FIRE-WUI: Community Resilience Assessment Framework for Wildfires using Multi-Resolution Digital Twins

Wildfires pose a growing threat to communities across the US, particularly in areas where homes and natural landscapes meet, known as the Wildland-Urban Interface (WUI). Each year, more than 60,000 wildfires occur in the US, often resulting in tragic loss of life, destruction of homes, and severe economic disruption. Effective wildfire response and prevention efforts are hindered by a lack of detailed, up-to-date information about vegetation and buildings in these high-risk areas. To address these challenges, this project will develop a framework that creates detailed, 3D digital representations, referred to here as digital twins, of both natural and built environments. These digital twins will provide accurate, location-specific information needed to simulate how fires spread and enable the development of more effective mitigation strategies. The intellectual merit of this project includes the creation of a groundbreaking framework for generating scalable, multi-resolution digital twins that combine data from satellites, drones, and ground sensors. By integrating AI and advanced modeling techniques, the system will enhance the understanding of fire behavior in WUI environments and support more accurate fire-spread predictions. This research represents a convergent and interdisciplinary approach that combines geospatial science, wildfire modeling, and data analytics in novel ways still unrealized at this scale. The broader impacts of this project are significant. The digital twin framework will help improve community resilience and preparedness against wildfires by providing first responders, planners, and policymakers with the tools to make faster, data-driven decisions. The system is designed to scale nationwide and can be adapted to other types of natural disasters. The project will also prepare the next generation of experts through interdisciplinary education and training programs, including new courses, online learning modules, mentorship, and professional training developed in collaboration with industry organizations. These efforts will ensure a more knowledgeable workforce ready to confront the complex challenges of wildfire management. The “Community Resilience Assessment Framework Against WUI-Fires using Multi-Resolution Digital Twins – (CRAFT)” project aims to develop a scalable, dynamic digital twin framework to support wildfire modeling and mitigation in the WUI. CRAFT will integrate heterogeneous geospatial data from satellite, aerial, and ground-based platforms using AI-based super-resolution, generative modeling, and QA/QC protocols to create multi-resolution, continuously updated 3D models. These digital twins will feed into enhanced fire-spread models to improve prediction accuracy and investigate the trade-off among resolution, performance, and computational complexity. The improved models are then used to evaluate mitigation strategies. Validation will occur through testbeds with distinct characteristics in the U.S. Eastern and Western WUI regions. Educational components will support workforce development in geospatial and wildfire sciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.