Cities are pivotal regulators of the global carbon cycle, yet the fine-scale balance between urban carbon emissions and sequestration remains poorly quantified. Here, we conduct a high-resolution spatiotemporal assessment of Beijing's carbon balance from 2000 to 2020, integrating emission inventories, ecosystem productivity modeling, and machine-learning–based attribution analyses. Over the period 2000–2020, carbon sequestration increased by 2.34 Mt. C while carbon emissions declined by 11.50 Mt. C. Consequently, the city-wide Carbon Balance Index (CBI) improved by 33.8%, narrowing the carbon deficit from 40.94 Mt. C to 27.10 Mt. C. Four distinct carbon transition pathways were identified, ranging from synergistic mitigation—concentrated in ecological zones—to persistent source-dominated regimes in the urban core. Among all drivers, green coverage ratio emerged as the strongest positive regulator, counteracting the negative effects of urbanization intensity, although diminishing marginal returns were evident in highly dense districts. These results demonstrate that targeted strategies—expanding green infrastructure, enhancing landscape connectivity, and moderating excessive urban density—can substantially buffer carbon losses. The integrative framework and empirical evidence presented here offer transferable insights for rapidly urbanizing cities worldwide seeking to accelerate progress toward carbon neutrality. • SHAP–MGWR–SEM framework disentangles nonlinear urban carbon driving mechanisms. • Beijing's Carbon Balance Index improved by 33.8% from 2000 to 2020. • Four distinct carbon transition pathways reveal core-periphery divergence. • Green coverage enhances carbon balance but saturates in high-density zones.
Zhou et al. (Sat,) studied this question.