Spatio-Temporal Analysis of Land Surface Temperature and Land Use/Land Cover Relationship in Antalya Basin

The Mediterranean Region is considered one of the most vulnerable regions to climate change, with annual temperatures projected to increase. With the additional heat load associated with anthropogenic activities, this region is expected to experience greater heat stress and negative health effects than other regions. The Antalya Basin, situated within the Mediterranean Basin, is characterized by rich biodiversity, important wetlands, intensive agricultural and tourism activities, and rapid urbanization, providing a relevant setting for examining land use/land cover (LU/LC) and land surface temperature (LST) relations. This study investigates long-term LU/LC and LST changes in the basin and assesses the relationship between them. For this purpose, initially pre-processing steps were performed to prepare the data sets and spectral indices were generated. Then the LU/LC classification was performed using Landsat imagery and auxiliary data for nine years at five-year intervals from 1985 to 2025. Afterwards the annual LST images were generated using Landsat imagery for the years between 1985 and 2025. Finally, accuracy assessment was performed and the relationship between LU/LC and LST was analyzed. The LU/LC thematic maps derived from Landsat images along with additional raster layers were obtained quite accurately with overall accuracies higher than 85%. The change detection results revealed increases in residential, industrial, and agriculture areas and were accompanied by decreases in vegetation and rangeland/bareland areas. Long-term annual LST analyses revealed distinct thermal trajectories among LU/LC classes from 1985 to 2025. Statistically significant monotonic warming trends were confirmed in residential, industrial, water surfaces, vegetation, and greenhouse areas, with residential areas demonstrating the strongest warming rate. In contrast, dry agriculture, rangeland/bareland, and irrigated agriculture exhibited high interannual variability and pronounced surface temperature peaks—most notably in 2024, when dry agriculture reached a maximum LST of 47.72 °C—but lacked statistically significant long-term monotonic trends. The highest LST values were measured in dry agriculture areas, followed by greenhouse, industrial, and residential areas. In contrast, water and vegetation areas exhibited the lowest LST values among all LU/LC classes.