A new study from the Environmental Defense Fund (EDF) finds methane emissions in the Barnett Shale are lower than what a previous study assumed. EDF claims this is because the methods used in the new study – including a component-based approach – are not capturing so-called methane “super-emitters,” and thus those methods, which are also used by the U.S. Environmental Protection Agency, underestimate total emissions. But a closer review suggests that this new study may actually undercut the super-emitter argument.

Before digging into the finer details, the presence of super-emitters is what EDF and other advocacy groups have used to try to justify costly new methane regulations from the EPA and other federal agencies. Their theory is that only a handful of sites are responsible for most of the emissions, but it’s difficult to determine which sites those are, and thus a blanket, one-size-fits-all regulation is needed to cover the entire system.

In the case of this study, EDF is comparing its results against a previous study in the Barnett, which used a site-based approach that attempted to measure total methane from a given site. The component-based estimates in the new study examine emissions from all equipment and operations present at production sites.

According to EDF, the fact that the component-based numbers are lower than the previous site-based estimates suggests that the component-based data are not capturing the super-emitters. EDF is assuming its previous study in the Barnett – which it admitted was “biased toward high-emitters” – is accurate, so the differences in the new study can only be explained by data that are not being captured.

In other words, EDF’s key finding is based off of something they did not observe, but rather assumed into existence in order to explain what they did find: significantly lower emissions.

Let’s take a deeper dive into the report and the data, shall we?

Confirming super-emitters are anomalous

The fact that the component-based inventories show lower emissions than site-based inventories pretty much confirms that “super-emitters” are episodic. EDF even attributes them to “abnormal process conditions.”

As noted above, the 2015 Barnett Shale study (against which this new EDF study compared its findings) was, as the report puts it, “biased toward high-emitters.” Even so, such sites still made up just two percent of all the facilities in that report.

Another EDF study collected data from 8,000 well pads in seven top U.S. oil and gas basins and found that just four percent were high-emission sites.

So not only are “super-emitters” more of the exception than the rule, they are exceedingly rare. To be fair, EDF doesn’t really dispute this. As they wrote on their blog describing their previous Barnett Shale study, “1% of natural gas production sites accounted for 44% of total emissions from all sites.”

If anything, this data gap suggests that more research is needed to definitively determine the source responsible for the discrepancies EDF found between component-based estimates and its site-based estimates.

Site-based estimates may be incorrect

EDF’s topline conclusion is premised on the assumption that its site-based estimates are correct. But as we also know, that study specifically sought out super-emitters, which EDF admits can be episodic. From the latest report’s accompanying blog,

“Because our new work tells us that super-emitter sites are characterized by abnormal behavior that is unlikely to persist indefinitely, we expect that different sites will be in the high-emitting group at different points in time.”

So it stands to reason – or it’s at least a reasonable possibility – that these site-based estimates were taken during episodic events, and thus not representative of what emissions actually are.

Those limitations of the site-based approach are largely ignored in the latest study. In fact, by assuming that something must exist to explain super-emitters, EDF is letting a conclusion get ahead of what the data indicate.

In summary, EDF’s site-based estimates could very well have been made during an episodic emissions event, which would explain why the component-based data in this new report show significantly lower emissions from the same region. But EDF takes the existence of super-emitters as a given, and uses the discrepancy in emissions data to try to justify the difficult nature of finding those emitters.

This new study does not make a definitive case for or against super-emitters. It does, however, suggest that additional research needs to be done to reconcile the different emissions figures. One could just as easily use the lack of information on super-emitters to say they’re difficult to find (as EDF has) as one could to question whether they even exist. Assuming that one explanation is better than the other, especially when the point of comparison (the 2015 Barnett Shale study) has its own limitations, would be premature at best.