The satellite that lost contact still had the last word
MethaneSAT stopped transmitting in June 2025, but the year of data it collected has now produced the mission's first global assessment of oil and gas methane emissions, and it says the inventories regulators and investors rely on are undercounting.
February 2026 · Global
Retrospective analysis of February 2026. Published August 2026.
On 2 February 2026, MethaneSAT released the first global assessment built from the roughly fifteen months of data it collected before losing contact with mission control in June 2025 (see our earlier Field Note on that loss). The analysis, published through MethaneSAT's own project updates and a companion paper in Atmospheric Chemistry and Physics, covers more than 221 scenes across 45 oil and gas producing regions in 16 countries, together accounting for about half of the world's onshore oil and gas production. Measured emissions across the sampled basins ran roughly 50 percent higher in aggregate than the figures reported in the Emissions Database for Global Atmospheric Research and the US EPA's Greenhouse Gas Inventory, the two most widely cited emissions inventories. The gap was not uniform: gas-dominant basins, where natural gas accounts for at least 20 percent of energy production, measured roughly three times higher than their reported inventory figures, while oil-dominant basins measured about 30 percent lower.
The data also identified where regulation appears to change outcomes and where small, low-producing wells hide disproportionate emissions. New Mexico's regulated Delaware sub-basin showed emissions less than half those of the adjacent, less regulated Texas side of the same formation. Separately, only three gas-focused basins, Algeria's Hassi R'Mel and the US Appalachian and Haynesville basins, measured methane intensity below one percent, and even those fell short of the oil and gas industry's own stated 0.2 percent intensity target. Across eight major US basins, wells in the lowest-producing areas, responsible for less than 7 percent of production, accounted for about 40 percent of measured methane emissions in production-only zones.
The Signal
A posthumous global analysis of MethaneSAT's year of operation, released 2 February 2026, found methane emissions across 45 sampled oil and gas basins running about 50 percent above the figures in the EDGAR and US EPA inventories, with low-producing wells responsible for a disproportionate share.
Why It Matters
This is a direct, basin-level test of whether the emissions inventories that underpin corporate disclosures, carbon pricing, and buyer due diligence actually match measured reality, and in aggregate they did not. The finding that regulated basins measured meaningfully lower than unregulated ones in the same geological formation is a rare piece of evidence that specific regulatory design, not just general policy intent, measurably changes emissions outcomes.
The System Connection
MethaneSAT's own operational life ended in June 2025, the loss we covered at the time. This assessment is proof that a monitoring asset's data value can outlive the asset itself, provided the archive is preserved and analyzed, a distinction diligence teams evaluating any single-mission monitoring commitment should weigh separately from the mission's operating status.
What We Are Watching
- Whether inventory compilers such as EDGAR and national GHG registries revise methodology in response to basin-level satellite measurement gaps this large.
- Whether other satellite operators (GHGSat, Carbon Mapper, Sentinel-5P) publish comparable basin-level assessments that corroborate or complicate MethaneSAT's findings.
- Whether the New Mexico-Texas regulatory comparison holds up as more basin pairs with differing state rules get measured.
