Methane Emissions Inventories Underestimated?


29th February 2012.

A recent study by researchers at the NOAA, Boulder, Colorado (1) suggests that the figures used in GHG inventories for methane emissions from oil and gas operations may be significant underestimates.

The NOAA study, led by Gabrielle Pétron, was prompted by results from air samples collected daily at the 300m tall tower located at the NOAA Boulder Atmospheric Observatory between 2007 and 2010. An analysis of the tower data filtered by wind sector indicated a high alkane and benzene signature from the direction of the Denver-Julesburg Basin, where, over the past forty years, over 20,000 oil and gas wells have been drilled. A survey in the summer of 2008, using a mobile laboratory unit, targeted air samples directly downwind of potential hydrocarbon sources in the region (oil and gas wells, a landfill, feedlots, and a waste water treatment plant). Multi-species analyses of these data strongly implicated the oil and gas operations as the emissions culprit.

When the research team went on to compare their top-down estimates of emissions from the oil and gas production activities with the bottom-up assumptions on which the state and national emissions inventories are based, they uncovered significant discrepancies. In particular, the team believes that the methane source from the Denver-Julesburg natural gas facilities is “most likely underestimated by a factor of at least two” by the EPA state inventory tool.

This startling conclusion prompts the question of whether methane emissions from oil and gas production operations have in general been underestimated. The issue is also highly relevant to the current debate about shale gas ‘fracking’ in the USA and beyond. The same technology used in the shale gas industry to release gas from the rock has also been deployed extensively in the Denver-Julesburg Basin wells.

Those interested in mitigating greenhouse gas emissions see tackling methane leaking from oil and gas operations as a potential win-win situation given that any gas retained can be sold for profit, and that the technology to reduce leakage – so-called ‘green completions’ for example – is readily available. However, not all those involved in estimating methane emissions are yet convinced that the results reported by the NOAA team are representative of oil and gas operations in general, or can be extrapolated to the shale gas industry. Sergey Paltsev, leader of an ongoing MIT-base study on shale gas emissions, is reported as saying the results are “surprising” (2).

Pétron and co-authors concede that their study does not pinpoint exactly where in the production process the gases come from, and that key uncertainties remain unresolved. Second author Greg Frost says “it’s certainly true that we need to measure in other basins” and adds that the research team is “making measurements right now in the Uintah Bain in northeastern Utah.” He also tells MethaneNet that the NOAA team plan to complement the tall tower and mobile unit measurements with continuous methane sampling by small aircraft, and to target the specific sources within the fossil fuel producing areas in detail. So work is already underway to address several issues raised by the Pétron et al. paper, and to improve estimates for methane and other climate and air quality relevant trace gases emitted by the hydrocarbon industry.


(1) Pétron, G., et al. (2012), Hydrocarbon emissions characterization in the Colorado Front Range: A pilot study. J. Geophys. Res., 117, D04304, doi:10.1029/2011JD016360.

(2) Tollefson, J. (2012), Air sampling reveals high emissions from gas field. Nature, 482, 139-140, doi: 10.1038/482139a.

Photo by Adam Hirsch, now at the National Renewable Energy Laboratory (NREL).