Transforming from 2D to 3D: A Review of Urban Green Space Visualization and Analysis Methods at Different Scales

With the development of city information modeling, digital twins, and smart cities, traditional 2D-based urban planning progress is now being transformed into 3D panoramic spatial planning. Current urban green space modeling and analysis methods largely rely on 2D landscape patterns and indicators, which cannot express the vertical characteristics of green spaces. This study reviews the commonly used methods for modeling green spaces in 3D environments and analyzes their applicability. Three case studies at different scales – regional, community, and site – are discussed, and an automated modeling and analysis process for each case study based on deep learning and parametric modeling tools is then constructed. The three case studies include the estimation of national-scale 3Dvegetation volume using GEDI for different cities in China, the construction of voxel models for green spaces in a middle school campus based on LiDAR point cloud data, and the application of L-system generative algorithms for simulating and predicting tree growth in the dormitory area. The proposed 3D green space modeling and analysis methods could support green space’s 3D modeling and analysis needs at different scales.