K-Means Clustering of Precipitation in the Black Sea Region, Türkiye

Abstract

In recent years, there has been a significant uptick in the frequency of disasters stemming from the impacts of global climate change. In response, both nationally and internationally, various studies are being conducted to mitigate these effects. Classifying regions affected by climate change into similar classes based on climate parameters is crucial for applying consistent methodologies in studies conducted within these regions. This approach will help determine the most appropriate strategies for mitigating the effects of climate change in these regions. The study utilized observational records of annual precipitation from 31 stations in the Black Sea Region, sourced from the Turkish State Meteorological Service, covering the data spans the period between 1982 and 2020. Cluster analysis was conducted using the k-means algorithm. The optimal cluster among those formed was determined through the silhouette index analysis. The study suggests that the optimal number of clusters is 2.