In The Omnivore's Dilemma, Michael Pollan writes that "the typical item of food on an American's plate travels some fifteen hundred miles to get there, and is frequently better traveled and more worldly than its eater." Observations of that sort have convinced many people that it's far better, from an energy and climate standpoint, to stick with locally grown food and try to reduce their "food miles" whenever possible.
The whole "food miles" concept has never lacked for nitpickers. Drake Bennett wrote a contrarian piece for The Boston Globe last year running through arguments for why it's not always a useful metric for measuring the true environmental impact of this or that grocery item. Moving food by train or ship can, in some cases, be more efficient than having a bunch of customers motor out to a local farm in their minivans. Shipping up a head of lettuce from Chile to New England late in the season can sometimes use less energy than growing that lettuce in a Vermont greenhouse. (Yes, better still would be to stick seasonal produce, but these admittedly simplistic examples are just meant to illustrate that the concept isn't all that straightforward.)
That brings us to the latest twist. Two Carnegie Mellon researchers, Christopher Weber and H. Scott Matthews, published a fairly novel study last month that attempts to pin down the total life cycle of greenhouse gases emitted by the food consumed by the average U.S. household. Not surprisingly, they found that what you eat matters much, much more than where it's grown:
Indeed, they traced 83 percent of the average household's food-related footprint of greenhouse gases to the origins of the food itself. Transportation only contributes 11 percent of greenhouse gas emissions on average—with the transportation leg from producer to retailer accounting for just 4 percent.
Now, this is not to disparage local agriculture in the slightest—it's still very much worthwhile for a whole garden of excellent reasons, one being to break away from the sort of industrial-farming practices that cause so much damage. The how of agriculture still matters. That said, it's striking to learn that, on average, replacing just 21 percent of the red meat in the "typical" diet with fish or chicken does as much, emissions-wise, as buying everything in that same diet locally. Another big surprise is that the far-flung globalization of the food chain that Pollan describes seems to have had a relatively small impact on global greenhouse-gas emissions, in part because ocean freight still accounts for 99 percent of all international transport, and that uses far less energy than trucking. That's unexpected, and it sounds like even the authors were surprised by the result.