All guidesClimate science10 min read

The climate footprint of food, measured properly

Food causes roughly a quarter of global greenhouse-gas emissions. Life-cycle data shows where inside that quarter the emissions actually sit — and it is not transport.

The climate footprint of food, measured properly

In short

  • Food systems account for about 26% of global greenhouse-gas emissions; animal products supply 18% of calories but over half of food emissions.
  • Transport accounted for about 6% of food-system emissions in the 2018 analysis; food choice often matters more than distance.
  • A global shift excluding animal products was estimated to cut food emissions by 49% and farmland use by 3.1 billion hectares.
  • Methane from ruminants is short-lived but powerful: cutting it delivers cooling within a human generation.

Key numbers

26%
Share of global GHG emissions from food
Poore & Nemecek, Science 2018
60 kg
CO₂e per kilogram of beef-herd beef, approximate supply-chain estimate
Poore & Nemecek
0.9 kg
CO₂e per kilogram of peas
Poore & Nemecek
76%
Reduction in farmland under a global plant-based diet
Science 2018

A large global food dataset shows wide differences

The 2018 Poore & Nemecek meta-analysis pooled data from about 38,700 farms in 119 countries. Impacts vary widely within foods, but even low-impact animal products typically exceed plant substitutes across several measures. Both dietary choices and production practices can reduce impacts.

Food miles are the wrong lever

For most products transport is 1–9% of the total footprint, dominated instead by land-use change, enteric fermentation, manure and feed production. Air-freighted produce is the exception — but it is a tiny share of trade, and shipping a kilogram of beans across an ocean still beats a locally raised kilogram of beef by an order of magnitude.

  • Beef from beef herds: about 99 kg CO₂e per kilogram as a global mean.
  • Tofu including transport and processing: about 3 kg CO₂e per kilogram.
  • Local grass-fed beef is not low-carbon; grazing extends land use and methane per kilogram.

Methane, and why the clock matters

Ruminant methane has an atmospheric lifetime of roughly twelve years but around 80 times the warming power of CO₂ over twenty years. Because it decays quickly, reducing livestock numbers produces measurable cooling inside decades rather than centuries — one of the few climate levers with fast feedback.

Land is the hidden emission

Half of habitable land is farmed and 77% of that farmland is used for livestock, which returns 18% of calories. Freed land is not idle: restored forest and grassland absorb carbon, which is why diet studies report a 'carbon opportunity cost' often larger than the direct emission saving.

What an individual change is worth

Replacing a high-animal-product diet with a plant-based diet can substantially reduce food-related emissions. The saving depends on the starting diet, replacement foods and production methods; costs also vary.

The data

Global mean kg CO₂e per kilogram of food (Poore & Nemecek, 2018)

  • Beef (beef herd)99.48 kg CO₂e
  • Cheese23.88 kg CO₂e
  • Pork12.31 kg CO₂e
  • Tofu3.16 kg CO₂e
  • Peas0.98 kg CO₂e

At a glance

Global mean impacts per kilogram of food (Poore & Nemecek, 2018)

Foodkg CO₂e / kgLand m² / kgFreshwater withdrawals L / kg
Beef (beef herd)99.48326.211,451
Lamb39.72369.811,803
Cheese23.8887.795,605
Pork12.3117.361,796
Poultry9.8712.22660
Eggs4.676.27578
Tofu3.163.52149
Peas0.987.46397

Common objections

ClaimBuying local is the biggest thing I can do.

What the evidence showsTransport is usually under 10% of food emissions. Replacing beef and dairy with plants beats local sourcing by a wide margin.

ClaimGrass-fed beef is climate-friendly.

What the evidence showsGrass-fed systems generally emit more methane per kilogram and use far more land, offsetting soil-carbon gains that also saturate within decades.

ClaimAlmond milk is worse than dairy.

What the evidence showsAlmond milk generally uses more freshwater than soy or oat milk, but has lower average emissions and land use than dairy. Rankings differ by environmental measure.

Questions people ask

How much CO₂ does going vegan save?
Savings depend on your starting diet, replacement foods and production methods. Avoiding animal products can also free land for ecosystem restoration.
Is cheese really that high-impact?
Yes — the global mean in Poore & Nemecek is about 24 kg CO₂e per kilogram of cheese. Impacts depend on milk production and cheese type.
Which plant milk is best for the climate?
Soy and oat milk generally have low impacts, but rankings depend on the measure. Soy, oat, almond and rice milk have lower average emissions than dairy milk.
Does eating less meat help faster than driving less?
Both can help. Their relative benefit depends on the foods replaced, distance driven, vehicle and timescale; methane reductions can slow warming relatively quickly.
What about food waste?
Waste adds roughly 6% of global emissions on its own. Reducing waste and shifting diet are additive, not alternatives.

Sources

Figures are drawn from the sources listed at the end of this guide.

Text on this page is free to reuse under CC BY 4.0 with attribution to Veg.ac.

Keep reading

Turn evidence into a week of meals

The 7-day starter plan translates everything above into a shopping list and simple recipes.