水文模型不确定性分析的多准则似然判据GLUE方法

Uncertainty Analysis of Hydrological Model Using Multi-criteria Likelihood Measure Within the GLUE Framework

水文模型不确定性分析GLUE(Generalized Likelihood Uncertainty Estimation)方法常用似然判据Nash-Sut-cliffe系数(确定性系数),这种判别标准侧重对整体过程误差的估计,对点状况如洪峰的表征不够,不能准确地反映模型的不确定性情况。本研究基于GLUE方法,建立了多准则似然判据,应用北方水库常用的大伙房模型,对碧流河水库洪水预报的不确定性进行研究,并给出预报误差分析。结果表明,将常用的确定性系数似然判据扩展为洪峰误差、洪量、峰现时间、确定性系数4个目标的多准则似然判据,能更好地反映模型的实际不确定情况,对模型参数的率定和不确定性研究具有重要意义。

GLUE(Generalized Likelihood uncertainty Estimation) is a very widely used methodology for uncertainty analysis of hydrological model.How to select an appropriate likelihood measure is still an open question so far.Usually the Nash-Sutcliffe coefficient is a traditional likelihood measure,but this measure focuses on the global performance without a special assessment in point condition especially for flood peak.A multi-criteria likelihood measure is presented within GLUE methodology,which consists of peak fo...