Collaborative research: caig: integrating artificial intelligence and neotectonics for data-driven global active fault mapping

The National Science Foundation Office of Integrative Activities awarded the Board of Regents of Nevada System of Higher Education, doing business as University of Nevada, Las Vegas (UNLV), $589,714 under the Geosciences program on October 1, 2026, for collaborative research integrating artificial intelligence and neotectonics for data-driven global active fault mapping.

The project will develop an artificial intelligence framework for mapping and interpreting active fault networks across tectonically variable regions, combining geophysical and geological constraints within physically informed learning models to infer fault properties. The work will employ interpretable learning approaches to advance beyond simple feature detection toward uncertainty-aware and physically meaningful fault-system mapping, with resulting workflows benchmarked against well-characterized fault systems. Reproducible tools and datasets will be released for broader community use. The research addresses gaps in fault mapping worldwide, supporting improved understanding of earthquake hazards and creating training opportunities for students at the intersection of computing and earth science.

Performance occurs in Las Vegas, Nevada. The period of performance extends from October 1, 2026, through September 30, 2029. No subawards are planned. This is a Project Grant (assistance type).

Abstract

Earthquakes can cause major loss of life, damage infrastructure, and disrupt communities, yet many active faults around the world remain poorly mapped or not mapped at all. This project addresses the need for better, more consistent ways to identify and understand active fault systems. The project will develop artificial intelligence tools that help researchers map faults more accurately by combining information from a variety of sources. The work will support improved understanding of earthquake hazards. IT will also create training opportunities for students at the intersection of computing and earth science. This project will develop an artificial intelligence framework for mapping and interpreting active fault networks and characterizing the dynamics of those fault systems across tectonically variable regions. The research will integrate geophysical and geological constraints within physically informed learning models to infer fault properties. The project will integrate interpretable learning approaches to move beyond simple feature detection toward uncertainty-aware and physically meaningful fault-system mapping. The resulting workflows will be benchmarked against well-characterized fault systems. Reproducible tools and datasets will be released for broader community use. 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.- subawards are not planned for this award.