Temporal trend and frequency of maximum precipitations in Cruzeiro do Sul, Acre, Brazil

Abstract

The verification of temporal trends in historical series and the frequency analysis of maximum precipitations events are important for the understanding of the dynamics of the hydro-meteorological processes. In this context, this study aimed at verifying series trends in the observed data from the annual maximum monthly precipitations in Cruzeiro do Sul, Acre State, Brazil, registered at 82704 station (7°38’S; 72°40’W) operated by the National Meteorological Institute, besides performing the frequency analysis through Gumbel probabilistic distribution for the maximum events. The results obtained through Mann-Kendall test suggest a light increasing temporal trend; however, it converges on rejecting the hypothesis of a significant trend, at the significance level of 5%, considering the analysis between 1971 and 2013. Regarding the frequency analysis, it was shown that Gumbel’s distribution was appropriate to represent the maximum precipitation, through the adherence obtained by the Kolmogorov-Smirnov and the Chi-Square tests, to 5% probability.