Urbanization-driven LULC change and heat intensification: a machine learning study of a small coastal cityscape

Urbanization has become a key factor in environmental change, amplifying urban heat and impacting ecological resilience in the world’s fastest-growing cities. The present study was conducted to elucidate the spatiotemporal dynamics of land use and land cover (LULC) and their correlation with land surface temperature (LST) and the Urban Heat Island (UHI) effect in the small coastal city of Balasore, Odisha, India. This study utilized multi-temporal Landsat data for 2000, 2012, and 2024, machine learning-based Random Forest (RF) classification, and spatial indices (normalized difference vegetation index (NDVI), normalized difference built-up index (NDBI), and normalized difference water index (NDWI)), to examine LULC transitions, biophysical properties, and thermal variations. Results show an increase in built-up area (18.6% in 2000 and 46.6% in 2024), primarily at the expense of cultivated and open lands. Meanwhile, vegetation cover and water bodies experienced some localized gains, but these were overshadowed by the widespread expansion of impervious surfaces replacing the previous land covers. Simultaneously, the mean LST increased from 31.78 °C to 38.31 °C between 2000 and 2024, and the maximum LST rose to 44.59 °C, indicating substantial increases in heat intensity in Balasore. Correlation analysis revealed a significant positive correlation between NDBI and LST, and a significant negative correlation between NDVI and LST, indicating the cooling effect of green cover. Both the UHI and UTFVI patterns indicate an increase in thermal stress in densely urbanized areas. Therefore, these findings suggest that urban expansion leads to increased heat stress, which can have significant consequences for ecological sustainability and urban livability.