Your diet is destroying the world.

Regular readers of this blog will already know this fact. But the more I read, the more I grow convinced that the data I have cited on here has downplayed the problem. Most figures put meat consumption as contributing to about a fifth of global emissions. Try changing that figure to more like a half. Mark Bittman explains:

Five years ago, the United Nations Food and Agricultural Organization published a report called “Livestock’s Long Shadow,” which maintained that 18 percent of greenhouse gases were attributable to the raising of animals for food. The number was startling.

A couple of years later, however, it was suggested that the number was too small. Two environmental specialists for the World Bank, Robert Goodland (the bank’s former lead environmental adviser) and Jeff Anhang, claimed, in an article in World Watch, that the number was more like 51 percent. It’s been suggested that that number is extreme, but the men stand by it, as Mr. Goodland wrote to me this week: “All that greenhouse gas isn’t emitted directly by animals.  ”But according to the most widely-used rules of counting greenhouse gases, indirect emissions should be counted when they are large and when something can be done to mitigate or reduce them.”

Robert Goodland himself illuminates his own findings:

The key difference between the 18 percent and 51 percent figures is that the latter accounts for how exponential growth in livestock production (now more than 60 billion land animals per year), accompanied by large scale deforestation and forest-burning, have caused a dramatic decline in the earth’s photosynthetic capacity, along with large and accelerating increases in volatilization of soil carbon…

[R]eplacing at least a quarter of today’s livestock products with better alternatives would both reduce emissions and allow forest to regenerate on a vast amount of land, which could then absorb excess atmospheric carbon to reduce it to a safe level. This may be the only pragmatic way to reverse climate change in the next five years as needed.

Elsewhere, Goodland makes similar points and adds:

[A]n astonishing 45% of all land on earth is now used for livestock and feed production. So we propose that, contrary to popular belief, the key to reversing climate change in the next five years — as needed — is actually the food industry.

More from Goodland here. But here’s Bittman again:

It’s seldom that such enormous problems have such simple solutions, but this is one that does. We can tackle climate change without inventing new cars or spending billions on mass transit or trillions on new forms of energy, though all of that is not only desirable but essential.

In the meantime, we can begin eating less meat tomorrow.

Reading all this, I don’t think there’s any way I can continue to hold this half-way house position which commits me to trimming meat consumption without eliminating it. I have consoled myself with self-deceiving weasel words about how I’m doing more than most, but there’s just no way anyone can become conscious of the consequences of their actions here and not feel compelled to stop eating chicken and beef and bacon. Through weakness of will, of course, we may slip back into these bad habits. But there’s simply no intellectual defence available here. Every time I buy meat, I send market signals which help to ensure we continue to fuck over future generations. And I’m supposed to justify this how? By insisting that the pleasure of tasting steak is more important than preventing the deaths of millions of people through floods? Please. Again, there’s no real moral dispute here. Everyone who knows the facts must know they should change. It’s just that being moral is hard, and when the consequences are as complex, convoluted and distant as they are with climate change, it’s easy to rationalise away one’s duties.

The lazy get-out is to invoke the necessity of protein, as if it isn’t perfectly possible to get it elsewhere. Yes, it takes effort and vast dietary reform. But learning to love lentils and discovering the delights of alternative foods like quinoa needn’t be difficult, and with care and attention the food can be equally rich in taste, if not richer. In that respect, Ottolenghi’s New Vegetarian blog archives are going to become my new best friend. Having to take the time to plan one’s diet and shop more at Holland and Barrett rather than the butcher’s is a small burden when put aside the stringency of the duty at hand here. And before any vegans jump in – yes, I need to look more into the status of dairy produce, and I need to firm up my understanding of the situation with fish. But, baby steps. In the mean time, no more meat. If you know me personally, you can hold me to that. And please do.


4 thoughts on “Your diet is destroying the world.

  1. 1) “More to the point, livestock (like automobiles) are a human invention and convenience, not part of pre-human times, and a molecule of CO2 exhaled by livestock is no more natural than one from an auto tailpipe.” (from the article you quoted)
    It doesn’t matter at all whether a carbon source is “natural” or not to be considered a carbon source. 100% of the carbon emitted by livestock (both methane & CO2) was acquired from the plants they eat, 100% of the carbon in those plants was acquired by them from atmospheric or surface-oceanic (algae) CO2 within a relatively short time. Where as the carbon emitted by cars comes from burning fossil fuels (usually) which was stored by plants over the course of millions of years. Therefore livestock respiration is immediately offset by the plants where as burning fossil fuels takes millennia to be offset by plant growth. So by their own calculation you should knock off 13.7%
    2) “vast amounts of potential carbon absorption by trees has been foregone.”
    Unfortuneately most natural forest are infact carbon neutral not carbon sinks (recall that before humans started burning fossil-fuels the atmospheric CO2 wasn’t decreasing). Are only carbon sinks while they are growing and soil carbon storage varies by ecosystem type but generally plateaus when the ecosystem stabilizes. Reclaiming land will only undo the carbon emissions from their original conversion to farmland and even that takes centuries or millennia for many ecosystems.
    3) “Or suppose that land used for grazing livestock and growing feed were used instead for growing
    crops to be converted moredirectly to food for humans and to biofuels.”
    Biofuels from corn release barely CO2 from fossil fuels (being conservative many claim they actually release more) than burning fossil fuels due to the fertilizer & processing required. Also notice they use a comparison to coal the worse fossil fuel rather than fuels like natural gas which may emit less GHG than biofuels.
    4) Their “other sources” is basically them picking the worse values where-ever uncertainty exists which will over inflate their numbers.
    5) “Third, the FAO uses citations for various aspects of GHGs attributable to livestock dating back to such years as 1964, 1982, 1993, 1999, and 2001. Emissions today would be much higher.”
    Excuse me for saying but this sounds like “we think it should be higher so we made it higher”.
    6) “Fourth, the FAO cites Minnesota as a rich source of data. But if these data are generalized to the world then they under-state true values, as operations in Minnesota are more efficient than operations in most developing countries where the live-stock sector is growing fastest.”
    OTOH Minnesota is much more industrialized agriculture compared to developing countries which will inflate GHG emmissions due to fertilizer & other chemical inputs, and reduces carbon sequestration due to use of animal wastes as fertilizers due to the high density production.

    So of their four undercounted inventories I found major issues with their accounting in three (I would examine their methane calculations as well but I don’t have time to do the research). Thus >=26.6 of their Percentage of world wide total should be immediately removed. And several of their “misallocated” I have issues with as well (another 4.7% which is highly suspect).

  2. Ms. Andrews writes as if analysis by Goodland and Anhang (which she critiques) has never been cited before. Perhaps at Oxford it hasn’t; in my country of Canada, it’s been widely cited, in Canadian sources such as the Journal of Sustainable Agriculture, Macleans, Canadian Dimension, the Montreal Gazette, and Le Devoir. I’ll provide the links to those 5 sources below.

    First I’ll reply to each of Ms. Andrews’ substantive points. She begins with a good, concise summary of the carbon cycle that many people learn at a young age: “100% of carbon emitted by livestock was acquired from atmospheric or surface-oceanic CO2 within a relatively short time-frame.” I suppose Ms. Andrews may not have seen that Goodland’s explanation of why the carbon cycle no longer operates as it once did has been published in the highly-regarded journal Nature Climate Change — having passed that journal’s editorial review — as follows: “[R]espiration has increased exponentially with livestock production, while intensified livestock and feed production accompanied by large-scale deforestation and forest-burning have caused huge increases in volatilization of soil carbon, resulting in a dramatic decline in the Earth’s photosynthetic capacity. Therefore, either carbon dioxide in livestock respiration, or its reflection in carbon debt created where land is used for livestock and feed production, must be counted as emissions.” It can be seen at

    The importance of carbon debt, as described by Goodland, has also been described in a video featuring an FAO specialist at

    Ms. Andrews’ second point involves a belief that reclaiming land can’t be effective in the near term. Indeed, that belief is so common that land reclamation has been almost entirely overlooked as a measure to address climate change. In fact, robust assessment has explained how within 5 years, enough land can be reclaimed to reduce atmospheric carbon to a safe level, at

    Ms. Andrews’ next criticism is to question (sensibly) why Goodland and Anhang have focused on replacing coal instead of replacing natural gas. The answer is available in widely-available data showing that worldwide coal production has increased by 7% each year from 2009-2012.

    Next, Ms. Andrews cautions that Goodland and Anhang may have included in their analysis anything that would inflate numbers. That’s a good cautionary prescription for anyone reading any quantitative analysis. But in the case of Goodland and Anhang, their analysis explicitly omits to account for whole categories, such as GHGs attributable to farmed fish — and it provides minimal estimates for other categories simply to be comprehensive, rather than to be inflationary.

    Another cautionary prescription that’s generally quite useful comes when Ms. Andrews fears that a criticism of FAO use of data from the 1960s is like saying “we think it should be higher.” However, in the case of livestock-GHG analysis, widely-available data on global livestock production shows that it has more than quadrupled since the 1960s.

    Ms. Andrews last comment says that the FAO, by using data from Minnesota, was maximizing its accounting of GHGs — as Minnesota’s use of fertilizer and other chemicals generates GHGs not generated in less industrial production. But it seems Ms. Andrews may not have actually read FAO’s analysis, according to which industrialised livestock are responsible for much less GHGs than non-industrialised livestock — to the extent that the FAO’s analysis even asserts: “The principle means of limiting livestock’s impact on the environment must be… intensification.”

    Finally, here are those five Canadian sources that cite Goodland and Anhang. The first one is authored by four Canadian government climate scientists, at The other four are at,,, and

  3. One just has to read G&A to be skeptical of the 51 percent figure. The original figures from the FAO report are obtained by dividing emissions attributable to livestock by total emissions. G&A start with the FAO figures and only revisit the livestock figures, despite many of the adjustments they make being applicable to non-livestock emissions as well. One example is their recalculation of the impact of methane from the livestock sector using the global warming potential (GWP) for the 20-year timeframe rather than the 100-year frame (which, in CO2 equivalents, is roughly three times higher than the 100 year figure used in the FAO study), but not doing the same for methane from other sources. They also conveniently ignore that carbon sequestration from alternative land use also applies to agricultural land not being used to grow feed. In addition, they update livestock using 2009 figures, but don’t make similar adjustments for all other sources of GHGs (the FAO study was released in 2006). Gavin Schmidt at has indicated that G&A is full of errors (see his comments in the following threads).

  4. Further clarification: The FAO study uses emission estimates for the year 2000, using numbers from World Resource Institute’s Climate Analysis Indicator Tool. The obvious question for the Goodland and Anhang analysis is “51% of what?”, and the answer is something like “51% of 2000 emissions, once a series of adjustments are made, including replacing livestock values with their 2009 levels and weighing only methane from livestock using the 20 versus the 100 year GWP.” Which is essentially meaningless.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s