Urban Transformation Analysis of Metro Corridor Expansion, LULC Changes, and openly accessible DEM Accuracy Assessments using ICESat-2

Metro expansion induces significant modifications to the terrain and land cover through construction activities such as flyovers, embankments, and excavation. Accurate elevation data is crucial to detect these horizontal and vertical changes, especially in rapidly urbanizing areas. This study utilized NASA’s ICESat-2, launched in 2018 with the Advanced Topographic Laser Altimeter System (ATLAS), which operates at a wavelength of 532 nm and a pulse repetition frequency of 10 kHz, producing ~70 cm ground footprints. ICESat-2’s high-resolution photon-counting laser altimetry provides more precise elevation data than conventional DEMs such as SRTM, ASTER, and CartoDEM. The Kolkata Metropolitan Area, situated along the Hooghly River in eastern India, was selected as the case study due to its ongoing metro expansion and urban sprawl. ICESat-2 elevation values were compared with six open-access DEMs—FABDEM, CartoDEM, TanDEM-X Edited DEM, SRTM, ASTER, and COP-DEM-GLO-30—at selected footprint locations. After filtering uncertainty and removing building footprints, 1636 ATL08 footprints were used for analysis of terrain height, and 38,101 ATL06 footprints were used for for the Digital Surface Model analysis. Outliers were eliminated using the 3-sigma method. Statistical metrics, including RMSE, MAE, mean, SD, and MAD, were used for analysis. The assessment of DEMs used ICESat-2 as a reference and a filtering threshold of 0.25m for elevation deviations, finding that the values of RMSE for FABDEM, CartoDEM, TanDEM-X EDEM, SRTM, COP-30, and ASTER are 1.41m, 4.36m, 3.42m, 4.78m, 3.07m, and 7.14m, respectively. The results indicated that FABDEM 1.41m (for ATL08) is the most reliable among the evaluated datasets. Additionally, LULC changes over the past 30 years were mapped using random forest classification and multi-temporal satellite imagery. The analysis showed significant increases in built-up areas from 1995 to 2025 among all the buffer classes of 250m, 500m, 750m, and 1000m used in this study. This integrated analysis offers critical insights for metro infrastructure development and urban planning in fast-growing metropolitan regions.