Ajuste de distribuições de probabilidades à série 1970-2010 de precipitação máxima anual de Rio Branco, Acre

Abstract

The determination of the magnitude of intense rains is of fundamental importance for the elaboration of hydraulic projects and management of water resources in engineering, in addition to the design of structures for the control of water erosion in soil conservation. The objective of this study was to analyze the performance of four probability distributions, applied to the historical series of maximum annual daily precipitation in the municipality of Rio Branco, Acre. The rainfall data were collected in a conventional meteorological station of the National Institute of Meteorology (INMET), referring to the period from 1970 to 2010. The non-parametric Kolmogorov-Smirnov statistical test was applied, at a significance level of 5%. The results show that the probabilistic models Gamma, Log-Normal with 2 parameters and Normal, satisfactorily adjusted the data of maximum annual daily precipitation. The Lognormal distribution with 2 parameters, presented higher estimates of probable precipitation, for return periods equal to or greater than ten years, in relation to the other probability distributions evaluated in this study.

Keywords: Statistical hydrology. Random variables. Extreme events.