Our World in Data's chart of greenhouse gas emissions per kilogram of food product, built on Joseph Poore and Thomas Nemecek's 2018 meta-analysis in Science, carries no line for turkey, no line for ham and no line for any branded plant-based roast. It records categories. The Thanksgiving main course falls into three of them: poultry meat, at 9.87 kg of CO₂-equivalents per kilogram; pig meat, at 12.31; and tofu, at 3.16.
All three sit in the lower half of the chart. Above them are beef from beef herds at 99.48, lamb and mutton at 39.72, beef from dairy herds at 33.3 and cheese at 23.88. The distance between those figures and the three candidate centerpieces is far wider than the distance among the centerpieces themselves. That is the framing of VegOut's video “Why We Stopped Telling People to Go Vegan for the Planet,” published December 24, 2025: the dominant divide in this data runs between ruminant meat and everything else.
What the figures cover, and what they stand in for
The mapping matters, because the dataset was never built around a holiday menu. Turkey is covered by Our World in Data's "poultry meat" category. Ham falls under "pig meat." A plant-based roast has no entry of its own and is approximated here by tofu, or by peas and other pulses, depending on what the product is made from. Potatoes, root vegetables and "other vegetables" stand in for the sides. None of those lines is a measurement of a turkey, a ham or a particular packaged roast, and no figure in the chart should be read as one.
The numbers are global mean values in kilograms of CO₂-equivalents per kilogram of product, covering the supply chain from land use change through the farm stage, processing, transport, retail and packaging. Poore and Nemecek drew on 570 suitable studies with a median reference year of 2010, covering about 38,700 commercially viable farms in 119 countries and 40 products that account for roughly 90 percent of global protein and calorie consumption. Our World in Data processed that data into the chart series, last updated in October 2019.
Emissions appear as CO₂-equivalents rather than carbon dioxide alone because agriculture is a large source of methane and nitrous oxide, each weighted by its global warming potential, usually measured over 100 years. For most foods in the set, and especially for the largest emitters, land use and farm-stage emissions account for more than 80 percent of the footprint.
Per kilogram against per 100 grams of protein
Weight is one way to compare foods that are not eaten in equal amounts. Our World in Data publishes a second series, emissions per 100 grams of protein, which reorders parts of the chart. On that basis beef from beef herds is 49.89 kg of CO₂-equivalents per 100 g of protein, pig meat 7.61, poultry meat 5.70, tofu 1.98 and peas 0.44.
The ranking of the five holds across both measures. Tofu stays below both meats, peas stay lowest, and beef from beef herds stays far above everything else. Poore and Nemecek put the general case in their abstract: "impacts of the lowest-impact animal products typically exceed those of vegetable substitutes, providing new evidence for the importance of dietary change."
Readers weighing a meat-free centerpiece on grounds other than emissions can compare the options in VegOut's roundup of vegan Thanksgiving roasts; the chart says nothing about taste, cost or what a given household will actually eat. It also says nothing about the table itself, a subject VegOut has covered separately for readers who are the only vegan at Thanksgiving.
The sides, transport and the range between farms
The accompaniments sit lower per kilogram than any of the mains. Potatoes are 0.46, root vegetables 0.43, other vegetables 0.53, peas 0.98, and wheat and rye 1.57. Lower per kilogram is not the same as negligible, and the dataset gives no total for a whole dinner; portion sizes, cooking method and waste all sit outside what the chart measures. What the figures do show is that the spread between the mains is larger than the spread between the sides.
Transport moves a smaller number still. In Our World in Data's analysis of food choice against eating local, transport accounts for less than 10 percent of emissions for most food products, and for beef from beef herds it is 0.5 percent. The same article cites work by Christopher Weber and Scott Matthews finding that substituting less than one day per week's worth of calories from beef and dairy for chicken, fish, eggs or a plant-based alternative reduces emissions by more than buying all of a household's food from local sources. The same dataset sits behind many of the food-footprint comparisons published around Earth Day.
One finding in the paper limits how far any single figure can be pushed. Impact can vary 50-fold among producers of the same product, Poore and Nemecek wrote, so the global mean for poultry meat, pig meat or tofu covers farms with very different footprints. Their conclusion supports an approach in which producers monitor their own impacts, meet environmental targets by choosing from several practices, and communicate those impacts to consumers.
