Why · How foods compare
How do foods compare on land, water and feed?
Mycoprotein, algae, precision fermentation and cultivated meat are absent from every chart and table on this page. They are not absent because they did well or badly. They are absent because this project found no comparable published figure for them: no land, water or feed figure for these four routes that uses the same functional unit as the charts above, which is per 100 g of protein or per 1000 kcal.
Every figure retrieved 19 September 2026
The comparison
The three datasets do not cover the same foods, and the land and water charts each show a selection of their dataset's rows rather than all of them. A food absent from a chart is not zero, and it may still be in the dataset. This is why a food can appear in one chart and not another.
Land per 100 g of protein
m²·year per 100 g protein
The unit here is protein: 100 g of it, whichever food it came from. Tofu and peas are at the bottom and lamb is at the top, and the distance between them is the point of the chart.
Every value in this chart, as text
- Lamb & mutton
- 184.8
- Beef (beef herd)
- 163.6
- Cheese
- 39.8
- Milk
- 27.1
- Beef (dairy herd)
- 21.9
- Pig meat
- 10.7
- Nuts
- 7.94
- Other pulses
- 7.27
- Poultry meat
- 7.06
- Eggs
- 5.65
- Grains
- 4.6
- Peas
- 3.36
- Wheat & rye
- 3.16
- Tofu
- 2.2
Land per 1000 kcal
m²·year per 1000 kcal
The same foods, measured by the energy they deliver instead of the protein. Potatoes and maize appear here because the calorie dataset covers them, and they are not in the protein chart above.
Every value in this chart, as text
- Beef (beef herd)
- 119.5
- Lamb & mutton
- 116.7
- Cheese
- 22.7
- Milk
- 14.9
- Pig meat
- 7.26
- Poultry meat
- 6.61
- Eggs
- 4.35
- Peas
- 2.16
- Nuts
- 2.11
- Wheat & rye
- 1.44
- Tofu
- 1.30
- Potatoes
- 1.20
- Maize
- 0.65
Water per kilogram of product
litres per kg
Every value in this chart, as text
- Beef cattle
- 15,400
- Sheep
- 10,400
- Pig
- 6,000
- Goat
- 5,500
- Chicken
- 4,300
- Eggs
- 3,300
- Cow milk
- 1,000
Feed energy returned, per unit of feed energy eaten
percent · A higher figure means the animal returns more of the energy it was fed. Beef returns under 2 units for every 100 it eats.
- Whole milk
- 24%
- Eggs
- 19%
- Poultry
- 13%
- Pork
- 8.6%
- Lamb/mutton
- 4.4%
- Beef
- 1.9%
Our World in Data · retrieved 19 September 2026
What this comparison cannot show yet
Mycoprotein, algae, precision fermentation and cultivated meat are absent from every chart and table on this page. They are not absent because they did well or badly. They are absent because this project found no comparable published figure for them: no land, water or feed figure for these four routes that uses the same functional unit as the charts above, which is per 100 g of protein or per 1000 kcal.
- Mycoprotein — listed in the directory under mycoprotein
- Algae — covered in the knowledge entries
- Precision fermentation — covered in the knowledge entries
- Cultivated meat — covered in the knowledge entries
A comparison that dropped them silently would flatter them, and one that filled the space with a company's own marketing number would mislead. The directory lists foods in the mycoprotein category and the knowledge entries cover algae, precision fermentation and cultivated meat. None of them can be placed on these charts, so none of them is.
What would fill the gap is a peer-reviewed life-cycle assessment for each route that reports land, water and feed use per 100 g of protein or per 1000 kcal, gathered the same way these three datasets were. Until one exists, the honest form is a named gap rather than a guessed bar.
Three things these charts are not saying
Grass-fed ruminants on pasture can return more human-edible protein than they eat
Cattle and sheep eat grass and leaves that people cannot digest, on land that people often cannot farm. The authors' own summary of Mottet et al. (2017) says that cattle need only 0.6 kg of protein from human-edible feed to produce a kilogram of protein in milk and meat, so a grazing system can add to the food supply rather than only subtract from it. These charts measure land, water and feed energy. They do not measure moral worth, and a long bar is not a verdict on the animal or on the person who farms it.
Global Food Security 14:1–8Livestock: on our plates or eating at our table? A new analysis of the feed/food debate · 2017 · retrieved 21 September 2026
Most livestock feed is material people do not eat
The same work found that 86% of the dry matter fed to livestock is material people do not eat, and that soymeal — the feed most often named in this argument — is 4% of what livestock eat. So a figure for feed energy is not the same thing as a figure for food that could have gone to a person.
Global Food Security 14:1–8Livestock: on our plates or eating at our table? A new analysis of the feed/food debate · 2017 · retrieved 21 September 2026
Land is not one thing that can be swapped for another
The authors' own summary of Mottet et al. (2017) says that most of the land used to feed livestock is grassland, with a large share of pastures that could not be converted to cropland and could only be grazed. That land could not simply be turned into a field of peas. It is the reason a chart of land per 100 g of protein answers a narrower question than it looks like it does.
Global Food Security 14:1–8Livestock: on our plates or eating at our table? A new analysis of the feed/food debate · 2017 · retrieved 21 September 2026
What the water report says itself
“Most of the total volume of water (98%) refers to the water footprint of the feed for the animals. Drinking water for the animals, service water and feed mixing water account only for 1.1%, 0.8% and 0.03%, respectively.”
So a water figure for meat is mostly a water figure for the crops grown to feed the animal.
UNESCO-IHE, Value of Water Research Report Series No. 48 · retrieved 19 September 2026
“The average water footprint per calorie for beef is twenty times larger than for cereals and starchy roots… the water footprint per gram of protein for milk, eggs and chicken meat is about 1.5 times larger than for pulses. For beef, the water footprint per gram of protein is 6 times larger than for pulses.”
The report states these as ratios, so they are quoted rather than charted: they compare foods this page does not otherwise carry.
UNESCO-IHE, Value of Water Research Report Series No. 48 · retrieved 19 September 2026
What this comparison does not show
- These are global averages. One food is not one number: how it was produced, and where, moves the figure a long way. Averages describe a food system, not a farm and not a plate.
- The three datasets do not cover the same foods, and the land and water charts each show a selection of their dataset's rows rather than all of them. A food absent from a chart is not zero, and it may still be in the dataset. This is why a food can appear in one chart and not another.
- A low figure here is a low use of land, water or feed energy. It is not a claim about the food's nutrition, its taste, its price or its worth, and nothing on this page is health advice.
- Every charted figure on this page was retrieved on 19 September 2026 and re-checked against its own dataset on 21 September 2026. The four Mottet figures in the prose were read on 21 September 2026 from the paper's abstract and the authors' own summary, because the paper itself is paywalled; that is recorded against that source. Each row carries the source it came from. Nothing here is rounded, re-derived, estimated or interpolated, and no figure was added from memory.
Sources
-
J. Poore and T. Nemecek. Reducing food's environmental impacts through producers and consumers. Science 360:987–992, 2018, retrieved as Our World in Data's published CSV of that dataset ·
https://ourworldindata.org/grapher/land-use-protein-poore.csv ·
2018, dataset year field 2010 ·
retrieved 19 September 2026
Our World in Data is the publisher of the dataset this page reads. The Science article itself returned HTTP 403 and was not read, so every land figure here is cited to Our World in Data's published CSV and not to the article. The dataset's year field is 2010.
The article's own text was not read; only the published dataset was. -
M. M. Mekonnen and A. Y. Hoekstra. The water footprint of farm animals and animal products. Value of Water Research Report Series No. 48, UNESCO-IHE ·
https://waterfootprint.org/media/downloads/Report-48-WaterFootprint-AnimalProducts-Vol1_1.pdf ·
2010, data period 1996–2005 ·
retrieved 19 September 2026
Global averages over the data period 1996–2005. The two sentences quoted on the page are verbatim from this report. -
P. Alexander et al.. Energy conversion efficiency of meat and dairy production. Alexander et al. (2016), retrieved as Our World in Data's published CSV of that dataset ·
https://ourworldindata.org/grapher/energy-efficiency-of-meat-and-dairy-production.csv ·
2016, dataset year field 2013 ·
retrieved 19 September 2026
The dataset's own header row is recorded here verbatim, so the column this page reads can be checked against it. The dataset's year field is 2013. -
A. Mottet et al.. Livestock: on our plates or eating at our table? A new analysis of the feed/food debate. Global Food Security 14:1–8 (2017) ·
https://doi.org/10.1016/j.gfs.2017.01.001 ·
2017 ·
retrieved 21 September 2026
The paper is paywalled: the publisher's page refused automated retrieval (HTTP 403) and no repository copy could be read. The 0.6 kg, 86% and 4% figures are read on 21 September 2026 from the FAO AGRIS record for the paper, which carries its abstract, and from the authors' own summary in Feedipedia. The 65% figure an earlier version of this page carried is withdrawn: it was attributed to Our World in Data's land-use page, which does not carry it, and the paper's own tables were not read.
Where to go next
These charts say what is known and what is missing. The directory is where the foods are, and the knowledge entries are where the evidence behind them is written out one question at a time.
The food directory: What is made, who makes it and where to look for it. · What we know: One fixed question per entry, answered from dated sources.