UN uses AI to scale methane leak detection and spark global reductions

The United Nations Environment Programme (UNEP) is leveraging artificial intelligence to identify and eliminate major methane leaks at a scale previously impossible for human analysts. By integrating AI into its Methane Alert and Response System (MARS), the UN is converting massive streams of satellite data into immediate mitigation actions to slow global warming.

Since becoming fully operational in 2024, MARS has contributed to more than 40 methane mitigation actions worldwide. The system integrates data from more than 30 satellite instruments and has analyzed over 1.3 million observations since 2023. According to UNEP, AI-assisted workflows identified 80% to 85% of confirmed methane detections before they reached expert review. This automation allows International Monitoring and Evaluation Office (IMEO) analysts to process 12 to 15 times more satellite data while maintaining scientific rigour.

The urgency of methane mitigation

These technological gains come at a critical time for climate strategy. Methane is approximately 80 times more powerful than carbon dioxide over a 20-year period, though it remains in the atmosphere for only about a decade. Because these emissions primarily originate from the waste, agriculture, and fossil fuel sectors, the ability to detect and stop leaks rapidly is essential for short-term climate stabilization. This effort is supported by the Global Methane Pledge and the UN Secretary-General's Call to Action on Methane, which have increased pressure on corporations and governments to act on emission alerts.

Bridging the gap to action

The primary challenge has not been a lack of data, but the human capacity to process it. The volume of satellite imagery has grown beyond the ability of manual review, creating a bottleneck between detection and repair. By using energy-efficient AI to filter environmental "noise" from actual leaks, the UN can now provide actionable notifications to the responsible parties in near real-time.

"The real lesson from this work is that AI can help convert the explosion of environmental data into practical action," said Martin Krause, Director of UNEP's Climate Change Division.

Global impact and economic potential

The results are already measurable. Mitigation efforts driven by the system have addressed sources estimated to have emitted 1.2 million tonnes of methane. UNEP estimates this climate benefit is equivalent to removing nearly 24 million gasoline-powered passenger cars from the road.

Beyond the environmental impact, the broader use of such data holds significant financial implications. A World Economic Forum study, conducted in collaboration with Deloitte, estimates that the aggregate economic potential of Earth observation data could surpass US$700 billion by 2030.

What remains

As MARS continues to scale, the focus shifts to the speed of response from the governments and companies receiving the alerts. While the AI has solved the detection bottleneck, the ultimate success of the system depends on the political and corporate will to implement the necessary repairs on the ground.