Cities are complex three-dimensional environments where buildings, vegetation, terrain, and infrastructure shape microclimate, energy use, flood risk, air quality, and human exposure. Yet urban analysis has long relied on two-dimensional data that miss critical vertical structure. Advances in multimodal sensing and AI-driven reconstruction now enable high-resolution 3D characterization at scale.

This session invites contributions across the pipeline of 3D urban perception, reconstruction, and application. Topics include: (1) multimodal sensing with LiDAR, UAV, satellite stereo, SAR, and street imagery; (2) 3D reconstruction and scene understanding using depth estimation, neural radiance fields, point cloud segmentation, and voxel models; (3) derived metrics such as sky view factor, green view index, canopy volume, and building morphology; (4) integration with models of microclimate, energy, flooding, mobility, and air quality; and (5) urban digital twins for climate adaptation, environmental justice, and planning. Submissions using foundation models, transfer learning, and scalable pipelines are encouraged.

• B - Biogeosciences
• GC - Global Environmental Change