PROBABILITY OF OCCURRENCE OF HARMFUL AND FAVORABLE MONTHLY MINIMUM TEMPERATURES FOR AGRICULTURAL CROPS IN MINAS GERAIS

Abstract

The objectives of this work were to verify the fit of the data series of monthly minimum air temperature (Tmin), from 47 municipalities in the State of Minas Gerais, to the functions of normal, log-normal, gamma and weibull probability distributions; determine, for each data series, value of Tmin whose probability of occurrence of Tmin less than or equal to this value is 90%; and prepare maps with the interpolation of these values ​​in isolines with a difference of 0.5ºC. In the 564 data series (47 municipalities × 12 months) of Tmin, the Kolmogorov-Smirnov adherence test was applied and the value of Tmin was determined, whose probability of occurrence of Tmin less than or equal to this value is 90%. The Tmin data fits the normal, log-normal, gamma, and weibull distributions. The estimates of the parameters of the normal distribution (mean and standard deviation) can be used to determine the value of Tmin whose probability of occurrence of Tmin less than or equal to this value is 90% and, for later elaboration of maps with the interpolation of these values of Tmin in isolines. Isolines are useful to verify the probability of occurrence of harmful or favorable Tmin for agricultural crops.