BIVARIATE FLOOD FREQUENCY ANALYSIS USING COPULAS

Abstract

In a flood hydrograph, the main variables are flood peak, volume, and duration. These three variables of floods are random in nature and are mutually correlated. The joint distributions of these variables are important in hydrologic design. In this study, the joint distribution of flood peak and volume, flood peak and duration, and flood volume and duration have been derived using the copulas. The benefit of using copulas is that it can join different distribution of different flood variables in a common marginal distribution. In this study, the copula method is also used to determine the conditional return periods by using 31 years stream flow data of Naramada River at Garudeshwar, Gujarat, India