Modelling microscale impacts assessment of urban expansion on seasonal surface urban heat island intensity using neural network algorithms

Investigation of surface urban heat island (SUHI) results from rapid urbanization and upsurge of land surface temperature (LST) has substantial socioeconomic and environmental impacts. This study investigates and simulates the impacts of rapid urbanization on LST and SUHI patterns in Sylhet City, Bangladesh, from 1995 to 2030. Landsat images and machine learning algorithms have been used to identify the urban growth, LST and UHI distribution patterns in several city directions. In addition, correlation analysis has been conducted between LST, SUHI and spectral indices (NDBI, NDBSI, NDVI, NDWI). Results suggested that urban expansion increased LST by 7 °C in summer and 6 °C in winter from 1995 to 2020. Increment has also occurred in summer high SUHI from 0.25 km 2 to 2.65 km 2 . Pearson correlation demonstrated that built-up areas have a strong positive relationship with LST (0.96) and SUHI (0.911). Future simulation of urban expansion for 2025 and 2030 shows a 9 % increase, leading to a significant increase in moderate to high SUHI intensity. The study's findings can act as an effective guideline for sustainable infrastructural development and ensure environmental stability by increasing the thermal comfort level of the city.