For the last four years, media outlets such as The New York Times, the Los Angeles Times, and Fox News have repeatedly cited a United Nations study which found that livestock production is responsible for about 18 percent of global greenhouse-gas emissions—a larger share than comes from all planes, trains, and automobiles combined.

Last week, news outlets revisited those claims, following a talk delivered by Dr. Frank Mitloehner, an animal scientist based at UC Davis, at an American Chemical Society meeting last Monday. Mitloehner criticized the conclusions of a 2006 report from the U.N. Food and Agriculture Organization, “Livestock’s Long Shadow.” The study’s assertion that meat (including eggs, dairy, and other animal protein) production is responsible for more greenhouse-gas emissions globally than the transportation industry is certainly untenable, and likely false, he said.

The problem Mitloehner highlighted is that the FAO performed what’s called a full “lifecycle analysis” for the livestock industry. In other words, it added up emissions from things like fertilizer production and land-use change in addition to those from cow burps and manure. It did not do the same for the transportation industry, however, tabulating the emissions from fossil-fuel combustion, but not from automobile manufacturing or road construction, for instance. Mitloehner has spent the last week explaining to reporters that this led to a classic apples-to-oranges comparison between the two sectors. In one of the first news articles on the subject, the FAO substantiated this criticism, saying that its comparison was indeed flawed.

“I must say honestly that he has a point – we factored in everything for meat emissions, and we didn’t do the same thing with transport, we just used the figure from the [Intergovernmental Panel on Climate Change],” FAO livestock policy officer Pierre Gerber, a co-author of “Livestock’s Long Shadow,” told the BBC. “But on the rest of the report, I don’t think it was really challenged.”

Unfortunately, most articles did not do a good job covering the rest. As the Knight Science Journalism Tracker astutely observed:

Some reports, however, interpret the asserted flaw and the concession from one of the UN-affiliated authors that it holds water to mean that climate science and climate worry has been stripped further of legitimacy. Which is to say, by implication, if it’s not climatically worse than using cars and other vehicles, then eating meat is as green as a bean sprout.

Few reporters, however, seem to have the wit to have homed in on the 18 percent figure. If that’s right, it’s a lot. If the figure is a serious overestimate, that’s a story. But to make the story only whether transport or meat are bigger boosts to the greenhouse is lazy.

In fact, some of the coverage displayed worse flaws than a narrow focus on the relative emissions of livestock and transportation. Not only did most reporters not scrutinize the 18 percent figure, some totally mischaracterized Mitloehner’s opinions about it. Rather than reporting that he thinks it might be an overestimate, they left the impression that he believes there is no link whatsoever between livestock production and warming. For example, Fox News misled readers in the second paragraph of its article:

The largely reported link between global warming and cattle farming — propagated by a United Nations report on “Livestock’s Long Shadow” — was also largely inaccurate, explains one scientist.

Fox News’s headline, “Eat Less Meat, Reduce Global Warming – Or Not,” while not glaringly wrong, was also inaccurate. Mitloehner does not think curtailing livestock reduction would make no dent in warming; he just thinks it would be an insignificant dent and that there are far more effective ways to reduce emissions. (A glaringly wrong headline, by the way, would be the one in The Washington Times, which read, “Meat, dairy not tied to global warming.”)

Some of the confusion seems to trace back to a press release from the American Chemical Society after Mitloehner’s talk at one of its meetings. The release quoted him saying, “We certainly can reduce our greenhouse-gas production, but not by consuming less meat and milk. Producing less meat and milk will only mean more hunger in poor countries.”

Mitloehner believes his original quote was slightly different, and his main concern is that the FAO’s flawed assertion that livestock accounts for more emissions than transportation will lead policymakers and consumers to make the wrong choices.

“I didn’t say that there is no reduction in greenhouse gases associated with animal protein consumption, but that it is a relatively small contribution and that consumers can do other things that have greater impact on this,” he said in an interview.

(Mitloehner’s talk was based on a peer-reviewed paper he co-authored last fall. The press release issued by UC Davis noted that the report “was supported by a $26,000 research grant from the Beef Checkoff Program, which funds research and other activities, including promotion and consumer education, through fees on beef producers in the U.S. Since 2002, Mitloehner has received $5 million in research funding, with 5 percent of the total from agricultural commodities groups, such as beef producers.” The more recent release from the American Chemical Society did not include this detail, and neither did many news articles, with the exception of a blog post in the Guardian. The post made it clear that the author did not think Mithloehner was “in the pocket of Big Beef,” and that his criticisms of the FAO merited further investigation, but rightly criticized other outlets for overlooking an important detail of the story.)

So what about that 18 percent figure, then? Mitloehner is indeed skeptical about its accuracy, but he does not claim to be certain that it is wrong. He suspects that the FAO did not do a lifecycle analysis of the transportation industry because the data needed to do so is simply not available (which the FAO later confirmed). Without that information, Mitloehner argues, a comprehensive analysis is difficult.

“If you want a global assessment, then you really need those detailed emissions numbers,” he said. “We might have them for the United States, Germany, and Britain, and so on, but I question whether we have them for Paraguay, Ethiopia, and Afghanistan. And if you don’t know what the pieces of the pie are, then how do you know what 100 percent is? And if you don’t know what 100 percent is, then how can you say livestock is 18 percent?”

Gerber, the livestock policy officer at the FAO and co-author of “Livestock’s Long Shadow,” conceded that the comparison between livestock and transportation emissions was deeply flawed, but defended his organization’s estimate of the former’s relative contribution to total global emissions. “We still stand entirely behind the 18 percent figure,” he said.

The FAO will post a “technical note” (and issue a press release) on its Web site this week, responding to Mitloehner’s criticism and explaining what was and wasn’t included in its assessment. Gerber argues that the Intergovernmental Panel on Climate Change’s figure for the total global emissions, which the FAO used for its calculations, is accurate, and that because the FAO used the same methods as the IPCC to do its lifecycle analysis of the livestock industry, comparing its subset to the total is fair. The lack of a lifecycle analysis for the transportation industry doesn’t change that, he added, arguing that such analyses have a greater effect on how the emissions pie is carved than on the size of the pie.

“In the IPCC’s analysis, you have everything—all transport, all agriculture, deforestation, manufacturing. You name it; it’s there,” Gerber said. “But if you were to add up lifecycle emissions for every industry, you would come up with more than 100 percent because of overlap [in other words, part of the livestock industry involves transportation, and part of the transportation industry involves livestock, so certain activities get counted more than once]. In depends on how they’re partitioned. So, lifecycle analyses are mostly a problem only when trying to compare different sectors.”

For instance, Gerber said the IPCC attributes only about 5 percent of global emissions to livestock (out of roughly 14 percent from agriculture overall) because it did not perform a lifecycle analysis like the FAO did.

Nonetheless, a recent report from the National Research Council (which advises the government on matters of science), titled “Verifying Greenhouse Gas Emissions: Methods to Support International Climate Agreements,” lent some credence to Mithloehner’s concerns about uncertainty in the IPCC’s estimate of total global emissions:

In many developed countries, uncertainties are reported to be less than 5 percent for national CO2 emissions from fossil-fuel use, which is the dominant source. With the exception of a few minor sources in the industrial sector, uncertainties are much higher for other greenhouse gases and sources and vary greatly by country. Uncertainties for the net CO2 emissions from agriculture, forestry, and other land uses and for emissions of CH4, N2O, PFCs, HFCs, CFCs, and SF6 from all sectors can be less than 25 percent in some countries and greater than 100 percent in others.

Stephen Pacala, an ecologist at Princeton University who chaired the committee that published the report on verifying emissions, couldn’t be reached to comment on the merits of Mitloehner’s and Gerber’s arguments.

[Update, 3/30: Reached today, Pacala said that we know, unequivocally, how much greenhouse gas is in, and is being added to, the atmosphere each year.

“If you use as the measure the atmospheric totals, or increase in atmospheric totals, then we know exactly what they are to within a couple percent for all these gases,” he said.

Some uncertainty comes into play when talking about the amount of certain gases that comes from anthropogenic versus natural sources, however. Take nitrous oxide. Scientists know exactly how much is in the atmosphere and that the amount is increasing steadily. “But we cannot attribute that global increase accurately to different sources, nor can we disaggregate that N2O geographically,” Pacala said. The situation is similar for methane and the so-called F-gases (HFCs, PFCs, etc.). Carbon dioxide, on the other hand, is very well understood and scientists have a firm handle and how much is being added to the atmosphere from various sources.

Because carbon dioxide accounts for slightly over three-quarters of the IPCC’s estimate of total global greenhouse-gas emissions (based on a 100-year global warming potential), that estimate is fairly reliable. Nonetheless, Pacala said, it might be better to base the relative contribution of industries like livestock on total carbon-dioxide emissions alone.

“If you really want to do people a service, you just take as a denominator something that is indisputably known, like fossil CO2 emissions,” he said. “And you say, okay, we’re going to evaluate everything relative to that. Then the uncertainty is in the thing you’re talking about – not in something else.”]

Given the availability of such information and perspective, it is unfortunate that reporters did not do more to analyze the dispute over the 18 percent figure. On the other hand, the conflict narrative isn’t ideal either, and there might be a richer story to be found in places where Mitloehner and Gerber see eye-to-eye.

For all his skepticism, Mitloehner concedes that, globally, the 18 percent may or may not be that far off. But, applied regionally, it is certainly a misleading number, he argues, and that is his main beef (pun intended) with the figure.

“In Paraguay, the contribution of livestock may be as high as 50 percent because they are clear-cutting a lot of forest, and that basically takes a unit of [greenhouse-gas] sequestration away and puts cattle, which is an emissions source, there instead. In the U.S. the contribution from livestock is only around 3 percent of the total,” Mitloehner said. “So, I don’t think that the 18 percent number is all that meaningful because it doesn’t apply regionally. It does not apply to Ethiopia, it does not apply to Paraguay, and it does not apply to the United States. Yet it’s being used in these countries [to influence policy and consumer choices].”

On this point, Gerber agrees. “[The 18 percent] is a global figure that covers a wide variety of farming systems and development objectives,” he said. “Different countries have different first priorities in terms of feeding their people, or preserving environment, or protecting public health. So, using that 18 percent figure to try to draw conclusions for the U.S., or for the U.K., or for the Sudan is of course wrong.

“This is why there is a major effort here now, which is requiring much more work, to disaggregate this 18 percent into various commodities – eggs, milk, beef, poultry, and so on – in different farming systems and different regions to understand where are the emissions and where are the big pockets of reduction potential. We’re also coupling that with economic work to see what the most cost-effective options are to reduce environmental impact.”

Mitloehner agrees that the FAO is taking admirable steps to improve its reports. He called the press’s single-minded focus on conflict between him and the organization unfortunate, and stressed that despite its faulty comparison between livestock and transportation emissions, the FAO is doing valuable research.

“We are predicting a massive increase in the demand for animal protein, so the question is: how do we satisfy that demand without having a major environmental impact? That’s why what the FAO is doing is so laudable,” Mitloehner said. “Livestock’s Long Shadow’ was a good first attempt and they are working on follow-up studies right now. I’ve had the privilege to look at some of them and they are much better because they are refined by region. So, the follow-up reports will talk about how the United States livestock industry compares to those in, say, sub-Saharan Africa, or India, or Brazil, and why some areas have a higher environmental impact per unit of meat or milk produced than others.”

Gerber noted that the FAO will not do country profiles, however. “We won’t disaggregate by country because our data is not accurate enough and I think there is also the risk of pointing fingers,” he said. “So, for now, we will do political regions, we’ll do climatic zones, we’ll do farming systems, but across boundaries.”

Mitloehner and Gerber complained, respectively, that the media’s focus on conflict and attempts to cast the former’s criticism as yet another blow to climate science are “regrettable” and “pour oil on the fire.” The two men clearly disagree on some fundamental points, and the FAO’s comparison between livestock and transportation emissions is indeed a newsworthy mistake. But the impression left by most news articles is one of scientists bogged down in disagreement—when, in fact, there is a lot of consensus about how research could be improved.

Indeed, it seems fair to say that Mitloehner and Gerber agree on three points: the need to do a lifecycle analysis for the transportation industry; the need to “disaggregate” livestock statistics; and the need to reduce the environmental impact of both industries.

“I think it’s time that we work across the globe really on transferring knowledge and help particularly those areas like India and China to produce in a way that is as environmentally benign as possible,” Mitloehner said. “I think we have that responsibility. So it would be nice if we would take some of the politics out of the discussion and really focus on getting things done and resolved and addressed.”

If you'd like to get email from CJR writers and editors, add your email address to our newsletter roll and we'll be in touch.

Curtis Brainard is the editor of The Observatory, CJR's online critique of science and environment reporting. Follow him on Twitter @cbrainard.