Collaborative Research: GEER - Collecting and Archiving AI-ready Perishable Data for Geotechnical Engineering Natural Hazards Research

This collaborative project addresses the urgent national need to increase community resilience and protect critical infrastructure against severe natural hazards, such as earthquakes, hurricanes, floods, and wildfires, which consistently cause significant loss of life and economic disruption. While laboratory experiments and computer simulations provide essential theoretical insight, many of the most important engineering advancements stem directly from observing and documenting real-world infrastructure performance during actual extreme events. Collecting this perishable ground-truth field data immediately following a disaster is critical before evidence is lost or cleaned up. Building on over two decades of experience, the Geotechnical Extreme Events Reconnaissance (GEER) organization has coordinated and supported post-disaster teams of geotechnical engineers and geoscientists to rapidly collect perishable field data after disasters and contributed to significant scientific discoveries. This project serves the national interest by advancing GEER activities into the modern age of digital technology and data science. By establishing standardized workflows to collect, document, and publish open-access, machine-readable datasets, the initiative enables researchers and engineers nationwide to benefit from a thorough process aimed at archiving field-performance observations during extreme events. The resulting discoveries will inform improved engineering design methodologies, enhance building codes, and support mitigation of infrastructure downtime during future disasters. Furthermore, the project encourages broad involvement in virtual deployment teams, engaging students, engineers, and public stakeholders to support disaster recovery and regional emergency planning. The project aims to advance post-disaster geotechnical reconnaissance by establishing integrated workflows for collecting and curating artificial-intelligence-ready perishable data. To accomplish this goal, the project introduces an updated organizational architecture featuring specialized scientific advisory groups spanning extreme weather, earthquakes, ground and infrastructure instability, and emerging technologies. Field operations will adopt an integrated three-tier structure consisting of a field deployment team, a virtual data-mining team, and a central communications unit. The virtual team will leverage crowd-sourced information, social media streams, and satellite remote sensing to support field planning and populate a centralized public event clearinghouse. In close partnership with national cyberinfrastructure and experimental facilities, the project will establish standardized protocols to ensure all gathered datasets adhere to principles of findability, accessibility, interoperability, and reusability, enabling direct integration into machine learning models without extensive data preprocessing. Methodological enhancements will be documented through updated field reconnaissance manuals, ethics protocols, and a range of educational, outreach, and information webinars, establishing long-term impact metrics to track the subsequent scientific utilization of published hazard data. 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.