Considering the socioeconomic importance of Presidente Prudente area, this study aimed at to estimate the maximum pluvial precipitation expected for different levels of probability and to check the fitting degree of the data to the Gumbel model, with the parameter estimation obtained by the maximum likelihood approach. The Kolmogorov-Sminorv (K-S) test showed that the Gumbel distribution has fitted with p-value larger than 0.28 for all of the time periods considered, showing that the Gumbel distribution presents a good fitting to the observed data representing the maximum pluvial precipitations values. The precipitation estimates obtained by the maximum likelihood approach are consistent allowing to reproduce with plenty reliability the rainfall regime for Presidente Prudente region. Therefore, the knowledge of the monthly maximum pluvial precipitation distribution and the expected maximum daily precipitation estimates permits a better strategic planning thus minimizing the risk of economical losses for this region.
Maximum precipitation; maximum likelihood estimate; Gumbel distribution; confidence interval
