A new method of three-dimensional green volume retrieval and its applications in urban greenery evaluation

Urban greening benefits plays a crucial role in promoting public health and aligns with regional development strategies. Three-Dimensional Green Volume (3DGV) has emerged as a precise metric for evaluating greening benefits, yet its large-scale application remains limited. This study developed a satellite-derived 3DGV retrieval model using multi-source remote sensing data to assess urban greenery in central Yunnan, China. Field measurements were collected to establish UAV-derived 3DGV retrieval models, integrating Canopy Height Model (CHM), Leaf Area Index (LAI), and Fractional Vegetation Cover (FVC) from RGB images. We established a power regression model in large scale retrieval linking these UAV-derived parameters to Sentinel-1 and Sentinel-2 images, achieving strong performance (R² = 0.72; RMSE = 139.71 m³/pixel; AE = 120.69 m³/pixel; MAE = 10.54 %). Spatial analysis was used to revealed the distribution of retrieved 3DGV, and it showed a pronounced west-to-east gradient (Moran’s I = 0.772) and an obviously increase trend from 2018 to 2022. This study demonstrated that Sentinel-1 and Sentinel-2 images enable accurate large-scale 3DGV mapping and reveal 3DGV dynamics to evaluate the greening benefits, providing a feasible and effectiveness approach for sustainable urban greenery evaluation and ecological management.