As a partner of the US National Oceanic and Atmospheric Administration, we at Western Energy Alliance provided much of the data for the reported study on methane emissions from a natural-gas field in Colorado's Denver-Julesburg Basin (Nature 482, 139–140; 2012). However, uncertainties remain over the extent of methane leakage from these operations.

For example, this study (G. Pétron et al. J. Geophys. Res., in the press) relied heavily on data from mobile-laboratory samples taken in summer 2008, before regulations reduced emissions of volatile organic compounds to 90%, and before low-bleed pneumatic devices were implemented to reduce methane emissions from the field. Much of the data were collected in an operating environment that no longer exists. Also, the venting you describe is at odds with industry practices.

The authors acknowledge that methane emissions from agriculture are of “comparable magnitude to the estimate from natural gas systems” in Colorado, but do not consider these.

Pétron et al. should have noted that methane readings were lower at the Boulder observation tower than at a tower in a Wisconsin forest that is far removed from significant industrial activity.