Author(s): Lan Zhang; Vijay P. Singh
Linked Author(s): Vijay Singh
Keywords: Flood variables; Random variable; Vine-Copula; Pseudo-MLE; SMLE
Abstract: For gauged watersheds, flood frequency analysis has been commonly done in one of three ways: (i) univariate frequency analysis of observed peak discharges at a gauged site; (ii) at-a-site bivariate frequency analysis of correlated flood variables, i. e., through the dependence of peak discharge and flood volume; flood volume and flood duration with the assumption of independence between peak discharge and flood duration; (iii) at-a-site trivariate flood frequency analysis of flood variables with the use of conventional trivariate distributions, symmetric 3-dimensional Archimedean copula, and asymmetric 3-dimensional Archimedean copula. However, for the three cases discussed above, there are some limitations as, for example: (i) the univariate analysis may not appropriately represent the flood risk; (ii) it biviariate Archimedean copula may not be extended to higher-dimensional Archimedean copula (
Year: 2013