Homogeneity and Change Point Detection Analysis of Seasonal and Annual Precipitation and Temperature Series Van, Turkiye

Abstract

Abstract: Researcher which study water resources, hydrological processes, and climate change fields are given of high importance to the accuracy and consistency of historical temperature and precipitation series. In this regard, Homogeneity tests are a helpful tool to manage the accuracy and consistency of the data. In this study, the homogeneity of seasonal and annual temperature and precipitation is examined using Standard Normal Homogeneity (SNHT), Pettitt (PT), Buishand Range (BR), and Cumulative Deviation (CD) tests at a significance level of 0.05. The four homogeneity tests are assessed using long-term annual and seasonal precipitation and temperature records obtained from five meteorological stations located in Van, Turkey. Finally, change point detection is investigated for each station. According to the results, the annual mean temperature rejects H0 hypothesis for all stations by four tests. While Ha alternative hypothesis is accepted for stations 17172 and 17852 and 17880 at annual total precipitation, stations 17784 and 17812 are observed as doubtful. Seasonal results revealed that although it is rarely seen that there is homogeneity, it can be concluded that the homogeneity is broken in the precipitation and temperature data, especially in the temperature data. It can be seen that stations 17784 and 17852 generally experienced changes after 2000, whereas stations 17172, 17812, and 17880 experienced changes before this year in the series. The results of this study constitute a source of information in terms of the reliability of the meteorological data series used in climate research, hydrological modeling, and statistical research to be carried out in this region. As a result, the reliability of the data should be questioned in the studies to be carried out in the Van region and the data should be made reliable in the projects to be carried out.