If you want to fight global warming with your diet, it is better to change what you eat than where it comes from, according to a recently published article in the peer-reviewed journal Environmental Science and Technology.*
Analysis by Christopher L. Weber and H. Scott Matthews, professors in the Department of Civil and Environmental Engineering at Carnegie Mellon University, found that the vast majority of greenhouse gas (GHG) emissions in the American food system occur during the production of food, not from transportation of the raw materials, inputs, or final product. They estimate that small changes in dietary habits — such as replacing 1/7 of the average person's red meat and/or dairy consumption with chicken, fish, or eggs, or an all-vegetable diet — can lead to reductions equivalent to those that could be obtained through "maximum localization."
People's "foodprints" — the GHG emissions associated with their diet — are receiving attention in climate-change circles because dietary habits are (theoretically) far easier to change than the source of our electricity, how we get around, and so on. In other words, it could be easier for the population to eat less beef than to switch from coal-fired power plants to solar panels or dramatically revamp the built-up landscape so that more people could walk or bike to work.
"All Local" vs. Dietary Shifts
As a thought experiment, the authors examine how an "all local" diet — i.e., a diet that has zero emissions between producer and eater — compares to shifts in diet in terms of GHG emissions. Since that is nearly impossible to achieve, they found that one could achieve equivalent reductions through the following changes:
The authors also use a automobile comparison to illustrate how changes in diet compare to changes in driving. Using a 25-mile per gallon car as their baseline, they provided the following estimates of mileage reduction through diet shifts:
This is not the first time I've seen dietary choices compared to driving choices. The article "Diet, Energy, and Global Warming," by Gidon Eshel and Pamela A. Martin in Earth Interactions (sub. req'd) compared various diets to various types of cars with similar results.
Get on the life cycle
The authors used a technique called "life cycle analysis" to dissect and examine the food system. Life cycle analysis (LCA) tries to quantify all of the environmental impacts (energy consumption, GHG emission, water pollution, etc.) from production to disposal of a product. An LCA of a glass of wine, for example, could consider production of the bottles, labels, and cork (including all necessary transportation); production and transport of fertilizer and other farm chemicals for the grapes; energy for irrigation; impacts of the actual wine-making process; bottling; climate control for the wine's fermentation, aging, and transport; and transportation to the wholesaler, retailer, and consumer. (Basic overviews of LCA are available from the American Center for Life Cycle Assessment and the United Nations Environment Programme.)
To create the model, the authors used several collections of economic, transport, and food data from 1997. The age of the data is one of the shortcomings of the study because much has changed since then. The rapid increase in food imports — between 2001 and 2007 the value of food imports rose by 73% — is one of the biggest changes that would affect their calculations. The authors performed additional analysis to include the import increase and found that the longer supply chains increase the average distance traveled by food by about 25%, while increasing the overall GHG emissions associated with transportation by only 5%. The reason for the smaller GHG increase is the frequent use of oceangoing ships in international trade, which are far more efficient than trucks. The recent rapid rise in oil and fertilizer prices could also affect the food system in significant ways.
Final transportation accounts for only 4% of emissions
One of the most interesting results is the average amount of transportation required to produce food for an average household. Assuming that 11 pounds of food is required for the average 4-person household per day, they estimate that 4,200 miles of transportation is needed throughout the entire production and supply chain (the "average" household diet was obtained by the authors from the USDA Economic Research Service). The average "direct" distance — the final leg from farm or production facility to the consumer, the segment typically called "food miles" — is 1,020 miles. In terms of GHG, production accounts for 83% of the emissions, while transportation accounts for about 11%. The "direct" segment of transportation causes only 4% of the GHG emissions.
Another interesting result is the clear demonstration that different foods can have dramatically different "foodprints." The figure below, which I created using some of the data from Figure 1-c of the paper, shows the source of GHG emissions for three food types: chicken/fish/eggs, fruit/vegetables, and red meat (the figure in the paper shows eight food types). The horizontal axis is the annual GHG emission per household from the food group in tons of CO2 equivalent**, with a longer bar corresponding to higher emissions. The colored sub-bars show the contributions from various parts of the food system: black is emissions from the "direct" leg, blue is carbon dioxide emissions, red is nitrous oxide emissions, and so on. The authors calculate that the average household's diet causes an annual release of 8.1 tons of GHG, so red meat consumption is responsible for almost one-third of the total food-related emissions, whereas chicken/fish/eggs and fruit/vegetables contribute only about 10% each. Note that for each food group, the delivery and transport emissions are relatively small.
Emissions of methane (orange bar) and nitrous oxide (red bar) are the primary reasons that red meat has such a high "foodprint." Methane is a natural byproduct of digestion in cattle and other ruminants (microorganisms in their gut create methane) and also is formed by decomposing manure. Nitrous oxide is emitted when nitrogen fertilizer breaks down in the soil, during various soil management processes, and when manure decomposes. Since the modern U.S. beef system relies so heavily on intensely fertilized corn and soybean fields for animal feed, the nitrous oxide emissions are relatively high. The far higher GHG emissions from red meat production can also seen in a Belgian study that compared the life-cycle GHG emissions of various animals.
The result in the figure above is for an average household, and therefore the emissions calculation is tied to the quantity of the foods in the diet. To separate the analysis from the average mix of product, the authors also calculated the relative GHG emissions on a per calorie and per kilogram basis. On a per-calorie basis, red meat has about three times the GHG emissions of fruit/vegetable or chicken/fish/eggs, and about twice the GHG emissions of dairy products. On a per-kilogram basis, the ratios are even higher, but that normalization is affected by the high concentration of water in dairy and fruits and vegetables.
Eating local and thinking global
Do the above results mean that locavoreanism has no role to play in fighting climate change? I don't think so.
First, reducing our GHG emissions will require many discrete actions, not just one or two big shifts. The stabilization wedge concept of Socolow and Pacala is a good example of this thinking. If we were to create a wedge strategy for the food system, reducing the amount of transport would certainly be one of the wedges. Second, rebuilding local food networks can help to create stronger, more self-sufficient communities. Third, teaching people about where their food comes from can be a way of introducing them to concepts about the entire economy (like life cycle analysis). Fourth, much of the innovation in agriculture (or, perhaps "old-ovation," since much of the new today is drawn from the old ways) is coming from small farms that serve local populations. Think, for example, of Joel Salatin's operation in Virginia that was profiled in Michael Pollan's "The Omnivore's Dilemma." Salatin is a "grass farmer" and uses his crop of grass to feed cattle and chickens, which then fertilize the land, leading to a high production of animal protein with a relatively low use of industrially derived inputs. Fifth, international trade has many other negative costs, like high levels of air pollution around large ports.
Energy conservation has never been at the top of my personal "why eat local" list. In my view, there are other more important reasons to choose locally-produced foods: the local produce has better flavor than imports, I can make a connection with the producer, local purchases can preserve nearby farmland, and it is one of the important parts of creating a more resilient food network. Considering these benefits, I see a theme that connects them, something that could be called "food mindfulness." Practicing food mindfulness, I consider how and where my food was produced, and how those practices affect the big picture (ecosystems) and the smaller picture (my pleasure and health). With the huge number of challenges facing our food system and society, it is possible that food mindfulness is more important than either food miles or food choices.
The full citation of the paper discussed above is "Food-Miles and the Relative Climate Impacts of Food Choices in the United States," Christopher L. Weber and H. Scott Matthews, Environmental Science and Technology, 42 (10), 3508–3513, 2008. DOI: 10.1021/es702969f
* Subscription required, available in libraries of most academic institutions with a science program. If you really want the article, the authors might be able send you a copy, as they often retain limited rights to distribute their own articles.
** The term "CO2 equivalent" is used to express the climate changing power of a mixture of greenhouse gases in terms of the equivalent amount of CO2 that would cause the same effect. Methane (CH4), for example, is 25 times more potent than CO2, so one ton of methane would can be expressed as 25 tons of CO2 equivalent. Another food related greenhouse gas is nitrous oxide (N2O), with a relative potency of about 298 times that of CO2. (The methane and nitrous oxide climate factors are from the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.)