A Study on Precipitation Trends in Türkiye via Linear Regression Analysis and Non-Parametric Mann-Kendall Test

Abstract

Climate change, a significant global issue, has recently become a pivotal area for many researchers. Climate is defined as the long-term average of meteorological parameters such as temperature, precipitation, wind speed, and air pressure of a region. Precipitation is an important climatic parameter that can be variable at the temporal and spatial scales. A half-century precipitation dataset was selected and analyzed to reveal the effects of global climate change on the rainfall amounts of Türkiye. Precipitation data of each geographical region have been analyzed with respect to annual and seasonal basis in the period of 1969-2018. For this purpose, the non-parametric Mann-Kendall trend test which is recommended by World Meteorological Organizations (WMO) and linear regression method have been implemented to each geographical region of Türkiye. As a result of analysis belonging to 85 meteorological stations, the presence of any increasing and decreasing linear trends in annual and seasonal precipitation series have been studied on a regional scale. While the Black Sea Region has the highest increase with 148 mm/50 years, the total annual precipitation in the Southeastern Anatolia Region has decreased by 3.2 mm/50 years. Another important finding of linear regression has been observed that increase in precipitation has occurred in the Black Sea Region in all seasons, as a consequence of seasonal analysis. To determine whether these trends are statistically significant, we used Mann-Kendall test results. The test proved the existence of an increasing trend at 99% significance level in the annual precipitation series of the Black Sea Region. A statistically significant increasing trend was also obtained for the autumn season of the Black Sea Region at a 95% confidence level.