US methane intensity falls faster than total emissions as oil and gas output grows
US oil and gas production rose during 2019–2024, while methane emissions per unit of output fell more sharply than the national total. A peer-reviewed satellite and modeling study estimates an 11.1 per cent intensity reduction alongside a 3.5 per cent decline in total emissions. The final annual change remains uncertain. The observed trend cannot isolate the effect of federal regulation.
Science··Midday
Output growth offsets some intensity gains
US oil and gas production increased by 8.5 per cent from 2019 to 2024, while methane emissions per unit of production declined by 11.1 per cent. Estimated total emissions fell by 3.5 per cent. More production therefore accompanied a smaller overall reduction than the intensity measure alone suggests. Emissions declined through 2019–2023; the estimated smaller rise in 2023–2024 had an uncertainty interval including zero.[1]
Satellite measurements reconstruct the sector’s emissions
The peer-reviewed study uses satellite observations and atmospheric modeling to examine methane from US oil and gas production during 2019–2024. Its accounting distinguishes output-normalized intensity from total emissions.[1], [2]
TROPOMI and GOSAT, satellite instruments measuring atmospheric gases, supplied observations. Models worked backward from methane concentrations to estimate sources. Alternative modeling assumptions tested sensitivity to transport, observation coverage and prior information. This approach provides regional estimates rather than a direct inventory of every leak at individual facilities.[1]
Federal rules leave an uncertain causal comparison
Researchers detected no abrupt national decline after federal methane regulation. Implementation delays affected the period under examination, while emissions without those measures are unknown. The observed curve thus cannot establish that regulation had no effect. Regional differences also limit what the national average says about each operator. The study is observational and model-based; it does not randomly compare otherwise equivalent groups with and without regulation.[1]