Assessment of the risk of inland flooding in a tidal sluice regulated catchment using multi-variate statistical techniques

In a catchment discharging to the sea through a tidal sluice a two step statistical analysis is carried out. In the first step the joint distribution of peak values for the runoff at two upstream locations and for the sea water level at the outlet is analysed. The second step consists of estimating the joint distributions of excess values of peak, volume and duration for the variables at the three locations. In order to select appropriate joint distributions, a multi-variate analysis of the extreme values is carried out in each step using copula functions from the Archimedean family. The main goal of the study is to assess the risk of inland flooding at arbitrary locations, which is addressed by combining the results of the multi-dimensional statistical analysis with a river flow simulation model. Data from the Vidaa River in Denmark are used in this study.