A UAV-Based Urban Digital Twin Framework for Building-Resolving Urban Climate Modeling

Abstract. Building-resolving urban climate models require high-resolution urban morphology datasets that are often labor-intensive to prepare from heterogeneous geospatial sources. Meanwhile, urban digital twins have emerged as promising platforms for integrating geospatial data, environmental observations, and numerical models to support climate-resilient urban planning. However, existing studies generally treat these components independently, with limited integration of UAV-based data acquisition, urban morphology preparation, climate-model preprocessing, and digital twin visualization within a unified geospatial workflow. This paper presents a UAV-based urban digital twin framework for building-resolving urban climate modeling that integrates UAV LiDAR data acquisition, three-dimensional city modeling, urban morphology generation, and web-based visualization into a systematic and reproducible workflow. The framework was implemented in the University of the Philippines Diliman Engineering Complex using UAV LiDAR and photogrammetric surveys. Point cloud processing generated Digital Terrain Models (DTMs), Digital Surface Models (DSMs), Canopy Height Models (CHMs), and LoD1 three-dimensional city models, while a Random Forest classifier produced an initial classification of major surface classes to support subsequent urban morphology generation. An interactive web-based digital twin prototype was developed using deck.gl to visualize the generated three-dimensional urban models and geospatial datasets. The proposed framework establishes a common high-resolution geospatial foundation for integrating UAV-derived products with complementary thematic datasets required for building-resolving climate modeling. The framework provides a transferable methodology for integrating geospatial data acquisition, urban climate model preparation, and digital twin development, contributing toward the development of digital twins for climate resilience and evidence-based urban planning.