Assessment of long-term urban vegetation quantity versus quality by decoding green vigor, atmospheric resilience, chlorophyll, and senescence analysis applying multi-spectral remote sensing data fusion

Rapid urbanization and industrialization have substantially transformed vegetation ecosystems worldwide, raising concerns regarding the long-term sustainability of urban green infrastructure and ecosystem services. The existing literature primarily relies on conventional indices and vegetation extent, often overlooking vegetation vigor, physiological condition, atmospheric-resilient greenness, and senescence dynamics. This study develops an integrated multi-index remote sensing framework to investigate long-term vegetation trajectories in the Haldia urban-industrial region (103.84 km²), India (1991-2021). Multi-sensor Landsat-5 TM, 7 ETM+, and 8 OLI/TIRS were processed to derive advanced vegetation indicators comprising the Enhanced Vegetation Index (EVI), Atmospherically Resistant Vegetation Index (ARVI), Modified Chlorophyll Absorption Ratio Index 2 (MCARI2), and Normalized Difference Senescent Vegetation Index (NDSVI). The framework quantifies vegetation vigor, atmospheric-resilient vegetation condition, chlorophyll-physiological health, and vegetation senescence of urban-industrial transformation.

The results reveal decline in vegetation extent (30.12-22.16 km²) (-26.4%), accompanied by rapid built-up expansion (22.87-53.37 km²) (+133.4%). Despite this reduction in vegetation area, mean EVI, ARVI, and MCARI2 values increased by 6.3%, 11.8%, and 16.7%, respectively, indicating enhanced vegetation vigor, atmospheric-resilient greenness, and physiological condition within the remaining vegetation cover. Concurrently, increasing standard deviation values for EVI (0.06-0.08), ARVI (0.12-0.15), and MCARI2 (0.03-0.04) revealed growing spatial heterogeneity of vegetation condition under continued urban-industrial development. In contrast, NDSVI remained relatively stable (mean = 0.11; SD = 0.06), suggesting limited long-term increases in vegetation senescence and degradation. Cross-index consistency analysis demonstrated strong positive correlations among EVI, ARVI, and MCARI2 (r = 0.671-0.953), confirming the robustness of vegetation vigor and physiological-condition assessments. Furthermore, inverse relationships between vegetation extent and vegetation condition indicators (r = -0.647 to -0.918) provided quantitative evidence of a transition from vegetation quantity to quality. The spectral findings highlight the ecological resilience of surviving urban green spaces despite continued urban-industrial expansion, providing a transferable Earth observation approach for urban ecological assessment and green space sustainability.