更好地推求模型参数的后验分布.
Hydrological model structure defects, model input and output errors, hydrological model parameters and complex nonlinear linkages, etc., led to the hydrological model of the river basin uncertainty.Based on the Bayesian theory of MCMC method and the GLUE method in recent years Which is widely used in the study of the uncertainty of the hydrological model of the basin.In order to compare the analytical results of the above two models and the advantages and disadvantages of the method, taking the Muma River in the Hanjiang River Basin as the research object, The conceptual hydrological model SMAR model is used as the experimental model to deduce the uncertainty of the model parameters and the posterior distribution of the parameters.The reliability of the confidence interval based on the measured flow data is used as the evaluation criterion.The experimental results show that for the SMAR model, The MCMC method can better predict the posterior distribution of model parameters.