Abstract

CAMPOS-ARANDA, Daniel Francisco. Correction of design floods due to hydrological uncertainty. Tecnol. cienc. agua [online]. 2020, vol.11, n.6, pp.400-425.  Epub June 15, 2024. ISSN 2007-2422.  https://doi.org/10.24850/j-tyca-2020-06-10.

The planning, design and/or hydrological inspection of hydraulic works is based on what are called Design Floods (DF) or maximum river flows associated with low probabilities of exceedance. DFs are estimated using Frequency Analysis (FA), a statistical method that ignores the hydrological uncertainties involved. When such indecisions are taken into account, DFs are not unique magnitudes but a range of values, and therefore, the design and inspection of hydraulic infrastructure becomes an indeterminate process. DFs can be corrected, according to hydrological uncertainty, using a simple practical procedure developed by Botto, Ganora, Claps and Laio (2017). The method is based on a correction factor ($\widehat{y}$), which modifies the DFs obtained with FA. The value of $\widehat{y}$ is obtained with several empirical equations that are a function of the size of the available records of annual maximum flows and of the return period in years or the inverse of the exceedance probability. There are five equations of $\widehat{y}$, one for each of the most used probability distribution functions (PDF) in the FA, these are: Log-Pearson type III, General of Extreme Values, Generalized Logistic, Log-Normal and Pearson type III. The corrective method was applied to seven flood registers, after objective selection of the most appropriate PDF, according to their L moment ratios. We obtained that the corrective factors ($\widehat{y}$) vary in the 50-year Tr from 1% to 5.6%, while in the 1000-year Tr corrections fluctuated between 14.9% and 43.9%.

Keywords : frequency analysis; design floods; hydrological uncertainty correction; L moments; L moment ratios; LP3; GEV; LOG; LN3 and PT3 distributions; minimum absolute distance; standard error of fit.