Livestock and climate: Whose numbers are more credible?
An argument has been brewing over the past half-decade about a key environmental-impact assessment: The role of livestock production in climate change, specifically in greenhouse-gas (GHG) emissions.
Livestock and climate: Whose numbers are more credible?
The science of assessing environmental impact is a relatively young one, having only gained traction over the past few decades. Methodologies are still being developed and perfected, and best practices are still in the process of being definitively established. So it's no surprise there is disagreement among experts as to the exact environmental impact of a given industry, with the "fracking" controversy being only the latest high-profile example.
Meanwhile, however, an argument has been brewing over the past half-decade about a key environmental-impact assessment: The role of livestock production in climate change, specifically in greenhouse-gas (GHG) emissions. I've been following this off and on and it's time to take a hard look at it.
At the outset I will remind you of my potential bias (especially since such transparency, or lack thereof, comes into play here): As a vegan, I deplore the institution of animal-based foods and know that they are unnecessary for humans to eat. You may well have a different attitude about the livestock industry, which is why I encourage everyone to read everything at the links herein and see what conclusion you come to.
To recap: Livestock's Long Shadow (LLS), a 2006 report from the UN's Food and Agriculture Organization, analyzed the livestock sector and concluded that it was responsible for 18% of GHGs, more than all of human transport. In late 2009, the Worldwatch Institute released a recalculation by Jeff Anhang and Robert Goodland that found some impacts uncounted and concluded, rather, that the livestock sector was tied to 51% of GHGs - in other words, more than all other anthropogenic factors combined.
I did one E2P post on the report and refrained from going back and expounding on it after that because, given my own veganism, I wanted to err quite heavily on the side of restraint. Subsequent developments have spurred me to reexamine the issue, and I now think it's of paramount importance that everyone interested in environmentalism use something close to the 51% figure, or else present a new and compelling case against it.
It's not that there weren't cases made against the Worldwatch report (including by some vegans) very soon after publication. However, just about all of these concerns, which echoed and/or amplified some of my own, were answered in a Spring 2010 point-by-point follow-up by the two authors
At the same time, the FAO was forced to admit an error in its own calculations, pointed out in a (beef-industry-funded) UC Davis report: In its LLS assessment, the agency had included "indirect" impacts of livestock - associated or related sectors that are causally linked - but had not done the same in the case of transport. Many news organizations gleefully blared headlines that meat was now off the hook for killing the planet.
Where things get interesting is in a 2011 commentary by some of the LLS authors called "Livestock and greenhouse gas emissions: The importance of getting the numbers right" ('Herrero et al.' henceforward). Readers who may have logically expected a clarification of the data in dispute, or alternately, a discussion of different methodologies for verifying the validity of such data, may have been surprised by the content: Herrero et al. ignored the FAO controversy and concentrated instead on attacking the credibility of the Worldwatch report and lauding the "well documented" and "widely recognized" LLS, without (and here's where that transparency thing comes into play) disclosing the very salient detail that this comparison of the two studies was being co-authored by the lead author of Livestock's Long Shadow itself (plus one of that study's other authors).
Then, on its first reference to the Worldwatch report, Herrero et al. call it "a recent non-peer-reviewed report." The implication is damning: Unlike LLS, the latter report lacks scientific rigor. However, the claim is patently false. As an employee of the World Bank Group’s International Finance Corporation, Jeff Anhang was required to have peer review on any report he had his name on. Via email, I pressed Goodland and Anhang on this question and received details on the researchers and instutions who had reviewed the Worldwatch draft prior to publication, as well as those that have cited it subsequently. On the other hand, Livestock's Long Shadow may or may not have been peer-reviewed. FAO cites no such process (nor does the Herrero et al. commentary) and I was unable to find any reference to peer review in any coverage of LLS. I emailed Mario Herrero for clarification on this but have received no reply.
Similarly, on the issue of livestock respiration, which LLS omitted but Goodland and Anhang have argued for including, Herrero et al. claim that "under the Kyoto Protocol, CO2 from livestock is not considered a net source of CO2." That statement is clearly false, as the listing of manure in Kyoto's "Annnex A" shows. If Herrero et al. meant to say "CO2 from livestock respiration," they would then have needed to specify why this would be excluded from the "other" listed under "agriculture." Again, I asked Herrero for clarification on this seemingly errant claim, and will update if I receive a response.
But it's not just a couple weird false statements. "The Importance of Getting the Numbers Right" as a whole is written in a very odd, haphazard style, with relevant criticisms (founded or not) freely intermixed with those that are irrelevant.
For instance: "[Goodland and Anhang] do not quantify the lost opportunity for carbon sequestration that results from other forms of land use, such as arable crop production for human consumption, or urban development." The first of these other forms of land use is very relevant to the discussion. Urban development, though, has nothing to do with the topic at hand.
And later, the commentary claims Goodland and Anhang's proposal to replace some consumption of animal foods with that of plant foods "would contribute to habitat destruction of native grasslands, an ecosystem that harbors a number of species at risk." Whether or not that's true, the subject here - i.e. the numbers that it's imporant to get right - is livestock's impact on GHGs, not all possible environmental impacts of livestock (which are, of course, legion). Threatening at-risk species' ecosystems is way outside of the topic area.
Even more oddly, in the midst of explaining a detailed distinction between the terms "poultry production" and "poultry biomass," the authors suddenly zoom back out to say, "It should also be mentioned that despite some shortcomings of the FAO statistics, FAO still remains the only globally recognized source of data on agriculture," as if needing to remind themselves of their own validity. It's a sentiment properly expressed in the intro or conclusion that almost seems to have been accidentally copy-pasted into the poultry paragraph.
As it happens, it's exactly that emphasis on animal agriculture expertise that may undermine those authors' case for higher credibility in terms of overall eco-assessment. The journal in which their commentary appeared is "Animal Feed Science and Technology," distributed, as you'll see in the PDF linked above, by the International Livestock Research Institute, whose slogan is "Better lives through livestock." Hmmmm, no possiblity of pro-livestock bias there, right?
More to the point, as Goodland and Anhang pointed out in a "Comment" that ran as a letter to the editor of that journal, none of the authors of Livestock's Long Shadow were environmental assessment (EA) specialists - a discipline devoted specifically to, well, assessing environmental impacts. Since generating such an assessment was the entire point of LLS, this would likely have been a good idea. Goodland and Anhang, who are EA specialists (Goodland, for example, served as president of the International Association of Impact Assessment), make this case by pointing to the infamous transport error:
It appears that evidence arose of the need for EA specialists to be involved in assessing livestock and climate change when the authors of Livestock’s Long Shadow were forced to retract their comparison of livestock related GHG emissions to transport related GHG emissions, after it was revealed that their estimate of livestock related GHGs involves some lifecycle analysis, but their estimate of transportrelated GHGs involved none, only CO2 from the tailpipes of all types of transportation vehicles (The Atlantic, 2010). We propose that such an elementary error would have been less likely to occur if an EA specialist had been involved in the writing of Livestock’s Long Shadow.
Later, the two pinpoint what bothered me the most about the 2006 report:
[Herrero et al.] assert that livestock are needed to provide livelihoods and food for people; they project continued growth in the livestock sector and recommend no measure to restrict its growth; and they propose no strategy to achieve a net reduction in GHG emissions attributable to the livestock sector worldwide. They continuously fail to consider concepts considered essential in professional environmental assessment – notably the avoidance of impacts and analysis of alternatives, which would examine more environmentally sustainable pathways than livestock to provide livelihoods and food for humans.
As anyone who read it knows, LLS constantly claims, without providing numbers to back it up, that "greater efficiencies" in livestock production will eliminate the threat of GHGs, thus no reduction in consumption will be necessary. Not only is this data-free Pollyanna projection unscientific, but they are clear on what they mean by efficiency: More highly-concentrated factory farms. How many of us truly believe that's the best solution in the years to come?
Something shared in common by Goodland, Anhang and most of the team Herrero et al. is employment within the UN system. Moreover, one UN agency has used the lower GHG estimate favored by Herrero et al. to recommend a global shift away from eating animal products for many reasons, the key being that growing plants and feeding them to animals instead of people is an inherently, grossly inefficient way of feeding people. Constantly making little "efficiency" tweaks to such a system is about as impactful as adding a decorative green garnish to a cow patty.
It seems that the FAO's next report might do well to begin with that baseline - reducing consumption - rather than with its current assumption that the livestock industry is untouchable. While recycling and changing light bulbs are at best ways to "get into the spirit" of environmentalism rather than high-impact remedies, altering our diet can have a huge impact - yes, one that may indeed be larger than all other changes combined. Goodland and Anhang make this point in their own conclusion:
While governments continue to struggle to agree on measures that would increase renewable energy infrastructure significantly, we propose that alternatives to livestock products could be scaled up quickly to reduce today’s grave risk of climate change significantly. Indeed, reducing animal feed production and replacing at least one quarter of today’s livestock products with substitutes could be the only way for governments, industry, and the general public collaboratively to take to take a single, powerful action to reduce climate change quickly.
UPDATED 10:45 p.m. to clarify some terminology characterizing scientific journal articles and U.N. agency structure.
From G&A's "point-by-point followup":
"We discovered after publication that the FAO’s own statistical division reported 56 billion livestock worldwide in 2007. This is many more than are counted in our article, and doubtless outweighs whatever the increase would be in non-livestock methane."
As a vegan myself, I would love to be proved wrong, but as far as I've been able to determine this is a serious flaw in G&A's analysis. I've never come across estimates higher than 60 or 65 billion for total number of farmed animals SLAUGHTERED/CONSUMED globally per year. But the number or quantity that matters when it comes to analyzing "livestock's long GHG shadow" is the average number of farmed animals - aka, "the herd size" - living and breathing at any given time (multiplied by the GHG footprint of the different species of animals, e.g., very high for every cow; much, much lower for every chicken). Most farmed animals have a short life. In North America, for example, most of the 10 billion or so animals slaughtered per year are "broiler" (meat) chickens, and their average lifespan is about 42 days. Similarly, pigs only live a few months before they're ready to "go to market." What that means is that at any given moment on a continent that consumes ~10 billion farmed animals per year, the number of living, breathing, methane- and CO2-emitting farmed animals is probably closer to 2 billion than to 10 billion. This is consistent with the choice by the authors of LLS of a number of livestock worldwide of approx. 22 billion. G&A claim that LLS used this figure for the total number of farmed animals "RAISED worldwide in 2002" (emph. mine). I've not been able to verify this figure when searching the pdf of LLS. Regardless, if you or anyone else can provide documentation that LLS significantly underestimated the "herd size" of the world's livestock population, please do. Otherwise, I remain unconvinced of the validity of this materially very important argument by G&A.— S_Gordon_B- Goodland and Anhang's analysis, while it does provide some useful correctives, is deeply flawed. They do indeed grossly overestimate the world herd size (mistakingly treating the total numbers of animals raised for food annually as the standing population of animals throughout the year) and very strangely include farm animal respiration (but not the respiration of humans and every other (non-farm animal) living being on the planet) in their accounting. Basically, they apparently fail to understand the carbon cycle. I've written a couple of constructive criticisms of their work. This one covers their strengths and shortcomings fairly well in an easy to read way: http://www.earthsave.ca/files/2010_Winter_Earthsaver_web.pdf— dsteele
- It is not ethical for a vegan to evaluate impact of livestock on climate. However honest a person is, he/she can't separate his/her feelings and beliefs from the subject's matter. It's just a human nature. Therefore, any such study or opinion can't be neither acceptable nor trusted. It's the same as with any other similar conflict of interest. I, for one, wouldn't trust any research or opinion on how safe is fracking, if this study/opinion was expressed by someone within the industry. Same applies here.— hollandpa
- Mr. Lehmkuhl notes that UNEP recommends more meat reduction than do Goodland and Anhang based on FAO's numbers. This provides just a hint of how comments by dsteele and S_Gordon are not constructive as they pretend to be. They both falsely claim that GHG accounting by Goodland and Anhang is based on livestock raised annually instead of livestock standing at any one time. Anyone who carefully reads the analysis of Goodland and Anhang can see that their GHG accounting is based precisely on FAO's acounting of livestock standing at any one time; they comment on the implications of different sets of statistics, but they do not increase their GHG accounting to reflect livestock raised annually. The idea that their work is deeply flawed is contraindicated by the treatment of their work by groups such as the FAO itself, and the editors of Animal Feed Science and Technology journal. Animal rights activists should forthrightly make a case - if they see fit - that it's more important to consider the rights of animals than the impacts of climate change. On the other hand, people may logically see going vegan as an ethical action after evaluating the impact of livestock on climate, and it seems unfair to judge that they couldn't thereafter be objective. For example, while dsteele implies that Earthsave doesn't support analysis by Goodland and Anhang, Earthsave's founder John Robbins has expressly supported and promoted their analysis - e.g., at vegnews.com/articles/page.do?pageId=2178&catId=5— World Watcher
- I'm not pretending, and I'm trying to be constructive in the sense that I believe accurate advocacy is more constructive and socially responsible than inaccurate advocacy.
You write: "Anyone who carefully reads the analysis of Goodland and Anhang can see that their GHG accounting is based precisely on FAO's acounting of livestock standing at any one time..." I've read their analysis carefully and am unable to see where they give a traceable/verifiable estimate (FAO-based or otherwise) of the standing population or "herd size" of livestock globally. All I can see is a straw man: their claim that LLS drastically lowballed the annual number of animals raised and killed at just 21.7 billion. THAT would indeed be a highly inaccurate number. But where is the evidence that this is what LLS actually says? I can't find it. If you can, please direct me to it, and I will happily stand corrected.— S_Gordon_B - At the top of p. 96 of Livestock's Long Shadow, a table states "Livestock numbers (2002)" where it should have clarified that it meant livestock alive at any point in time in 2002. Mr. Steele has written: "Goodland and Anhang counted emissions from each of those lives as if each had lasted all year." But they didn't. He's apparently citing a paragraph 4 pages into their article, where they briefly discuss some differences in various published statistics. It seems the intention of this paragraph is to say that where FAO and NGOs provide various numbers but don't distinguish between animals alive at any point in time and animals killed per year, they open up a potential for huge differences in GHG tallies – even above 10 percent in difference. But they actually add 0 emissions to their GHG tally as a result of this paragraph. This can be understood by looking above and below that paragraph, where they say they're adding 8.7% in GHGs for all "other sources" –- and the 8.7% is actually composed of 4% and 4.7% from other paragraphs in that section.— World Watcher
- I agree that the numbers in that table implicitly represent "livestock alive at any point in time in 2002." But this merely backs up my point. If you add all the numbers up, you get about 22 billion animals, which is what one would expect in a world where, as G&A themselves suggest, over 50 billion farmed animals are raised (and killed) per year. As I wrote in my first comment, "G&A claim that LLS used this figure [21.7 billion] for the total number of farmed animals 'RAISED worldwide in 2002' (emph. mine)." Having made this error, G&A then proceed to "correct" LLS. To quote from their paper: "If the true number is closer to 50 billion than to 21.7 billion, then the percentage of GHGs worldwide attributable to undercounting in official livestock statistics would likely be over 10 percent." — S_Gordon_B
- To put it another way, G&A's "correction" implicitly assumes that LLS was basing their GHG estimates on a number of "livestock alive at any point in time in 2002" capable of producing just 21.7 billion animals per year. That would imply an LLS figure for "livestock alive at any point in time in 2002" of less than 10 billion. — S_Gordon_B
- In Goodland and Anhang's original article, they fundamentally accepted FAO's 21.7 billion global "snapshot" statistic for 2002, except they factored in what they understood had been a 12% increase in the global population of livestock from 2002 to 2009. They included some text to alert readers to what can happen if/when one conflates the 21.7 billion "snapshot" figure with the 50 billion "annual" figure that was used by some NGOs in 2009. In their point-by-point follow-up, they reported that they found FAO itself using an "annual" figure of 57 billion -- which both flouted their caution regarding conflating "snapshot" with "annual" statistics, and exceeded the 12% increase that they had factored in. In any event, you may be correct that the ratio of "annual" vs. "snapshot" statistics is about 5:1 in the U.S. -- but animals in most other parts of the world are raised with much less genetic engineering and intensification than in the U.S., suggesting that the FAO's global ratio of about 2.5:1 makes sense. Anyway, there are virtually no sets of livestock statistics in the world as sophisticated as FAO's, so everyone is probably stuck with them whether they like it or not.— World Watcher
- World Watcher, this is verbatim from G&A's original article:
"The report [LLS] also states that 21.7 billion head of livestock were raised worldwide in 2002, while many nongovernmental organizations report that about 50 billion head of livestock were raised each year in the early 2000s. If the true number is closer to 50 billion than to 21.7 billion, then the percentage of GHGs worldwide attributable to undercounting in official livestock statistics would likely be over 10 percent."
"21.7 billion head of livestock were raised worldwide" means "they fundamentally accepted FAO's 21.7 billion global 'snapshot' statistic for 2002"?
I'm not getting it. 2+2 doesn't equal 7.
Also, I never speculated that "the ratio of "annual" vs. "snapshot" statistics is about 5:1 in the U.S." It may well be, given the extremely high proportion of extremely short-lived chickens and, as you point out, the shorter lives in general in intensive livestock production. But I agree that a global ratio of 2.5:1 sounds reasonable and is consistent with LLS's estimate being based on a global "herd size" of 21.7 billion in 2002 when probably around 50+ billion animals were being "produced" per year.
No matter how you crunch the numbers, livestock has a disturbingly long shadow, but in this instance, I can find no support for G&A's argument that 10 points should be added to the LLS estimate because it erroneously "states that 21.7 billion head of livestock were raised worldwide in 2002." It doesn't state that. It's G&A who are erroneous on this point. Let's suck it up and base our estimates on the facts, which are damning enough as it is.— S_Gordon_B - My bad. I see you perhaps misread a 5:1 ratio where I wrote "at any given moment on a continent that consumes ~10 billion farmed animals per year, the number of living, breathing, methane- and CO2-emitting farmed animals is probably closer to 2 billion than to 10 billion." — S_Gordon_B
- To repeat, they didn't add 10 points due to the difference in statistics. They added zero points. In fact, in the section of their World Watch article that you're referring to, they did exactly what you are doing; that is, they merely cautioned that anyone who doesn't properly distinguish between snapshot numbers and annual numbers can end up with a large error in GHG tallies, even above 10%. You can see they added 0 percent themselves to their GHG tally by looking in that section above and below the paragraph in which their cautionary note appears. You'll see they say they're adding 8.7% in GHGs for all "other sources" in that section –- and the 8.7% is actually composed of 4% and 4.7% from other paragraphs in that section.— World Watcher
- I see what you mean now about the most important point: while they float the figure of 10% in that passage, for some reason (I know you think you understand why, but I don't, unless they "mispoke themselves") it's nowhere to be found in their tally in the table at the top of the paper and, as you point out, it's clearly not included in the tonnage of GHGs they tally in the "other sources" section where the passage is found. But I continue to read the passage the same way as I did before. For any readers who may be following this, here it is again in the full context of the paragraph it's in. Judge for yourselves:
"Second, the FAO and others have documented frequent undercounting in official statistics of both pastoral and industrial livestock. Livestock’s Long Shadow not only uses no correction factor for such undercounting, but in some sections actually uses lower numbers than appear in FAO statistics and elsewhere. For example, Livestock’s Long Shadow reports
that 33.0 million tons of poultry were produced worldwide in 2002, while FAO’s Food Outlook of April 2003 reports that 72.9 million tons of poultry were produced worldwide in
2002. The report also states that 21.7 billion head of livestock were raised worldwide in 2002, while many nongovernmental organizations report that about 50 billion head of livestock were raised each year in the early 2000s. If the true number is closer to 50 billion than to 21.7 billion, then the percentage of GHGs worldwide attributable to undercounting in official livestock statistics would likely be over 10 percent."— S_Gordon_B
The experts at Philadelphia's Energy Coordinating Agency answer your energy questions in our regular feature Stay Warm, Stay Green. Send in your question or questions to energy@phillynews.com.
Look for Jenice Armstrong to supply tips on green living as well as occasional columns on the subject of Green. She also blogs at Hey Jen.
Becky Batcha stays tuned for the here-and-now practical side of conservation, alternative energy, organic foods, etc. - stuff you can do at home now. Plus odds and ends.
Laurie Conrad recycles from her ever-growing e-mailbag to pass along the latest travel deals, fashion statements, household strategies, gadgets, cool local events and other nuggets of interest to those who appreciate a clean, green world.
Vance Lehmkuhl looks at topics like eco-conscious eating, public transportation and fuel-efficient driving from his perspective as a vegetarian, a daily SEPTA bus rider and a hybrid driver, as well as noting the occasional wacky trend or product. Contact Vance with your 'green' news.
Ronnie Polaneczky sees the green movement through the eyes of her 12-year-old daughter, who calls her on every scrap of paper or glass bottle that Ronnie neglects to toss into the house recycling bins. Ronnie will blog about new or unexpected ways to go green. She also blogs at So, What Happened Was...
Sandra Shea and the DN editorial board opine on any green-related legislation or policy. And we'll pass along some of the opeds on the subject that people send us.
Jonathan Takiff will be blogging mainly about consumer electronics - those things that we love to use and that suck too much energy. He'll spotlight green-conscious gizmos made in a responsible fashion, both in terms of materials used and the energy it takes to run them.
Signe Wilkinson draws the comic strip Family Tree, which follows the Tree family as they try to live green in the face of nattering neighbors, plastic-wrapped consumer products, and the primal teenage urge to spend vast quantities of money on hair care products of dubious organic quality.
In addition to these updates from our newsroom bloggers, watch for an occasional feature, Dumpster Diver Dispatches, from Philadelphia's original "green" community of artists, the Dumpster Divers. You'll learn about creative ways to reuse and recycle while you reduce, and about the artists who are making little masterpieces from what others throw out.
Sept. 17: A head start on cold season
Oct. 8: Help with fuel bills
Oct. 29: Last-minute ways to winterize
Nov. 19: Budgeting for the big bite
