Peaks over threshold method: One stochastic model for flood volumes

Abstract

Peaks-over-threshold method (POT) is the way for analysis of the stochastic structure of extreme hydrological events. For a chosen base flow and for characteristic value obtained form partial hydrographs, the POT method encompasses detection of discrete probability distribution of number of events in chosen time interval and continuous distribution of the exceedance values (peaks). This article presents a stochastic model for the analysis of the base flow exceedance volumes and accompanied cycle times between the ends of the successive exceedaence events. The model is based on the Markov's discreteds model principles and the assumptions about the form of the process intensity functions. The number of occurrence discrete distributions are discussed according to chosen forms of the time and volume intensity functions. The continuous distributions of the base flow exceedance characteristic values are modelled for the base series of values and their aggregation. The distribution of the maximum exceedance volume over the base flow in chosen time interval is formulated. The article presents an application of the suggested procedures on the mean daily flows hydrographs from the Bezdan gauging station on the Danube river.

Keywords

• POT method
• flood volumes
• stochastic model