Temporal analysis of land surface temperature dynamics in response to the trajectory of urbanization

Abstract

This study investigates land use/land cover and land surface temperature changes for 1989 and 2023 in Kharkiv, Ukraine, utilizing remote sensing, machine learning algorithms, and climate data analysis. Landsat and Modis satellite images were analyzed using a Random Forest classifier to categorize LULC into five types, enhanced by vegetation indices. The analysis revealed a increase in urban areas and decrease in liquid vegetation. Climate trend analysis showed significant temperature fluctuations, with a general increase in LST, particularly in summer months, highlighting urban heat island effects. It indicates the impact of urban expansion on UHI and emphasizes the necessity for effective urban planning and environmental management to mitigate these effects.