Heavy rainfall events in the state of Acre, western Amazonia

Abstract

Abstract Currently, humanity is unable to prevent extreme hydrological events from occurring. However, the use of appropriate tools to estimate the probability associated with different hydrological variables, such as heavy rainfall, can help mitigate social and environmental damage. The objective of this study was to verify the hypothesis that the maximum rainfall records (over different intervals of duration) from the three main rainfall stations in the state of Acre fit the Gumbel distribution. Additionally, the periods of intense rainfall were evaluated based on the association between rainfall duration and return period. Using the maximum likelihood method, and once the baseline stationarity assumption established by the Mann-Kendall test was met, the relevant equations were fitted to the classical Gumbel model. Adherence to the model was assessed using the Anderson‒Darling test. The fitted distributions allowed the estimation of rainfall intensity for different intervals of duration and return periods for the cities of Cruzeiro do Sul, Tarauacá and Rio Branco. These results are relevant for the implementation of plans aimed at strengthening the water security in the state of Acre.