Assessment of spatio-temporal variation in seasonal land surface temperature and its relationship with spectral land use indices in Varanasi

The dynamics of Land Surface Temperature (LST) are intricately linked to changes in land use and land cover, necessitating comprehensive spatio-temporal assessments to inform sustainable urban and environmental management strategies. The cities of developing countries like India have witnessed significant urbanization in the past few decades, which has significantly influenced the thermal environment. Therefore, this study investigates the seasonal variations in LST and explores its relationship with key spectral land use indices—Normalized Difference Vegetation Index (NDVI), Normalized Difference Built-up Index (NDBI), and Normalized Difference Water Index (NDWI)—in Varanasi, India, a region experiencing rapid urbanization and ecological transformations. To retrieve LST, thermal bands of Landsat data have been used, while for the calculation of spectral land use indices, multispectral bands (Band 3 to 5 for Landsat 5 and Band 4 to 6 for Landsat 8 and 9) have been used. The descriptive statistics show that the LST initially increased in Varanasi during 1993–2009 but later declined during 2009–2023. The mean LST of the summer season increased from 32.47 °C to 34.43 °C during 1993–2003 but declined to 30.34 °C during 2009–2023. Similarly, for the winter season, the mean LST has increased from 17.66 °C to 20.01 °C during 1993–2009 but declined to 18.54 °C during 2009–2023. The correlation analysis shows a varying but significant relationship between spectral land use indices and LST. The study shows that the dynamics of built-up area, vegetation cover, and soil moisture, as shown in spectral land use indices, play a critical role in modulating LST. The findings may help policy-makers, government bodies, and urban planners in implementing appropriate solutions for sustainable urban planning and maintaining the urban thermal environment.