By Richard Lovett of Nature magazine

In the calculus of global warming, natural gas is generally considered to be preferable to coal as a fuel. That's because, on a per-joule basis, burning methane, the primary constituent of natural gas, produces less carbon dioxide than burning coal.

But, earlier this week, the conventional wisdom was shaken by researchers from Cornell University in Ithaca, New York, who argue in a study to be published in Climate Change that, over a 20-year period, the use of natural gas extracted from 'gas shales', porous rocks that hold the gas in minute pockets throughout the rock, could be worse for the climate than coal. Their study also estimates methane releases from conventional gas production and transportation over a 20-year period, which are less than from gas shales, but could also be significant. However, gas shales are expected to represent the bulk of US production of natural gas in the future.

Culprit methane

The additional emissions from shale gas arise because the process of extracting gas from the shale, called hydraulic fracturing or fracking, releases some of the methane directly into the atmosphere. The rocks are shattered by high-pressure injection of water in order to release the methane more easily. According to the new study, by biogeochemist Robert Howarth and his colleagues, between 0.6 and 3.2% of the gas can escape to the atmosphere during this process. And methane is a much more potent greenhouse gas than carbon dioxide.

This study is guaranteed to be controversial. Even before the paper was officially published, the natural-gas industry was denouncing it as "bunk".

"This study lacks credibility and is full of contradictions," Russell Jones, senior economic adviser for the American Petroleum Institute, based in Washington DC, said on his organization's website. "The main author is an evolutionary biologist and an anti-natural-gas activist who is not credentialed to do this kind of chemical analysis."

There are several basic problems with the study, Jones told Nature: the 20-year time horizon is too short; state-of-the-art power plants fired by natural gas are considerably more efficient than the coal-fired plants; and the data is of poor quality.

Time horizon

Independent experts agree - in part.

The biggest concern is the 20-year timescale. "It's an outlier position," says Henry Jacoby, former co-director of the Program on the Science and Policy of Global Change at the Massachusetts Institute of Technology in Cambridge. His overall opinion is that the study is "very weak".

The time-horizon problem stems from trying to compare carbon dioxide emissions, the primary climate culprit in coal burning, with methane leaks.

Although methane is a powerful greenhouse gas it breaks down in decades, unlike carbon dioxide which persists in the atmosphere for hundreds to thousands of years. So although methane is more than 70 times more powerful at heating the atmosphere than carbon dioxide over a 20-year period, says Michael Wang, senior scientist on life-cycle energy and environmental effects of energy production at Argonne National Laboratory, Illinois, after 100 years it's only 25 times more potent.

Not everyone is so critical. Alan Krupnick at Resources for the Future, an economic think-tank in Washington DC, notes that much of the controversy is due to media headlines that cite the study as showing that natural gas is worse than coal.

"That's not exactly what the paper says," he notes. "It presents two numbers, one for a 20-year time period and one for a 100-year time period." Overlooked in the media fray, Krupnick says, is the fact that even on a 100-year timescale, coal and natural gas from shale came up roughly comparable -- still a serious blow to the latter.

But that's not the only issue.

Jacoby thinks that the industry is correct to cry foul over the study's failure to account for differences in energy efficiency between coal-fired power plants and those using natural gas. "Howarth is making the wrong comparison," Jacoby says. Instead of greenhouse emissions per joules of combustion energy, he says, what should have been used are emissions per kilowatt-hour of electricity produced -- a statistic that would have favored natural gas.

The methane leaks are still important, but none of the experts consulted for this article were sure how much methane is lost during production. Krupnick notes that some gas fields have been drilled so many times the rocks have been compared to Swiss cheese. "There are all these old boreholes that can be a pathway to the atmosphere."

"That's an area that deserves study," Jacoby agrees. "The blowback of fracking does bring up gas."

Lifetime study

But flawed or not, experts say, the study is a reminder that combustion isn't the only aspect of fuel production and usage that should be taken into account in discussions on emissions. "Sometimes we forget that the debate should always be done on a life-cycle basis," Krupnick says. "I think the Cornell study is very important to remind people that the whole life cycle is what matters, not just the immediate emissions."

As for concerns about data quality, Howarth is the first to admit it's a problem. "I fully agree the data weren't what one would like. I'm hoping this will help free up the logjam of industry refusing to turn over data, so that one can get a better handle than we've been able to do so far."

This article is reproduced with permission from the magazine Nature. The article was first published on April 15, 2011.