TY  - JOUR
AU  - Gupta, Rupesh Kumar
AU  - Tiwari, Yash Kumar
AU  - Gupta, Arpit
AU  - Yadav, Grinedge
AU  - Gupta, Swati
PY  - 2026
DA  - 2026/10/23
TI  - Urban Expansion, Thermal Transformation, and Ecological Stress in Chennai, India: A Multi-Temporal and Predictive Analysis Integrating Landsat and VIIRS
JO  - Advances in Environmental and Engineering Research
SP  - 035
VL  - 07
IS  - 04
AB  - Rapid urban growth transforms land cover and alters surface thermal conditions, especially in fast-changing tropical coastal cities. This study examines the spatial and temporal relationships among land-use/land-cover (LULC) transformation, land surface temperature (LST), relative thermal stress, and nighttime illumination in Chennai, India. Landsat observations for 2005, 2015, and 2025 were processed in Google Earth Engine to derive LULC for 2005 and 2025 and seasonal LST and Urban Thermal Field Variance Index (UTFVI) measures for the thermal reference years, while annual VIIRS nighttime radiance was analysed for 2014-2025. We simulated future LULC using a Cellular Automata-Artificial Neural Network (CA-ANN) framework, then used Random Forest regression to estimate a conditional thermal response associated with the simulated 2045 land configuration. Between 2005 and 2025, built-up land increased from 258.83 to 305.69 km2, whereas cropland and barren land declined by approximately 60% and 75%, respectively. Under the CA-ANN scenario, built-up land is projected to reach 352.55 km2 by 2045, representing approximately 80.56% of the study area. Historical thermal analysis shows substantial redistribution among LST classes, with the 35-40°C class becoming increasingly dominant; however, the observed class redistribution is distinguished from a statistically significant temporal warming trend. The 2045 thermal surface is similarly concentrated within the 35-40°C range, while the highest-temperature classes contract under the modelled scenario. This redistribution is interpreted using fixed temperature thresholds and possible upper-tail attenuation in the Random Forest estimates and is not interpreted as evidence of future cooling. UTFVI analysis indicates substantial spatial redistribution of relative thermal stress, while mean absolute UTFVI (MAUTFVI) provides a more informative measure of interannual variation in stress magnitude. Annual VIIRS nighttime radiance increased significantly from 20.992 in 2014 to 28.990 in 2025, whereas the seasonal LST and MAUTFVI series did not show statistically significant linear temporal trends. The bivariate VIIRS relationships with seasonal LST and MAUTFVI were also not statistically significant at the 5% level, indicating that nighttime illumination alone does not adequately explain interannual thermal variability. Overall, the results indicate that nighttime illumination, land transformation, and thermal conditions represent related but non-synchronous dimensions of urban change. The integrated framework therefore provides a scenario-based basis for understanding Chennai’s evolving thermal environment and for supporting climate-responsive and spatially targeted urban planning.
SN  - 2766-6190
UR  - https://doi.org/10.21926/aeer.2604035
DO  - 10.21926/aeer.2604035
ID  - Gupta2026
ER  -