Characterization of the actual evapotranspiration series, in a ten-day period, in the state of Sao Paulo-Brazil

The scientific analysis of drought is directly associated with the meteorological characteristics of a given area. Therefore, the climate characterization of that area is the first step to be pursued by a study of drought. Considering agricultural drought as a deficit between the estimated actual evapotranspiration over a period and the expected value for the period under evaluation, the aim of this study was to analyze four different types of probability density function (normal, log-normal, gamma and beta). The Kolmogorov-Smirnov test was used to verify the tack of the ETR series, in a ten-day period to the above-mentioned theoretical distributions. Air temperature and precipitation data of 13 locations of the State were used to calculate the climate water balance. Beta distribution has proved the best adjustment compared with the others, and is recommended as a basis for an agricultural standardized drought index.