AI Identifies Thousands of Undocumented Orphaned Oil Wells Across the U.S.

Researchers are using artificial intelligence and historical maps to locate hazardous wells that threaten the environment and public health.

Undocumented orphaned oil and gas wells (UOWs) across the U.S. are leaking pollutants like methane, benzene, and hydrogen sulfide, posing environmental and health risks.
AI systems, trained on digitized historical topographic maps, are helping researchers locate these wells by identifying specific symbols used to mark them on maps from 1947 to 1992.
In a case study of four counties in California and Oklahoma, researchers identified 1,301 potential UOWs, with 44 confirmed through satellite imagery and fieldwork.
The Consortium Advancing Technology for Assessment of Lost Oil & Gas Wells (CATALOG) is developing tools like drones, methane sensors, and magnetometers to verify and assess the wells' conditions efficiently.
Experts estimate there could be between 310,000 and 800,000 undocumented wells nationwide, with ongoing efforts to refine AI methods and scale up detection technologies.