Suppose that you can no longer rely on any other consistent source of food, except what you can grow on your own land. Your stored food supplies have been exhausted. Some disaster has wreaked havoc with the commercial food supply. How much land would you need to grow ALL of your own food? For this particular article, I’m setting aside the question of raising your own chickens, fish, goats, cattle, pigs, etc. For the sake of simplicity, let’s consider how much land it would take to grow a complete vegetarian diet, per person. Now I’m not a vegetarian, but I eat a fairly healthy diet, and if the necessity arises, I’d adapt.
Certain assumptions are necessary to this type of calculation. For example, we have to estimate the total calories per day per person, and the percent of calories attributed to protein, fat, and carbohydrates. [Technically, what we call 'calories' are kilocalories (kcal).] The USDA nutrition labels assume 1800 kcal per day for an adult woman and 2200 kcal per day for an adult man. So the average is that ubiquitous 2000 kcal diet figure found on so many product labels. I suggest that this figure is ridiculously low for anyone who is growing all of his or her own food, largely by manual labor. My target for kcal per day is 2740, which is one million kcal per year per person — a nice round figure, and plenty of calories for a non-sedentary lifestyle. If you think that my numbers are off by, say, 10 or 20%, you can easily increase or decrease the final tally by 10 or 20% to get a result that you prefer.
As for the percentage of calories from protein, fat, and carbohydrates, I’ll use a simplification of the Institute of Medicine’s recommendations: 15% of kcal from protein, 30% from fat, 55% from carbs. [Dietary Reference Intakes: Macronutrients -- PDF] This fat percent might seem a little high at 30%. The average sedentary overweight American would be better off at 15 or 20% of kcal from fat (68% of Americans are overweight). But if you are physically active, and for kids and teens in general, a higher percent of fat is preferable. For someone using manual labor to grow all their own food, they will burn those calories off, no problem.
So we need 1 million kcal per person per year, with 300,000 kcal from fat, 150,000 kcal from protein, and 550,000 kcal from carbs. In addition to considering protein, fat, and carbohydrates in terms of kcal per year, we can also use kilograms per year. The above stated kcal levels for macronutrients work out to some fairly simple numbers in kg of protein, fat, and carbohydrates per person per year, of approximately:
protein: 35 kg per year
fat: 35 kg per year
carbs: 145 kg per year
I’ve already calculated how many kg of protein, fat, and carbohydrates various common and uncommon crops produce per unit area of land. To save you some reading and math, I’ll refer to those numbers without a long explanation. This data is based on the USDA nutrient database, the FAOSTAT database, and various other online sources in agriculture.
All these numbers should work fairly well, to a first approximation. But you are never going to be able to make a precise calculation of calories or of crop yield, in advance. Caloric and nutritional needs vary from person to person and from time to time. Agricultural data will also vary greatly. The same crop planted in the same field in two consecutive years will give different yields. There are no exact numbers in agriculture, so a first approximation is all you can use.
I’m going to assume that the grower is inexperienced, and so I’ll keep the expected yields at a modest level. There are a number of estimates of how much land you need to grow a complete vegetarian diet, from various sources, but they tend to over-idealize the yields. What if there is less rainfall than expected? What if the grower makes a few bad decisions in terms of which varieties to plant, when to plant, how much fertilizer to use? Many factors can reduce yields, and if you are relying on those yields for your survival, you would be better off using very conservative estimates. So that is my approach below. Relatively low expected yields are used, well below what experienced farmers obtain, about one fourth of an optimum yield and about half of a typical commercial yield. Yields were also reduced to account for losses when a crop is dried, cleaned, and milled into grain.
Carbs are relatively easy to grow in a backyard mini-farm: wheat, rice, corn, oats, potato, barley, quinoa, amaranth, etc. The chart that follows has my estimates for the kg of carbs and of protein per 100 square meters for one crop. Numbers have been rounded because, in agriculture, exact numbers are illusory. The values in parentheses are yield per hectare (10,000 sq m); the values below each crop name are per 100 sq m. (1/100th of a hectare). A hectare is 2.471 acres.
Amaranth (1000 kg/ha)
6.50 kg carbs
1.35 kg protein
Barley (1000 kg/ha)
6.25 kg carbs
1.05 kg protein
Corn (1250 kg/ha)
7.70 kg carbs
0.70 kg protein
Oats (1000 kg/ha)
5.00 kg carbs
1.25 kg protein
Potato (7000 kg/ha)
9.15 kg carbs
1.05 kg protein
Quinoa (1000 kg/ha)
6.40 kg carbs
1.40 kg protein
Rice (1000 kg/ha)
6.00 kg carbs
0.50 kg protein
Wheat (at 1000 kg/ha yield)
5.75 kg carbs
1.05 kg protein
All of the above numbers are arguable. With some luck and/or experience, you should do better.
So if you need 145 kg of carbs per year, you will want to plant 2417 sq m of rice, or 2266 sq m of quinoa, or 1585 sq m of potatoes, or 2900 sq m of oats. I would like to simplify those numbers further and say that about 2500 sq m of land will provide enough carbs, from a variety of different staple crops, for one person per year.
If you are growing for 5 persons, you could pick five of these carb crops to grow in one season, and be set for the year. If you have a crop failure or two, you will simply need to try again in the next growing season that same year (assuming you have 6 or more months of frost-free weather per year). So I’m going to consider that all your macronutrients are grown in one season, and that the remaining growing season and any ability to overwinter a crop will be in reserve, in case of initial crop failure.
If all goes well, you can plant a second crop, and have food to give away, to sell, or to barter. This extra food is important. In developing nations, when people live off the land, it is called subsistence farming. But families in that situation are much better off if they are able to grow extra food to sell or trade. It improves their quality of living immensely. So growing only as much food as you need is not a good goal.
Next, let’s consider protein sources. First, you have the protein from the above mentioned carbohydrate crops, which averages about 1 kg of protein per 100 sq m per crop. So with a carb crop of 2500 sq m (50 m by 50 m), you are getting 25 kg of protein, out of the 35 kg that you need. What is the best supplemental source of protein, when your main protein source is grains? — beans, peas, lentils, chickpeas, soybeans, in short: legumes of every kind.
The best legumes for supplemental protein are cowpeas (blackeyes), dried green peas, yardlong beans (the full pods or just the seeds), chickpeas, lentils (sprouted for a better essential amino acid profile) and soybeans. All of the aforementioned legumes are complete proteins, except soybeans, which are a little short on methionine, but otherwise good. Soybeans have the added benefit of being high in dietary fat, with plenty of omega-6 and omega-3 fatty acids.
Yields for legumes are variable. But given a low reasonable estimate of yields, a 100 square meter plot will give you approximately:
chickpeas (1000 kg/ha)
1.90 kg protein per 100 sq m
cowpeas, dried (1000 kg/ha)
2.35 kg protein
lentils (800 kg/ha)
2.05 kg protein
peas, dried (1200 kg/ha)
2.95 kg protein
soybeans (1000 kg/ha)
3.65 kg protein
yardlong beans (8000 kg/ha, whole pods)
2.25 kg protein
The average of the above examples is about 2.5 kg of protein per 100 sq m of legume crop. To gain the additional 10 kg of yearly protein that we need to add to our previous calculation, we will plant only 400 sq m of legumes, in only one crop per year. This assumes that you are going to dry much of the legume crop, to store for later use.
So now we are at 2500 plus 400 = 2900 sq m of land, one crop per year (3 to 4 months), to grow all the protein and carbs that you need per person for a full year. But the next macronutrient is problematic: dietary fat. It is easy enough to grow carbs and protein in a backyard garden or mini-farm. But dietary fat is more difficult. You basically have two options:
1. grow a high-fat food and consume the whole food (e.g. sunflower seeds, peanuts, soybeans)
2. grow a high-fat food and cold press it for the oil
With the first option, your body efficiently extracts the fat from the seed, but you have no cooking oil. With the second option, your yield from a small home oil press will be modest. Piteba claims percent oil extraction efficiency for their small press in the 70 to 85% range; I suggest that this level is rather optimistic. So you are better off with both options.
The only truly small-scale economical home oil press that I’ve found is the Piteba oil press. If you search on YouTube, you can find many different videos of this press in action example. Canola seeds, safflower seeds, sunflower seeds, camelina sativa seeds, pumpkin seeds, peanuts and more can all be manually pressed to produce vegetable oil for cooking. However, several reviews I’ve read say that the yields are limited, and the device takes some practice and skill to use effectively.
Some high-fat foods that you can consume as a whole food (rather than pressing for oil) include:
The above foods can all be pressed for oil, except soybeans, which are too low in oil (18%) for a small home press. In addition, a small oil press can be used to extract oil from:
How much dietary fat can you obtain per 100 sq m of land? I set the yields of home-pressed oilseed below fairly low. This takes into account a possible lower yield of the oilseed crop, and the lower extraction efficiency of small home oil presses.
camelina sativa (1000 kg/ha oilseed)
1.70 kg oil per 100 sq m
canola (1000 kg/ha oilseed)
2.10 kg oil
chia seed (1000 kg/ha oilseed)
1.55 kg oil
chufa (2000 kg/ha tuber)
3.00 kg oil
flax seed (1000 kg/ha oilseed)
2.10 kg oil
peanuts (1000 kg/ha shelled)
2.50 kg oil
pumpkin seed (650 kg/ha hulless variety)
1.60 kg oil
safflower seed (1000 kg/ha oilseed)
1.90 kg oil
sesame seed (500 kg/ha oilseed; even commercial yields of sesame seed are low)
1.25 kg oil
soybeans (1000 kg/ha legume)
2.00 kg oil — but only if you eat the whole seed
(cannot be pressed for oil in a home press)
sunflower seed (850 kg/ha)
2.2 kg oil (black oilseed variety only)
The average of the above yields is 2.0 kg of oil per 100 sq m of land. For the sake of simplicity, let’s consider that all the dietary fat that you need, 35 kg per person per year, is met by the above oil sources, neglecting the oil content of your protein and carb crops (which are generally low in fat anyway). You would then need 1750 sq m of land per person to grow 35 kg of oil in one crop per year. At an average specific density for the oil of 0.92, your 35 kg of oil is about 38 liters in volume.
So we’ve provided for the dietary needs of our macronutrients: protein, fat, and carbohydrates.
But what about vitamins, minerals, antioxidants, and a variety of fruits and vegetables to make our meals healthier and more enjoyable? For that you will simply need a large backyard garden, not a mini-farm. I suggest that a garden of about 50 sq m (just over 500 sq ft.) per person is more space than you need to grow every kind of fruit or vegetable to add to your main protein, fat, and carbohydrate crops.
Now let’s add the above numbers:
main crops: 2500 sq m
(carbs and protein)
legumes: 400 sq m
oilseed: 1750 sq m
fruits and vegetables: 50 sq m
(vitamins, minerals, fiber, and enjoyment)
Total: 4700 sq m of land per person
which is 50590 sq ft., or 1.16 acres per person
Most estimates of the amount of land you need per person to grow a complete vegetarian diet are lower than 1 acre per person, many are lower than 1/2 acre per person. I could certainly lower my above estimate by allowing for 2 crops per year, rather than one. This would make the estimate into 2350 sq m of land, or 0.58 acres per person. That extra 0.08 acres is about 324 sq m. So we could round the estimate down to one half acre per person (a little less than one quarter hectare).
My take on the above analysis is that an ideal amount of land is over one acre per person. I would plan on 1.5 acres of growing space per person, taking into account possible crop failures, land that is used for compost crops, maybe a small area for raising and feeding poultry or cattle, a perhaps a small area for some fruit or nut trees. This estimate works for nearly any location, even one with a long winter, since you are only counting on one crop per year. I conclude that 1.5 acres per person is ideal; it adds some additional margin for error and some land for additional food production to the 1.16 acres determined above.
You could certainly do well with only 1 acre per person. This area of land gives you a wide margin for error, and in all likelihood ample additional food to give away, sell, or barter. With two crops per year, and perhaps some over-wintering crops, you would have a surfeit of food.
It also helps if the total number of persons is larger. The more crops you grow and the more land, the more you can off-set one crop failure with another crop success. For example, with 10 acres for 10 persons, a complete failure on one acre would still leave you with enough food for all 10 persons (0.9 acres per person).
But a half acre per person is, in my view, too little land. In such a case, you are banking your survival on the hope that every crop produces optimum yield, no crop ever fails, the weather is optimal every season, you never have any major pest problems, you never plant the wrong variety of crop, and so on. Theoretically, you could live off of a half-acre per person, but practically speaking it is not enough land.