A physically based analytical model of flood frequency curves. Issue 17 (7th September 2016)

• Record Type: Journal Article
• Authors: Basso, S.; Schirmer, M.; Botter, G.
• Is Part Of: Geophysical research letters. Volume 43:Issue 17(2016)
• Page Start: 9070
• Page End: 9076
• Publication Date: 2016-09-07
• Subjects: flood frequency curves -- high flows -- river flow regimes -- physically based -- stochastic model -- data scarcity; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1002/2016GL069915
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract: Predicting magnitude and frequency of floods is a key issue in hydrology, with implications in many fields ranging from river science and geomorphology to the insurance industry. In this paper, a novel physically based approach is proposed to estimate the recurrence intervals of seasonal flow maxima. The method links the extremal distribution of streamflows to the stochastic dynamics of daily discharge, providing an analytical expression of the seasonal flood frequency curve. The parameters involved in the formulation embody climate and landscape attributes of the contributing catchment and can be estimated from daily rainfall and streamflow data. Only one parameter, which is linked to the antecedent wetness condition in the watershed, needs to be calibrated on the observed maxima. The performance of the method is discussed through a set of applications in four rivers featuring heterogeneous daily flow regimes. The model provides reliable estimates of seasonal maximum flows in different climatic settings and is able to capture diverse shapes of flood frequency curves emerging in erratic and persistent flow regimes. The proposed method exploits experimental information on the full range of discharges experienced by rivers. As a consequence, model performances do not deteriorate when the magnitude of events with return times longer than the available sample size is estimated. The approach provides a framework for the prediction of floods based on short data series of rainfall and daily streamflows that may be especially valuable in data scarce regions of the world. Key Points: Analytic expression of seasonal flood frequency curve based on physically based parameters that embody information on daily flow dynamics Daily flow regime (flow variability) controls the observed shape of flood frequency curves, including step changes for long return times The approach is suited to assess magnitude of events featured by return period longer than the sample size