Occurrence probability of maximum air temperature in scale of fifteen days in Iguatu city, Brazil

This study aimed to calculate the probabilities of occurrence of maximum air temperatures in Iguatu city, Ceará, Brazil, on a scale of fifteen days for each month of the year at different levels of probability. Were used daily data of maximum air temperature, obtained from the 3rd District of the Instituto Nacional de Meteorologia (INMET). The distribution used was the Log-Normal and its performance was tested according to the Chi-square and KolmogorovSmirnov test. The probability levels used were 25, 50, 75 and 90%, representing a payback period of 1,3, 2, 4 and 10 years respectively. The Log-Normal distribution fit obtained satisfactory and can be used for the representation of the data used in the study according to both tests of adhesion. After applying the Log-normal distribution was possible to generate tables with the values of maximum air temperature expected for different levels of probability. The results of the tables show that for every month of the year, on a scale of fifteen days, to a maximum temperature of Iguatu at all levels of probability studied, will be greater than 30 °C.