Satellites Pass Biggest Ever Controlled Test of Methane Detection From Space

In the largest controlled experiment ever conducted to test methane monitoring from orbit, satellites designed to track the potent greenhouse gas correctly identified nearly all of the methane plumes released on clear days, according to a yearlong study led by Stanford University scientists. The findings, published October 14 in the journal Nature, offer the most rigorous real-world benchmark yet for the rapidly expanding fleet of space-based instruments that regulators, environmental groups, and energy companies now rely on to find leaks and flag so-called super-emitters of the gas responsible for a substantial share of near-term global warming.

For decades, the promise of spotting methane leaks from hundreds of kilometers above Earth has been accompanied by persistent doubt. Methane is invisible to the naked eye, and detecting it from space requires measuring faint absorptions of reflected sunlight in narrow infrared wavelengths, then distinguishing a genuine plume from noise in the data. Senior study author Adam Brandt, a professor of energy science and engineering in the Stanford Doerr School of Sustainability, said the technical difficulty of the problem has fueled significant skepticism over the years about whether methane tracking really works. The new results, he argued, should give regulators, policymakers, methane producers, and the public considerable confidence moving forward.

To put that confidence on firm statistical ground, Brandt, lead author and PhD student Frances Reuland, and collaborators staged 481 controlled methane releases at a research site in Casa Grande, Arizona, throughout 2025. Each release was timed to coincide with overpasses by 11 government, nonprofit, and commercial satellites. The releases were generated by piping compressed natural gas, which is mostly methane, through a release stack, with rates controlled through an online interface to approximate the point sources typically seen at landfills and oil and gas facilities. Individual releases lasted eight to ten minutes and ranged from roughly 10 to 1,200 kilograms of methane per hour.

Fifteen independent research teams drawn from corporations, nonprofits, government agencies, and universities across nine countries then analyzed the satellite observations without knowing the true release rates, assessing whether each image contained a methane plume and, if so, estimating how much gas had escaped. When the researchers compared these blind estimates with the known quantities, a clear hierarchy emerged. Instruments purpose-built for methane monitoring delivered the strongest performance, correctly detecting the presence of 92 to 100 percent of the test plumes under favorable conditions.

Quantification, while less precise than detection, proved serviceable. Most estimates based on the dedicated instruments’ observations fell within about 50 percent above or below the actual emission amounts, and the overestimates and underestimates roughly balanced out on average. That statistical balance matters, the authors said, because it makes the estimates reliable enough to flag large leaks for follow-up inspection, to inform the regional greenhouse gas inventories that governments compile, and to compare the relative emissions intensity of different oil and gas producing regions. In many cases, Brandt noted, detection itself is the crucial outcome: as long as operators know approximately how big a leak is, they still know they need to go fix it.

Under ideal conditions, with few clouds and moderate winds, the dedicated sensors reliably detected plumes when methane flowed at rates down to about 50 to 200 kilograms per hour, depending on the system. That sensitivity range is significant because it brackets the threshold of 100 kilograms per hour that triggers mandatory reviews under some state and national super-emitter programs. Reuland, a member of Brandt’s lab in the Doerr School of Sustainability, said the results were great to see and that the team hopes they build confidence in the field overall.

The study also exposed clear limits. Satellite methane detection depends on infrared light reflected off Earth’s surface, so even the specialized instruments aboard the GHGSat and Tanager-1 satellites struggled to see plumes through clouds. The authors noted that this poses problems in some key methane-emitting regions, including tropical oil and gas producers Nigeria and Venezuela, as well as cold, cloudy climates like Canada and Russia. More research, they added, is needed to understand how the satellite systems and their algorithms perform over non-desert terrains, across varied weather conditions, and over open water, since the Arizona test site offered predominantly clear desert skies.

Instruments repurposed for methane sensing showed a wider performance spread. Germany’s Environmental Monitoring and Analysis Program satellite, known as EnMAP, detected 62 to 88 percent of plumes when skies were clear. General-purpose land-scanning systems aboard United States and European government satellites fared far worse, catching just 29 to 37 percent of releases and proving especially error prone for emissions below 1,000 kilograms per hour. The reason is spectral: these land-imaging instruments gather far less of the spectral data required to isolate methane’s unique absorption signature. Their extensive coverage and long historical archives still hold value, the researchers said, for documenting long-term trends and identifying very large leaks.

False positives received their own rigorous test. The teams were given observations from 265 overpasses during which no methane was released at all, allowing the researchers to measure how often algorithms cried wolf, a problem that can erode public trust and waste resources chasing non-existent leaks. False positives proved relatively rare with most methods, with one striking exception: 79 percent of all false positives originated from commercial teams applying emerging artificial intelligence approaches to data from land-scanning government satellites. The researchers said the AI methods tested need improvement, and that the Casa Grande dataset, which contains orders of magnitude more controlled-source data points than ever before available for testing space-based methane detection, could help refine them.

The experiment’s environmental footprint was deliberately modest. Over the full year, the team released less than 50 metric tons of methane, equivalent to the annual emissions of a herd of about 500 dairy cows, or roughly 13 seconds of methane emissions from global fossil fuel energy operations. The stakes of getting this monitoring right are considerable. Methane is the biggest driver of climate change after carbon dioxide, and although it does not persist as long, it traps about 80 times more heat than carbon dioxide during its first 20 years in the atmosphere, giving it an outsized influence on near-term temperatures. Rapidly cutting methane emissions is a major focus of international efforts to slow global warming.

Fixing leaks also pays off beyond the climate. Some relatively low-cost repairs to oil and gas infrastructure, one of the largest sources of human-caused methane emissions, can yield financial gains for operators by keeping saleable gas from escaping, the authors said. Brandt pointed to additional motivations: methane is flammable, creating safety hazards, and it contributes to ground-level ozone, a health hazard. Supporting satellite monitoring, he argued, is a win-win-win for the environment, the producers, and the general public. With the technical question of whether these systems work now largely answered for clear-sky conditions, Reuland and her co-authors hope the conversation shifts toward how, given algorithmic trade-offs, and where satellite methane monitoring can be most effectively deployed to drive global methane mitigation. Co-authors were affiliated with the University of Michigan, Ann Arbor; the University of California, Berkeley; the University of Iowa; and the Max Planck Institute for Chemistry, with analysis teams including Carbon Mapper, the International Methane Emissions Observatory, NASA JPL, SRON, GHGSat, Google, Kayrros, and others. Funding came from the Global Methane Hub, the Grantham Foundation, Saudi Aramco, and the Stanford Knight-Hennessy Scholars program.

Subject of Research: Evaluation of satellite-based methane emission detection using controlled release experiments

Article Title: Satellites accurately detect methane emissions in large-scale test

Article References: Satellites accurately detect methane emissions in large-scale test. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: methane, satellites, greenhouse gas emissions, remote sensing, controlled release experiment, Stanford University, Nature, super-emitters, oil and gas, artificial intelligence, climate change, leak detection