Further Improvements to classical standard precipitation index (SPI) calculations

Abstract Global warming coupled climate change impacts show their end results in the form of extreme events including floods and droughts. Droughts are creeping phenomena and their prediction is more related to precipitation events and in many parts of the world droughts occurrences become temporally and spatially effective. In the literature, there is the concept of standard precipitation index (SPI) that provide classification of the drought based on the standard normal (Gaussian( probability distribution function (PDF) which is a product of the original time series record data PDF that may have various forms such as Gamma, Weibull, Log-Normal, Pearson, and like. As for SPI there are publications that are based on statistical standardization procedure whereas it should be probabilistic standardization methodology. On the other hand, many publications concentrate on conversion of the original PDF of the time series to Gamma PDF and then apply the normalization procedure. In this paper conversion of any PDF to Gamma PDF is noticed as misconception except in the case of original PDF confirms with Gamma PDF: It is well known that almost all hydro-meteorology time series have various PDF not abide with Gamma PDF: The mistakes of conversion to Gamma PDF prior to SPI is explained verbally and numerically and the genuine proper SPI procedure methodology is proposed in detail. The application of the proposed methodology is given for seven meteorology station records from seven internal climate zones of Türkiye.