Utilization of MOLUSCE tool and GEE cloud to predict the variation of LST through LULC, NDBI and NDVI in Tollygunge-Panchannagram Basin of Kolkata

Abstract

Kolkata, a metropolitan city in India with over 15 million residents, has experienced a steady rise in urban heat islands over recent decades. This study investigates the spatial and temporal dynamics of land use/land cover (LULC) and their impact on land surface temperature (LST) in Tollygunge-Panchannagram (TP) basin of Kolkata using Landsat data from 2000, 2010, and 2020 within the Google Earth Engine (GEE) environment. Additionally, it projects LST for 2030 based on LULC, the Normalized Difference Built-up Index (NDBI), and the Normalized Difference Vegetation Index (NDVI). The prediction employed the Methods of Land Use Change Evaluation (MOLUSCE) plug-in in QGIS, which integrates Artificial Neural Networks with Cellular Automata (CA-ANN). CMIP5-based RCP 4.5 climate data were incorporated to simulate future climate conditions and forecast LST for 2030. The findings revealed that the maximum LST increased from 36.15 °C in 2000 to 47.65 °C in 2020, showing a strong positive association between NDBI and LST. Elevated temperatures were primarily concentrated in the western and northwestern sectors, driven by urban expansion, industrial growth, and infrastructure development. According to the CA-ANN MOLUSCE analysis, built-up areas expanded by 6.31% between 2000 and 2020 and are anticipated to rise by another 8.1% by 2030, leading to further reductions in vegetation, open spaces, and water bodies. These spatiotemporal shifts, resulting from rapid urbanization, have significantly altered the city’s microclimate and reduced the extent of water-related land cover. The study further emphasizes the necessity of implementing sustainable Urban Heat Island (UHI) management strategies through increased green cover and water body restoration.