Abstract:
A copula is a function that joins or 'couples' multivariate distribution function to their one dimensional marginal distribution functions. The strong theoretical background of copula draws the attention of researchers in recent years in many application fields, including hydrology and water resources. In this paper, a brief introduction of copula is presented. A new methodology is proposed for uncertainty quantification based on the theory of copula, which is applicable to any distributional form of hydrologic time series. Thus, basic requirement of normality, as in the case of many hydrologic models, can be relaxed which is very important in its own right. The proposed methodology is explained in the context of rainfall-runoff modeling. The proposed methodology is shown to capture and provide the information of uncertainty associated with prediction. The proposed methodology is shown to be promising and, being general, can be applied to any other modeling approach.
