Abstract
Abstract. This study addresses the question of the existence of a parent flood frequency distribution on a European scale. A new database of L-moment ratios of flood annual maximum series (AMS) from 4105 catchments was compiled by joining 13 national data sets. Simple exploration of the database presents the generalized extreme value (GEV) distribution as a potential pan-European flood frequency distribution, being the three-parameter statistical model that with the closest resemblance to the estimated average of the sample L-moment ratios. Additional Monte Carlo simulations show that the variability in terms of sample skewness and kurtosis present in the data is larger than in a hypothetical scenario where all the samples were drawn from a GEV model. Overall, the generalized extreme value distribution fails to represent the kurtosis dispersion, especially for the longer sample lengths and medium to high skewness values, and therefore may be rejected in a statistical hypothesis testing framework as a single pan-European parent distribution for annual flood maxima. The results presented in this paper suggest that one single statistical model may not be able to fit the entire variety of flood processes present at a European scale, and presents an opportunity to further investigate the catchment and climatic factors controlling European flood regimes and their effects on the underlying flood frequency distributions.
Highlights
The first step for any assessment of the flooding potential or flood hazard is the estimation of the design flood associated with a given annual exceedance probability, often quoted in terms of a recurrence interval T measured in years
In a number of countries, the generalized extreme value (GEV) distribution is among the recommended choices (e.g. Austria, Germany, Italy, Spain), but a variety of two, three- or four-parameter distributions are used, including the Gumbel (GUM) distribution in Finland and Spain, the three-parameter generalized Pareto (GPA) distribution in Belgium, the three-parameter generalized logistic (GLO) distribution in the UK, or the four-parameter TCEV in Italy and Spain
Sample estimation uncertainty, for high values of L-Ck, could play a role by augmenting the variability in the observed L-moments, but the systematic underestimation of the dispersion points out to the fact that the GEV distribution alone is not complex enough to fully describe the variability of sample L-Cs and L-Ck values estimated for the FloodFreq database, being these L-moment ratios surrogates for the entire spectrum of flood generation processes occurring across Europe responsible for the diversity of shapes of flood frequency distributions
Summary
The first step for any assessment of the flooding potential or flood hazard is the estimation of the design flood associated with a given annual exceedance probability, often quoted in terms of a recurrence interval T measured in years This information is most commonly obtained using flood frequency estimation techniques based on statistical analyses of series of observed flood peak discharges. This study addresses explicitly the question of the existence of a parent flood frequency distribution on a European scale It presents the results of an assessment based on the analysis of a newly established pan-European database of an-. In the companion paper by Salinas et al (2014), the link between catchment and climate attributes and the choice from a set of potential parent regional flood frequency distributions is investigated on a subset of the database presented in this article
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
