I've written before about the biggest problem when it comes to fracking and climate change: methane leaks. Natural gas is a much cleaner fuel than coal, which theoretically could be useful in cleansing our electricity generation system of the worst pollutants, even if natural gas is not nearly enough by itself to stop climate change. But because it is a much more powerful greenhouse gas than carbon dioxide over the short term, the release of methane could cancel out any benefits natural gas might provide.

A new study is the latest to confirm that view. Absent regulation, fracking could very well be a net negative when it comes to climate change. But there are some new wrinkles to the story that suggest preventative regulation could be cheaper and easier than we think.

Here are the results:

During two days of intensive airborne measurements, oil and gas operations in Colorado's Front Range leaked nearly three times as much methane, a greenhouse gas, as predicted based on inventory estimates, and seven times as much benzene, a regulated air toxic. Emissions of other chemicals that contribute to summertime ozone pollution were about twice as high as estimates, according to the new paper, accepted for publication in the American Geophysical Union's Journal of Geophysical Research: Atmospheres. [CIRES]

What does this mean? First of all, there are two ways of measuring such leaks. The EPA has based its leak estimates by examining samples at every point in the production process (the drill, the well head, the pipelines, etc.), and then adding up what they found. That's what "inventory" refers to.

This study, by contrast, flew a plane over the extraction sites and measured the difference between upwind and downwind methane levels, thereby measuring leaks from every source in an entire area. They found, in concert with previous measurements, that leaks were much higher than the EPA method would suggest.

I spoke to the Environmental Defense Fund's* chief scientist Steven Hamburg, who calls these two methods the "bottom-up" and "top-down" approaches. He emphasizes that both are necessary to provide a complete picture of the emissions situation. The EPA's bottom-up approach may not be enough to get a complete picture, but it will be necessary to figure out where the leaks are happening and plug them.

So where are the leaks? Without more research it's hard to say for sure, but he suspects they may come from a tiny minority of "super emitters" that are leaking vastly more than everyone else. If true, this would explain why the EPA's estimate came in low, since you would need a very large sample size to capture a small minority of huge emitters. It would also fit with other pollution profiles, like that of cars, in which a tiny fraction of vehicles are responsible for the preponderance of emissions.

If this speculation is correct, then the solution is obvious: Just imitate those states that test for annual emissions to make sure no car is an egregious polluter. In this case, EPA folks will have to go well to well and make sure every drilling operation is within tight limits.

The upside is that the vast majority of operations likely will already be in compliance, so they won't have to do any expensive upgrading. Only a small minority will have to refit and retool. That would be nearly painless, so fingers crossed.

Hamburg says that there are several more studies on methane leaks that will be released throughout the year. I'll be keeping a close eye on them as they're released. Watch this space.

*Full disclosure: The EDF partially paid for this study, which was carried out by an independent research team.