95
Land animals spared
Mostly chickens — the species killed in the largest numbers.
Pick your current diet and how long you'd commit to plant-based eating. The estimates below come from peer-reviewed lifecycle and food-system studies.
Mostly chickens — the species killed in the largest numbers.
Fish, shrimp, and bycatch you would otherwise have eaten or financed.
≈ 804 litres of petrol burned.
≈ 5k days of household water use.
≈ 0.48 football pitches.
From land-use change driven by beef and feed soy.
Globally, ~77% of soy and ~36% of cereal grains feed livestock.
Lower blood pressure, lower LDL cholesterol, reduced risk of type 2 diabetes and several cancers.
Different ways to see the same truth: small daily choices add up to enormous lifetime totals.
Swapping one beef-based meal for a bean-based one saves, on average:
Per-kg figures: Poore & Nemecek (2018), Science.
Pick a date and we'll show the running totals you've already saved.
Multiply one omnivore year by your group size to see what a single year of collective change looks like.
One new plant-based habit a day, built on the ingredients you already cook.
Per-omnivore-year baselines combine Poore & Nemecek (Science, 2018), FAO statistics, faunalytics estimates of animals killed per capita, and the Water Footprint Network. Diet multipliers approximate dietary impact relative to an average omnivore in a high-income country.
Numbers are rounded estimates intended for orientation, not precise accounting.
Method
Every figure the calculator returns is a modelled average built from life-cycle assessment data covering roughly 38,000 farms in 119 countries. Knowing what the model includes — and where it is weakest — makes the output far more useful.
The underlying dataset accounts for land-use change, on-farm emissions including enteric methane and manure, animal feed production, processing, transport, packaging and retail. Because it spans thousands of producers, it captures the enormous spread within each product: the highest-impact beef producers emit around twelve times more per kilogram than the lowest. The calculator uses medians rather than means so that a handful of extreme operations does not distort your result.
Ruminant digestion produces methane, a gas with roughly 80 times the warming effect of CO₂ over twenty years, and pasture expansion converts forest and savannah into grazing land, releasing stored carbon. Together these two mechanisms explain why beef and lamb sit an order of magnitude above legumes on almost every metric, and why 'local' rarely rescues them — transport accounts for under 10% of most products' emissions.
Soil carbon sequestration under well-managed grazing is real but modest and saturating; published estimates offset between 20% and 60% of a grass-fed system's emissions for a few decades before levelling off. Aquaculture varies wildly by species and feed regime. And opportunity cost — the carbon that would be stored if farmland reverted to forest — is not included in standard footprints at all, though it is arguably the single largest term in the food-climate ledger.
Treat the output as a comparison rather than an audit. The absolute tonnage depends on assumptions you cannot verify from a kitchen; the ratio between your current diet and an alternative is robust across almost every published dataset. If the calculator says one change halves a category, that finding survives most reasonable changes to the underlying assumptions.
Median values across global production; kg CO₂-equivalent, m² land and litres of freshwater per kg.
| Food | kg CO₂e / kg | Land m² / kg | Water L / kg |
|---|---|---|---|
| Beef (beef herd) | 99 | 326 | 15,400 |
| Lamb & mutton | 40 | 370 | 10,400 |
| Cheese | 24 | 88 | 5,600 |
| Pork | 12 | 17 | 6,000 |
| Poultry | 10 | 12 | 4,300 |
| Eggs | 4.7 | 6 | 3,300 |
| Tofu | 3.2 | 2 | 2,500 |
| Beans & lentils | 0.9 | 3 | 4,000 |
| Oat milk (per L) | 0.9 | 0.8 | 48 |
Poore & Nemecek (2018), Science; Water Footprint Network.
Aspirational input produces flattering output. Use a week that includes the takeaway and the rushed dinner.
Cheese is the most commonly under-reported item in food footprints because it appears as an ingredient rather than a meal. Milk in coffee adds up too — five flat whites is roughly a litre.
Run your baseline, then change a single category. The isolated deltas tell you where effort pays; the combined total tells you very little about priority.
Habits drift in both directions. A quarterly re-run catches regression early and is far more informative than a single snapshot.
For animal welfare and local biodiversity it can be. For climate it is usually worse per kilogram, because grass-fed cattle grow more slowly and emit methane for longer, and because pasture requires far more land. Soil carbon offsets part but not all of that difference.
Yes, consistently. Across emissions, land and water, oat, soy and pea milks come in at roughly a quarter to a third of dairy. Almond milk uses more water than the others but still far less than dairy per litre.
Less than almost anyone expects. Transport is about 6% of global food emissions; for beef it is under 1% of the product's footprint. What you eat matters far more than how far it travelled.
No — food only. Heating, flights and driving are separate and often larger. Food is typically a quarter to a third of a household footprint in high-income countries.
Different datasets, different system boundaries and different treatment of land-use change. Comparisons within one calculator are reliable; absolute figures across calculators are not.