Rising Urban-Rural Temperature Gradient in Indian Cities: Analysis and Characterization

Urbanization and regional climate change-induced warming, known as the Urban Heat Island effect, result in urban areas experiencing temperatures 1–4 °C higher than their rural counterparts. This phenomenon poses significant risks to biodiversity, human health, and regional climate systems, necessitating an in-depth understanding of its spatiotemporal patterns and characterization to inform effective adaptation strategies. In this study, we investigated the diurnal and seasonal dynamics of  Surface Urban Heat Island intensity (SUHII) for 141 Indian cities over two decades (2001-2022) using MODIS satellite-derived Land Surface Temperature (LST), Normalized Difference Vegetation Index (NDVI), evapotranspiration (ET), and Land Use Land Cover (LULC) data. We employed the urban-rural method to calculate SUHII, used the Mann-Kendall Test and Theil-Sen slope estimator to identify trends, while five-year interval analyses captured the evolution of SUHII hotspots. Further,  to characterize SUHI variability, we used a Multilevel Modeling (MLM) approach, incorporating time-varying NDVI and ET, alongside city size as a time-invariant factor. Our findings reveal a significant rising trend in nighttime SUHII across most cities, while five-year average change analyses highlight emerging daytime SUHI hotspots during both summer and winter seasons. The MLM approach explained more than 90% of SUHII variability in both seasons. While SUHII generally showed negative associations with ΔNDVI and ΔET across most cities, except in warm deserts, city size exhibited a negative yet weak association. Overall, our findings demonstrate the escalating SUHI effect in Indian cities and underscore the importance of vegetation and water dynamics in regulating urban thermal environments at a regional scale. These insights emphasize the urgent need for sustainable local-scale urban planning to mitigate the adverse impacts of SUHI on ecosystems and human well-being.