In rapidly urbanizing areas, there is a notable aggregation of vitality in high-density urban environments, accompanied by an increasing discrepancy between the supply and demand of urban green space (UGS). This study presented an integrated framework comprising a model for UGS supply-demand coupling coordination and a measure of urban vitality. Using downtown Beijing as a case study, the Gini coefficient assessed UGS supply-demand disparities across different vitality types. The study examined how UGS supply and demand factors interact with urban vitality, revealing the impact of UGS supply-demand imbalances on various dimensions of vitality and the UGS mismatches experienced by different vitality groups. The study showed that: (1) 63.29% of central Beijing’s areas had low UGS supply-demand coordination, with 39.23% experiencing UGS mismatches; (2) UGS supply and demand were significantly correlated with urban vitality spatial distribution; (3) these factors significantly impacted urban comprehensive vitality; (4) and there were notable UGS distribution disparities among vitality groups, with economic vitality group perceiving the greatest inequity (Gini = 0.311), followed by social vitality (Gini = 0.289) and cultural vitality group (Gini = 0.247). These findings offer valuable insights for a more refined assessment and enhancement of UGS, aiming to achieve balanced, high-quality, and sustainable urban development.

Sustainable Development Goal 11 emphasizes inclusive, resilient, and sustainable cities, in which high-quality urban green space (UGS) is essential. However, traditional two-dimensional (2D) metrics, such as Green Coverage Ratio (GCR), fail to capture the vertical complexity of UGS, leading to incomplete assessments in planning practice. With accelerating densification and land scarcity restricting horizontal expansion, sustainable UGS growth faces critical challenges. To address this gap, we developed a Three-Dimensional Green Quantity (3DGQ) mapping framework and an integrated 2D-3D assessment approach for urban green-space evaluation. We leveraged global open-source satellite data-including spaceborne LiDAR, canopy height, canopy coverage, and optical vegetation information-to retrieve growing-season Leaf Area Index (LAI) using a machine-learning model, and introduced the Three-Dimensional Green Coverage Ratio (3DGCR) to enhance 3DGQ performance assessment. Applying this framework to Beijing's central urban area, the tuned Random Forest model achieved an R² of 0.76, RMSE of 0.51, and MAE of 0.30 m²/m² for LAI retrieval, supporting 3DGQ mapping at resolutions as fine as 1 m. Results revealed a spatial divergence between 2D and 3D metrics: 3DGCR showed a "low-south, high-north" gradient, contrasting with the 2D "low-center, high-periphery" pattern. At the subdistrict level, 38.4% of units were classified as 2D-3D green imbalance under the coupling coordination framework. We identified four "2D-3D matching patterns" to provide spatial guidance for targeted UGS regeneration. Moreover, the equity assessment unveiled a "double inversion" between spatially and population-weighted results; specifically, 66.7% of subdistricts had lower population-weighted Gini coefficients for 3DGCR than for GCR. Our results further suggest that vertical UGS structural optimization in high-density cores may function as an "equity amplifier." These findings demonstrate that integrated 2D-3D assessment can uncover hidden urban green deficits and inequities overlooked by planar indicators, providing a diagnostic basis for structural optimization, equity-oriented planning, and sustainable urban green-space governance.