Abstract

Urban trees are critical for climate mitigation, yet the inhibitory effects of the urban matrix remain underquantified. This study leverages multi-source LiDAR to reconstruct the carbon sequestration trajectories of 45,576 trees in Helsinki (2015–2023). Regression analysis reveals that urban environmental factors are deterministic drivers, explaining 13–52% of growth variance. Crucially, we identify a non-linear “stress penalty” where solar radiation and heat—typically growth-promoting—act as metabolic stressors alongside anthropogenic noise. Based on differential sensitivities, we classified taxa into structure-dependent specialists (Quercus, Pinus), atmosphere-sensitive indicators (Sorbus, Tilia, Acer), and resilient generalists (Betula). These findings challenge aesthetic-based planting, advocating instead for performance-based zoning: utilizing resilient species in high-stress canyons and reserving park sanctuaries for high-potential giants to urban climate mitigation infrastructure.