Análise da precipitação mensal provável para a fazenda Esmeril - Patrocínio-MG

Abstract

The adjustment of probabilistic models to daily rainfall data and provide a brief summary of these data is an efficient technique for the analysis of this information. This study aims to determine if the daily average for each month of Emery Farm area in the municipality of Patrocínio-MG, from 2000 to 2012, adjusts a Normal distribution function, predict the rainfall amount of certain months using Normal distribution and verify the data of the time series studying its rainfall behavior in this interval of thirteen years using the concepts of time series. First it was found that the data set the distribution function using the Kolmogorov-Smirnov test with 5% significance level. After checking the adherence of values to the normal distribution, a prediction using the Central Limit Theorem for the first ten months of each interval was made, but this method was not very efficient. The study of the time series showed that the rainfall data of the range of the period considered for the total monthly variable daily average of each month, weekly total and daily average of each week have no tendency has seasonality and have irregular variations.