Examining the best probability distributions of meteorological parameters based on annual scale

Climate and climate related events are still taking the interests of societies as it has a huge impact on live life, energy, economy and sustainability. Meteorological parameters such as precipitation, temperature, humidity provide significant information to humanity to adopt with nature. Within this scope, the aim of this study is to analyze the spatio-temporal variability of mean and maximum temperature, mean relative humidity and total precipitation series and obtain the best fitted probability distributions of them for Karaman, Konya and Ankara Regions of Turkiye considering annual scale by obtaining required data from General Directorate of Meteorology of Turkiye. The spatial distribution of selected meteorological variables has been analyzed by Inverse Distance Weighting (IDW) technique and possible temporal variability of these variables have been examined by the trend tests of Mann Kendall (MKT), Spearman’s Rho (SRT), Sen’ Slope(SSE). The best fitted distributions of parameters have been analyzed by utilizing the most commonly used probability distributions of Normal, Lognormal, Logistic, Gamma, Weibull, Gumbel, Generalized Extreme Value considering their performance in a set of goodness of fit tests which are Kolmogorov-Smirnov, Anderson Darling, Chi Square, Cramer Von Mises, Akaike Information Criterion, Bayesian Information Criterion and Maximum Likelihood Estimations. The trend results of MKT and SRT have demonstrated that although there has been a variety of trend types (increasing or decreasing) in data series, a few times these trends have been significant when considering α=0.01 significant level while SSE results have been generally detected with increasing trend slopes. The best-fit distribution results have shown that generally Generalized Extreme Value, Logistics and Gamma distributions have been detected as the best fitted distributions while Lognormal, Gumbel and Weibull have been detected with less numbers. The results of this study are expected to be beneficial for climate change action and water resources planning and urban management areas as the study presents significant findings.