The aim of this work was to characterize annual series of absolute daily value of maximum (Tmaxabs) and minimum (Tminabs) air temperature in the State of São Paulo, Brazil, detecting possible trends, periodicities and, associated probabilities. Using non parametric methods and spectral analysis, no trends and periodicities were detected in all analyzed series, excluding the Tminabs series of Ubatuba-SP, where increasing trends was detected. The probability density function associated to the General Value Distribution and to the normal distribution can be used, respectively, to estimate the probability associated to the Tmaxabs and Tminabs values in all series. The wavelet power spectrum shows appreciable power concentrated within the most severe event of meteorological drought observed in the scientific literature of the State of São Paulo. The same feature was not observed in the Tminabs series.
Climate trends; probability of occurrence
