基于贝叶斯理论的TOPMODEL参数不确定性分析

Bayesian theory is applied to the parameter uncertainty analysis of TOPMODEL in this paper. The sensitive parameters of TOPMODEL are determined firstly by the Monte Carlo Method. The results show that the parameters SZM and RV are sensitive, any a slight change of them will influence the simulated results; Moreover, the sampling distribution of flow process at the basin outlet can be obtained by the later distribution sampling, which can be calculated with the algorithm of Markov chain Monte Carlo (MCMC). And the paper analyzes the parameter uncertainty and its influence on forecast based on MCMC method. The paper also provides the analysis process of the parameter uncertainty, and constructs the 90% confidence interval with the sampling distribution of the model forecast results.