Snow cover detection using remote sensing techniques over different climate zones of Türkiye

Abstract

Snow-covered land surfaces can be easily mapped using remote sensing technologies. Accurate estimation of snow cover on the land surface allows for the construction of water resource management today. Using Landsat TM/ETM + satellite images, this study tried to assess how much snowfall covered the soil in Trabzon, Gümüşhane, and Bayburt provinces between 1999 and 2023. Satellite images were classified into three categories using the ERDAS 9.1 TM software. These classes are classified as snow-covered surfaces, other places, and data loss (cloud-shadow). When performing image analysis, it was important to verify that the cloudiness rate in the images was less than 15%. Images with cloudiness rates of more than 15% were not used. Seasonal and annual trend analysis of snow-covered areas (SCA) over three distinct regions (Trabzon, Gümüşhane, and Bayburt) were examined using the Mann-Kendall test. When three distinct study regions were examined together, Bayburt had the highest SCA rate, followed by Gümüşhane and Trabzon. In all three fields, the highest SCA was recorded in 2000, while the lowest SCA was recorded in 2017. The trends noticed that SCA on both annual and seasonal scales did not reach the statistical significance level of 0.05. Although snowfall in Trabzon, Gümüşhane, and Bayburt was beneficial in the autumn and spring seasons, no statistically significant association was found. The research concluded that the existing data are inadequate to make any statements on the impact of global warming in the area. However, the study figured out that satellite data may be effectively used to identify snowy places as a result of the study. Comparable investigations need to be undertaken in regions with varying climates, utilizing diverse remote sensing data and classification methodologies.