SciELO - Scientific Electronic Library Online

vol.10 issue6Graph theory based single and multiple source water distribution network partitioningInformation system for the management of drought risk in Camagüey, Cuba author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • Have no similar articlesSimilars in SciELO


Tecnología y ciencias del agua

On-line version ISSN 2007-2422


CAMPOS-ARANDA, Daniel Francisco. Contrast of the Generalized M5 Method to estimate Extreme Predictions and PMP in 24 hours, in the state of Zacatecas, Mexico. Tecnol. cienc. agua [online]. 2019, vol.10, n.6, pp.222-242.  Epub May 15, 2020. ISSN 2007-2422.

Probable maximum precipitation (PMP) is the basis for the estimation of the probable maximum flood, with which large hydraulic works are dimensioned and hydrologically revised. There are two groups of methods to estimate PMP: meteorological and statistical. Meteorological methods are the most reliable, but require a lot of data that is usually not available. Statistical methods are much simpler and only use annual maximum daily precipitation (PMD) values. The classic method of this group is the David M. Hershfield method, published in 1961. Subsequently, in England (NERC, 1975) another statistical method was developed based on the prediction of duration 24 hours and return period (Tr) 5 years, designated M5; this approach allows predictions with various Tr. Jónas Elíasson (Elíasson, 1997; Elíasson, 2000) generalized the M5 method, in a regional technique that only requires two statistical parameters: the M5 and the coefficient of variation (Cv). In this study, the results of the generalized M5 method (MM5G) are compared with those of the Hershfield method, previously calculated based on the PMD data for 81 localities in the state of Zacatecas, Mexico. The MM5G was applied using the available values of M5 and Cv. Results allow the recommendation of the use of MM5G to estimate predictions of PMD with Tr less than 100 years, in no-data sites in the state of Zacatecas, Mexico. It is also recommended for estimations of PMP of 24 hours duration, remembering that such method underestimates less than 16.4%, with respect to the result of the Hershfield method, when the Cv is less than 0.251 and overestimates, of the order of 38.0% when the Cv exceeds 0.386.

Keywords : Probable maximum precipitation; probable maximum flood; extreme predictions; relative error; coefficient of variation.

        · abstract in Spanish     · text in English | Spanish     · English ( pdf ) | Spanish ( pdf )