Análise Bayesiana da precipitação máxima de Petrópolis-RJ

Abstract

The city of Petrópolis, located in the mountainous region of the state of Rio de Janeiro

(RJ), Brazil, frequently suffers from damage caused by heavy rains, such as those oc- curred in 2011 and 2013. Therefore, analyze and predict the occurrence of extreme

rainfall in Petrópolis is fundamental for planning activities vulnerable to their occur- rence, such as farming and and the removal of people from risk areas, the first being

fundamental for the livelihoods of many local families and the second being extremely important to avoid the loss of human life. The modeling of the behavior of this extreme event is usually done through the Generalized Distribution of Extreme Values (GEV). The Bayesian methodology has shown good results in estimating the parameters of the GEV distribution. Thus, the purpose was to adjust the GEV distribution to the historical series of maximum rainfall in Petrópolis, and to evaluate different structures of prioris,

informative and non-informative in the forecasting of extreme rainfall values in Petrópo- lis at different return times. It was possible to conclude that the priori distribution based

on information from the city of Teresópolis, given the variance, provided more accurate and accurate results in the prediction of return levels for the city of Petrópolis. In this way, forecasts for the maximum rainfall expected for the return times of 2, 5, 10, 25, 50, 100, 115 and 135 years were made using this priori structure. It is expected that, in an average time of 5 years, there will be at least one day with maximum rainfall greater than or equal to 100.7mm in Petrópolis.