Rainfall is an important meteorological element, and the knowledge on its historical records is relevant for monitoring the impacts caused by its prolonged excess or lack. This study aimed to evaluate the monthly and annual rainfall variability and the probability for its occurrence, by using the incomplete Gamma distribution model, for the rainfall data from Cáceres, Mato Grosso State, Brazil. A series of 26 years data, provided by the National Institute of Meteorology, which supports a station in Cáceres, was used. The annual rainfall amounts showed great variability, with minimum of 972.9 mm, in 1985, and maximum of 1,624.1 mm, in 1998. The monthly data pointed out two well defined seasons: the dry season (May to September) and the rainy one (October to April). The lowest monthly averages occurred in June, July, and August, with 16.59 mm, 17.90 mm, and 20.09 mm, respectively. The estimates for the α parameter ranged from 0.9, in June and August, to 13.4, in March. The β parameter ranged from 13.2, in March, to 33.1, in January. The highest rainfall probability takes place from December to March, while June, July, and August show the lowest rainfall probability.
Incomplete Gamma function; climate; meteorology
