Editorial: Study assaults produce eco footprint, puts data in a twist
By Fresh Fruit Portal editor Matthew Ogg
It was a headline that first defied belief, then science. Was this a joke? "Vegetarian and "Healthy" Diets Could Be More Harmful to the Environment" read the press release from Carnegie Mellon University.
"Errr…guys, I think you might have spelled 'less' incorrectly," I thought.
As a representative of the fruit and vegetable industry and an aspiring vegetarian (I mostly have a plant-based diet but do indulge in seafood a couple of times a week), this assertion flew in the face of everything I had read and I knew it had the potential to spread like wildfire. After having some of their products linked to colorectal cancer by the WHO recently, the meat industry really needed a win and my inkling was this might be it.
After the breakthrough agreement from the Paris climate change talks to keep global warming below 2°C, here we had a university press department effectively discouraging people from eating more produce due to environmental concerns? Surely that can't be right.
So I read through the study. Many times, trying to understand how such a conclusion could be reached that was contrary to the existing studies out there, and which as a meta study was itself based on existing studies.
What became clear very quickly was the study had nothing to do with vegetarianism and the attention given to produce was just for the headline. The researchers looked at the USDA's dietary recommendations that the public eat more fruits, vegetables, dairy and seafood, coupled with less consumption of meat products, sugars, oils and fats.
Sounds like a reasonable exercise – let's think about both the health and environmental outcomes of our food chain. The authors also included a fairly speculative measure that included the impacts of dietary change on food waste, and that may have contributed to their conclusion that the USDA's recommendations would result in greater greenhouse gas (GHG) emissions, energy usage and blue water footprints; the latter meaning water that is drawn from the surface as well as from the ground.
Regardless of how we estimate food waste between different diets, it is arguably the biggest challenge facing the sector in reducing its environmental footprint. If there’s one practical take-home from the investigation, that's probably it.
But where I have a lot of beef with this paper is how its message was communicated. Here are a two points that emphasize just how misguided this PR attack on the produce sector was:
1) The increased greenhouse gas emissions come squarely from the seafood and dairy components of the study. The fruit and veggie impact is very low, and meat generates the highest emissions.
2) The water footprint attributed to fruits and vegetables per calorie is extremely high in the study compared to meat, especially considering the point of reference – published by Mekonnen and Hoekstra in 2012 –states, "As a general picture we find that animal products have a larger water footprint per ton of product than crop products."
The reference study also backs this up in liters used per kilocalorie which is by far the highest for beef at a level of 10.19, compared to vegetables with 1.34 and fruits at 2.09. Other products are also bigger water users than fruits and vegetables, including sheep/goat meat (4.25), pig meat (2.15), chicken meat (3) and eggs (2.29).
Using the raw data for the blue water footprint used in Mekonnen and Hoekstra, I found the amount of liters used in ham was still greater than higher water-using produce crops like lettuce and table grapes, albeit by a smaller differential.
The blue water footprint claim was right though in some crops. Avocados and tomatoes had higher levels than beef but were below ham, while peppers and watermelons came in just below beef but were higher than goat's meat.
Searching for answers
So I reached out to one of the report authors, Michielle S. Tom, asking how she calculated the water impact and why the press release had focused so much on fruit and vegetable-rich diets when the study was actually about broader USDA recommendations. She pointed me in the direction of Mekonnen and Hoekstra, and gave a surprising response regarding the vegetarian aspect.
"As for leading with the 'vegetarian' angle, it came out of the PR department. We probably should have caught it, and in subsequent interviews we've tried to clarify our findings," she said in an email.
So we know the press angle was a mistake, but for some reason at the time of writing it is still present on the Carnegie Mellon University website, which press outlets around the world have continued to publish.
But what about the water footprint issue? The study's findings still seemed strange given so much irrigation goes towards crops that feed animals.
"Most of the seed we use for animals I believe comes from corn and other types of grains, and those actually have a relatively low blue water footprint per calorie compared to the fruits and vegetables that we consume directly, so that is part of what leads to the lower bluewater footprint for animals," Tom said over the phone.
"Feed production for animals is included in our study, yes. I think that the main thing here is that we’re looking at this on a bluewater footprint per calorie as opposed to per ton or per pound, and as meat is a lot more caloric-dense than fruits and vegetables, the per calorie basis is going to be higher."
Of course, the removal of meat from any diet is unlikely to be matched by a similar caloric increase in fruits and vegetables, but rather food products with more protein like soy-based foods, lentils, beans and chickpeas. Compared to lentils, the blue water footprint for beef is 142% higher, while the level is also greater in ham (212%) and goat's meat (67%).
Tom also claimed the U.S. had higher blue water footprints than the rest of the world in many crops, due to their production in drought-stricken California.
"As a result it requires a lot more irrigation, whereas a lot of the meat - especially cattle - is raised on farms," she said.
"A lot of their [livestock's] water is coming from I think the green water footprint, so that's rainwater. We’re only looking at the blue water footprint which is drawn from the surface as well as from the ground. So that's going to be very different than if we’re going to be looking at the total water footprint.
"That's something we have not done, but would actually make for a good study in the future."
The real issue at hand
In a later email, Tom pointed to a very interesting observation which perhaps should have been given more weight in the paper.
"A key point that I want to emphasize, though, is that while most of the attention has been on meat, in fact it's the replacement of processed sugars, fats, and oils that is driving our results," she said.
"Unfortunately, the vegetables, fruits, and dairy foods have higher emissions per Calorie than sugars, fats, and oils.
"Thus switching out these lower impact foods with higher impact foods results in an increase in overall emissions when shifting to the "healthier" diet."
It's an issue that was buried towards the end of the results and one that was not mentioned in the paper's discussion, but it is certainly worth reflection. Are current consumption trends of these unhealthy products helping us maintain some kind of balance in the environment?
My gut feeling would be no on sugar, because it induces appetite and therefore contributes to increased calorie intake, while on the other hand many fruits with soluble fiber - including apples, oranges, grapefruit and plums - actually suppress your appetite. The study used an estimate that in 2010 the U.S. population ate an average of 2,390 calories per day, which is still above what is considered a healthy level as an average across adult men and women.
As for fats and oils? If the finding is true, this genuinely does represent what the authors described as a "tension between public health and environmental sustainability". But there are sure to be nuances of this topic that warrant further research, just as Tom emphasized the nuances of the water footprint findings.
"Not all fruits and vegetables are the same. Nor are all meats the same in terms of their environmental impact," she said via email.
"As for the differences in water footprints, again we use conversion factors (Calories per kg) extrapolated from the USDA food availability site that differ from those in the Mekonnen and Hoekstra study," she said. Prior to this comment she had emphasized the Mekonnen and Hoekstra study was the water footprint element's basis.
At the time of writing, Tom had not responded to requests to demonstrate her calculations. The USDA data looks at daily kilocalories per capita rather than per weight in some generic categories such as fruits, citrus, other vegetables and noncitrus, and is not as extensive as the information provided by Mekonnen and Hoekstra.
Due to these limitations I have been unable to cross-check any claims based on the USDA data, but I can say that the agency would have had to place a very different calorie level per weight unit on crops for them to be so different to the levels used globally. I doubt a tomato produced in the U.S. would have a drastically different amount of calories to one grown in the Netherlands or India, for example.
Beyond GHG, energy and water footprints
We also must consider there are myriad metrics that can be used to assess the environmental impact of different food industries, including how they affect biodiversity, soils, watersheds and how they are transported.
One issue not covered in the Tom et al study is excrement. While in the traditional sense manure was useful for farming, in high density livestock operations it is a different story. In his book "Eating Animals", Jonathan Safran Foer points to how the average pig farm produces 7.2 million pounds of manure, a broiler facility 6.6 million pounds and a typical cattle feedlot 344 million pounds.
Safran Foer highlights the enormous waste animals produce compared to the human population, the high polluting strength of that waste and the inability of government agencies to effectively regulate factory farming operations. A lot of the excrement is put into massive, toxic lagoons, and also has been known to contaminate the air of nearby communities.
He also mentioned Environmental Protection Agency (EPA) data that showed chicken, hog and cattle excrement had polluted 35,000 miles of rivers in 22 states, while in the space of three years the waste led to two hundred fish kills; instances where entire populations of fish have been wiped out in a given area.
The produce industry is not without its environmental impacts either, with the most common topic in public discussions being the matter of pesticides and monoculture. There is always room for improvement in this area, but as government agencies tighten up their regulations growers have been forced to turn to more biological pest and disease control options.
Regardless of how we want to interpret the Carnegie Mellon University meta study, at least it has served the purpose of putting the environmental footprint on the map. The produce industry cannot rest on its laurels with this issue. The Mekonnen and Hoekstra report does show the U.S. has a substantially higher blue water footprint in some crops compared to the rest of the world, which prompts action from industry players.
The sector also cannot sit back idly while press releases proliferate saying healthy fruit and veg-based diets are bad for the environment, discussing produce as if it were the same thing as milk and seafood. While Tom might say she's sought to clarify the angle of this release....let me just check....yep, it's still there on the university site. Let's hope it's removed or adjusted very soon.