How many humans are needed in order to sustain and grow a steady population in a habitable area? Assume resources are adequate and the environment is favorable.

To further refine this question, let's assume this is an early spacefaring technology scenario, such as putting a colony on a neighboring planet (same system) with earth-like conditions with a technology level either current or near-future. Assume the population has a mix of professions suitable for the new environment (resource gatherers, physicians, etc.).

  • 14
    $\begingroup$ I feel like this question should also require that the population be sufficient to ward off genetic defects due to inbreeding. Population bottlenecks like these can be disastrous after many generations even if the numbers are physically sufficient to care for one another and maintain shelter, food, water, etc. $\endgroup$
    – KRyan
    Commented Sep 16, 2014 at 18:37
  • 13
    $\begingroup$ The closer a population gets to the minimum, the more cross-breeding would be required. With a high population, couples can produce several offspring. With a low population, it would be advisable for each woman to have a pregnancy only once with any given man, and for subsequent pregnancies to be with men as distantly related from the previous partners as possible, to boost genetic diversity. $\endgroup$
    – Monty Wild
    Commented Sep 17, 2014 at 2:13
  • 3
    $\begingroup$ In my research for another answer I found that there's been new estimates by Cameron Smith in 2013 that place the minimum number at 14e3 to 44e3 individuals. This estimate accounts for accidents, illness etc. and this is what I used in the answer that was about undersea colonization. It greatly affected the size of the ship that I assumed would be required for the colony to start. $\endgroup$
    – mechalynx
    Commented Oct 8, 2014 at 1:28
  • 2
    $\begingroup$ If you want colony completely independent from the outside world, this question is relevant. $\endgroup$
    – Irigi
    Commented Nov 30, 2014 at 16:13
  • 7
    $\begingroup$ You can use frozen sperm and eggs to lower the number significantly. $\endgroup$ Commented Nov 12, 2016 at 9:44

21 Answers 21


This is quite dependent on technological/medical knowledge.

For example, it has been calculated at 160 for space exploration, provided the explorers return home after 20 generations. This of course assumes pretty good medical availability and actually fairly low risks to individuals. The resultant reduction in genetic variability has been analyzed as not being very detrimental. Certainly it could but not guaranteed to and even small infusions of genetic material would majorly reduce negatives. Also starting out with high genetic variability will help reduce those risks.

A good example of a relatively genetically non-variable population is Ashkenazi jews; a recent study has suggested that in semi-recent history (25-30 generations) their total population was around 350 individuals; yes there are some genetic defects that are common but there is no generalized unhealthiness (or genetic unthriftiness), and some genetic defects are rare because they weren't very present in that initial population group.

I would say that 160 is a fairly good minimum for a colony; yes- less than that may survive but any minor disaster has an unpleasantly high chance of wiping out the colony.

The lower the technology/medical treatment availability the higher the population would need to be to provide a 'buffer' for injuries.

  • 4
    $\begingroup$ One could also imagine that, under the assumption of sufficiently advanced medicine, potential colonists would get a screening for genetic defects, and only those with no significant defects are admitted. Of course, there's always the possibility that additional inheritable defects happen later (especially during a space journey, where there's increased radiation). $\endgroup$
    – celtschk
    Commented Oct 8, 2014 at 8:45
  • 4
    $\begingroup$ Note that, as portrayed in some SciFi (I forget where), with a small population mate selection is dictated by genetic requirements, which is something of a problem in a liberal society. The sexes would have to be separated to maintain The Plan. The minimum without this requirement would be rather higher. $\endgroup$
    – Phil H
    Commented Oct 8, 2014 at 9:18
  • 4
    $\begingroup$ While this seems to be a good estimate to sustain the population, it would be much, much harder to sustain technology. Martian colony would depend on technology heavily and number of people needed to reproduce everything from ore mining to computers would probable go to millions. The same problem, although not so severe, is with qualification, as educating newborn colonists with small population and knowledge database might be problematic. $\endgroup$
    – Irigi
    Commented Oct 10, 2014 at 12:12
  • $\begingroup$ The cited study indicated that 160 was minimum for a 10-generation generation ship that already had a genetically diverse population at the destination. So a viable genetic population is definitely higher than 160. But as others pointed out, it might take more to keep the technology & industry base going than to keep the population diverse. $\endgroup$
    – Jim2B
    Commented Jun 25, 2015 at 4:22

15 individuals, 8 males and 7 females are certainly enough to create a sustainable population, capable of growth without any out-breeding, albeit with some genetic diseases due to in-breeding.

The reason we know this is because of a small island on the Atlantic Ocean, Tristan da Cunha, which has a robust, growing population of 243 (2021), and have been inhabited for about 200 years. All people living on the island can derive themselves from 15 individuals, 8 males and 7 females. The population is a textbook example taught in genetics courses.

Tristan da Cunha - Demographics (Wikipedia)

  • 4
    $\begingroup$ There's also a similar population on Pitcairn Island in the Pacific. (Descended from the Bounty mutineers.) But of course they have some contact with the outside world. It's also worth noting that a previous Polynesian population died out some centuries before the Bounty folks arrived. $\endgroup$
    – jamesqf
    Commented Nov 9, 2016 at 5:06
  • $\begingroup$ I've always assumed that around 50 people would give you all the trades you need to keep a minor colony going with not too advanced technology; something around 1700ish technology, but with minor improvements that don't require advanced tools, like knowing how to make high-grade steel or use fertilizers. $\endgroup$
    – Clearer
    Commented Dec 13, 2017 at 10:10
  • 9
    $\begingroup$ +1 for citing a real-world example. There are also other isolated population groups that have had no genetic input from outside, and have maintained themselves for generations. en.wikipedia.org/wiki/Uncontacted_peoples $\endgroup$ Commented Dec 21, 2017 at 19:24
  • $\begingroup$ @Clearer Two problems: You need more of some skills than others and you need backups. Someday someone will die before the next person to take over the position is trained and now that skill is gone. $\endgroup$ Commented Oct 2, 2022 at 5:14

The answer by @Nick Wilde is very good, but the quoted number of 160 (the source actually claims 80 is possible with social engineering) refers to what is minimally required to sustain a stable population for 10 generations.

This could be relevant if your planet would be very far away, however if you look at the distance between Earth and Mars, it just takes about a year to move from one planet to the next. As such, the population that you would need to fit inside the spaceship could be reduced significantly, as you can simply let the population grow on arrival.

It is hard to quantify this effect, but as 80 already can be sufficient for a stable population, I dare say:

Starting out a growing population can be done with less than 80 people, if chosen carefully.

That being said, the main concern seems to be genetic diversity. Sources vary a bit on the subject, but it appears that with the current technology sperm can be frozen for at least 1 generation. So, let's do a quick calculation:

Suppose we need to reach a population with 80 unique genetic sets to go past the tipping point. Let's do a quick calculation on how this can be achieved if fertile women are able to give birth to 4 children on average and have taken a bunch of frozen sperm samples with them:

  • Generation 0: 10 (fertile) women and 10 frozen samples
  • Generation 1: 20 women and 20 men and 80 frozen samples
  • Generation 2: 40 women and 40 men

From this point on the women of generation 2 could continue to expand the population with the men of generation 1 and 2. It is true that a little bit of bad luck could already mess up the system, but being on the safe side it seems like:

A spaceship with 20 women and a freezer full of sperm is likely enough to start a growing population.

  • 8
    $\begingroup$ Good points. Indeed, it would even be conceivable that after the first colonists, only sperms are sent up in order to provide genetic variety. That would be much more cost-effective since the further ships would not need life-support systems. $\endgroup$
    – celtschk
    Commented Oct 8, 2014 at 8:50
  • 4
    $\begingroup$ Men aren't necessary at all in a scenario like this. All the genetic material could be gathered from two eggs and in vitro fertilization. There would be no y chromosomes and therefore never any boys would be born. $\endgroup$
    – Octopus
    Commented Oct 8, 2014 at 18:20
  • 13
    $\begingroup$ @Octopus - men would not be needed in the first several generations but eventually they would if the colony was to be self-sustaining. Therefore, some of the sperm would have to male. Moreover, you'd want a backup so live men would be the best choice. In theory, since sperm are generated outside the primary immune system, you could transplant the sperm stem cells of many men into one, turning that one guy into say 50 genetic individuals. $\endgroup$
    – TechZen
    Commented Oct 18, 2014 at 18:12
  • $\begingroup$ For a interestellar spaceship I guess the best approach is to send female only embrios and lots of sperm (bonus: human bodys cannot sustain too much acceleraiton). Defroze the first female generation just a few years before arriving. Males are not really necessary in the first generations and a unique male can let lots of females pregnat. Free of actual society vices and relogious/constumary rules and taboos breeding and raising humans cannot be a problem at all. It can be done in a collective way without pressures, like realtionships and breeding become total unrelated $\endgroup$
    – jean
    Commented Oct 22, 2015 at 17:46
  • 6
    $\begingroup$ Haven't you learned anything from Jurassic park? Epigenetic behavioral and cultural info also needs to be preserved. You lose all that with just frozen sperm. I do not believe a functioning society can be restored if the males are removed. $\endgroup$
    – Innovine
    Commented Aug 11, 2019 at 6:37

This is known as the Minimum Viable Population, and many computer models and studies based on various circumstances and species have been run.

For Humans, including the desire to ward of genetic defects due to inbreeding the median MVP reported is 4,169 individuals. You can read up more on this on the wiki article here.

  • 13
    $\begingroup$ According to that article, 4169 is the median MVP across multiple species, and MVP does not account for human intervention. The number is likely far lower for a sentient species, and even lower for one capable of performing genetic analyses. $\endgroup$
    – Attackfarm
    Commented Sep 16, 2014 at 20:12
  • 1
    $\begingroup$ @Attackfarm But then you get into issues of a completely new, and thus highly risky, environment, which should push your MVP back up. You also have the problem that humans low genetic diversity to begin with. $\endgroup$
    – John
    Commented Jan 23, 2017 at 1:13
  • 1
    $\begingroup$ The question specifically assumes a safe environment with adequate resources. And other answers are offering evidence of 15-20. Regardless, 4,169 is a median across multiple species, not humans. Further, again, our ability to perform genetic testing would improve genetic diversity significantly. $\endgroup$
    – Attackfarm
    Commented Jan 25, 2017 at 1:42
  • $\begingroup$ We have solid evidence that, after Toba exploded, the number of breeding humans in the entire world may have dropped to as low as 80, so the MVP of humans is likely at most 80. $\endgroup$
    – Michael W.
    Commented Dec 23, 2017 at 0:00
  • 3
    $\begingroup$ The idea that humans nearly became extinct 75,000 ago because of a super-volcano eruption is not supported by new data from Africa, scientists say. pnas.org/content/110/20/8025.abstract $\endgroup$ Commented Dec 31, 2017 at 11:09

I'm going to ignore genetic diversity because the necessary population to sustain a technological environment will always demand enough people to satisfy that requirement.


  • Maintenance of the basic technology that brought them to the new planet. Not necessarily rocketry... but vehicles, automation, manufacturing, etc. In other words, I'm assuming we didn't spend trillions of dollars to move a population to a new world so they could wear skins and fish with spears sharpened by broken rocks.

  • Self-sustaining colony. All the mining, farming, logging, maintenance, manufacturing, etc. must exist within the colony.

  • The colonization process brought a single set of all the tools needed to get through, say, a year, but after that the colonists must provide for themselves 100%.

  • Automation is commensurate with what's available in 2017. Automation changes these numbers RADICALLY, but the OP doesn't really tell us anything other than it's a intrasystem space-faring society in the near future.

Family Size

Your average family in the United States is 2.53 people. I'm not even going to speculate on what 0.53 people looks like. Worse, even if you round it to 3, that's two parents and one child, which is a decreasing population. You need 4 to break even and 4.1 (better known as 5) to grow. Let's favor 5.

Out of that 5 people let's assume you have two adults, one sub-adult, and two children. That's 2 laborers, one laborer/secondary education, and 2 dependents requiring child care and education.

Your biggest problem is children. They can't be ignored. If you arrive with no children and expect to start making babies you need adults for child care immediately and both education and medical by age 5-6. Therefore, you really must account for children at the outset or your fooling yourself. You could squeeze my final number by minimizing children based on age distribution so that you're getting replacement adults "just in time," but that's much more complicated.


From here we learn that on average one farmer can feed 155 people. This assumes an established farm in 2010. If we assume the farmer has a family and that he can draw on a "basic labor pool" to get him up and running, then our farmer +4 (family) can feed 150 people or 5:155.

farming = 5T/155 where T = Total population (keep this in mind).


I thought I could assume if we can get this group to another planet then we can construct with plastic and metal and therefore there is no need for logging. However, it's likely you need crates, paper, rubber, composting, and who knows what other non-foodstuff organics. I can't find statistics for how many people are involved in this. Let's use the farming number and assume one family can serve an additional 150 people with non-food organics. So, another 5:155.

Logging = 5T/155


This is one of the more problematic issues. Different materials appear in different locations on a planet. This means our colony has many labor centers, all requiring administration, law enforcement, etc. I'm going to define mining as "anything we take from the ground that we can't eat," so it includes petrochemicals.

According to here the average citizen of the U.S. needs 40,000 pounds (20 tons) annually of materials of over 15 types. (I'm being optimistic about the "and other" categories and counting them as one each.)

Mining requires blasting/digging/drilling, hauling, and processing. I have friends who work the hard-rock mines in northern Idaho, and even with automation, they have hundreds of laborers. Granted, it's production mining rather than subsistence colony mining, but still.... Let's assume you need 2 blast/dig/drill, 1 haul, 2 processing for 5 people (25 with families). I must assume a massive ratio or this simply doesn't make sense, so let's assume 25:3000 per item on average.

Remember, that's 2 blasters/diggers/drillers producing their share of 60,000 tons per day. It can be done. I must be done, because...

Total: 825:3000 (see my problem?)

mining = 825T/3000

We need to sub-calculate the number of labor centers for later calculations. That would be T/150 + T/150 + T/2175 = 30T/2175.


We need to get things from one place (e.g., farm or mining location) to someplace else (at least another farm or minining location). If we only assume two drivers + families (trucks or trains... pray we're not dealing with trains...) per labor center.

trans = 10(30T/2175)


Keeping the equipment running is a very difficult variable to define. Let's assume two full-time mechanics per labor center.

maint = 10(30T/2175)


The United States has 5.49 teachers per thousand students. We have some austerity going on, so 5:1000. Secondary eduction is 5.5, but this is more important, so 6:1000. Plus families.

education = 55T/1000.


Keeping all these people alive will be a problem. Keeping them healthy an even bigger one. You've gotta love the CIA, who suggest 2.55 physicians per 1,000 people in the U.S. I doubt this includes specialists and it certainly doesn't account for multiple labor centers. So, 2:1000 MDs and 1 MD per 3 labor centers and 5:1000 specialists. Plus families.

medical = 2T/1000 + 10T/2175 + 5T/1000 ≅ 25T/2175 + families = 125T/2175


This includes everything from paper-pushing bureaucrats to police, firemen, code enforcement, phone answerers, etc. My right-big-toe tells me we need 3 per labor center + families.

admin = 3(30T/2175) = 90T/2175 + families = 450T/2175


According to here, manufacturing jobs are 8.5% of the U.S. workforce. Let's assume that's only one parent in a family (like everything else).

manufacturing = (0.085)T/5 + family = 0.085T

I'm going to assume this number includes drivers, administrators, maintenance, etc. It's going to be a bit low because of the compounding affect of unassigned workers, but all that would happen is fewer unassigned workers increased by the same amount of additional manufacturing jobs. So the total population estimate should remain "accurate." (ahem).


I'm going to ignore entertainment of any and all kinds. Colonists should plan on bringing kazoos.

Unassigned Workers

I've intentionally not tried to compound the spouses or sub-adults into the workforce. This is because there will be jobs ranging from street sweeper to technical assistant that I'm not even going to try to estimate. All those jobs must draw from the unassigned labor force (and sub-adults, as necessary).

What am I missing? I'm missing retail outlets, which would presume some kind of town or village. For an initial colony, central distribution or coordinated transportation could solve this until the community begins to grow. I'm sure I'm overlooking/ignoring a lot of other things, but this is a long enough post.

That's a mess... how do you calculate the minimum population?

  • We're going to iterate through the equation.

  • We'll use T=1 to find our starting point and ignore the divisors so we have one of each primary laborer plus their family, which means our "initial population" is 2,070. We know this isn't right because we don't have enough raw goods to feed/supply all those people.

  • We want to assume that we never increase a number unless the population has actually exceeded the amount required for the increase. No fractional doctors, please, otherwise all the fractions add up to wholes that run out of control. However, this means people are a bit overworked... but that might be expected on a new colony.

  • With the exception of our raw materials. We must have enough farmers, loggers, and miners. So we'll round those numbers rather than keeping them at the floor.

Total minimum population: 2,260

  • Children: 904
  • Sub-adults: 452
  • Unassigned Workers: 452
  • Farmers: 14
  • Loggers: 14
  • Miners: 123
  • Transport: 61
  • Maintenance: 61
  • Education: 24
  • Medical: 25
  • Administration: 92
  • Manufacturing: 38

Is this realistic? To be honest, in real life you probably need 10X this number of people... but I can't prove it without spending 10X the time to analyze the situation.

My Program (PHP)

$a = 100; $t = 1; $ifar = 5; $ilog = 5; $imin = 825; $itra = 300; $imai = 300; $iedu = 55; $imed = 125; $iadm = 450; $iman = 5; $t = $ifar + $ilog + $imin + $itra + $imai + $iedu + $imed + $iadm + $iman; $count_check = 0; $count_max = 1000; while($a > 0.01){ $far = 5 * round($t/155); $log = 5 * round($t/155); $min = 5 * round(165*$t/3000); $tra = 5 * floor(60*$t/2175); $mai = 5 * floor(60*$t/2175); $edu = 5 * floor(11*$t/1000); $med = 5 * floor(25*$t/2175); $adm = 5 * floor(90*$t/2175); $man = 5 * floor(0.085*$t/5); $pop = $far + $log + $min + $tra + $mai + $edu + $med + $adm + $man; $a = abs(($pop - $t)/$t); $t = $pop; if($count_check >= $count_max){echo "\n\nFAILED TO CONVERGE!\n\n"; exit;} $count_check++; } echo "\n\n"; echo "Total Population:\t".$pop."\n"; $children = 2*$pop/5; echo "Children:\t\t".$children."\n"; $subad = $pop/5; echo "Sub-adults:\t\t".$subad."\n"; $spouses = $subad; echo "Unassigned Workers:\t".$spouses."\n"; $far /= 5; echo "Farmers:\t\t".$far."\n"; $log /= 5; echo "Loggers:\t\t".$log."\n"; $min /= 5; echo "Miners:\t\t\t".$min."\n"; $tra /= 5; echo "Transport:\t\t".$tra."\n"; $mai /= 5; echo "Maintenance:\t\t".$mai."\n"; $edu /= 5; echo "Education:\t\t".$edu."\n"; $med /= 5; echo "Medical:\t\t".$med."\n"; $adm /= 5; echo "Administration:\t\t".$adm."\n"; $man /= 5; echo "Manufacturing:\t\t".$man."\n"; echo "\n";

  • $\begingroup$ I think this is way low because it considers all such workers generic when in reality there are specialties. Also, that manufacturing involves a lot of machinery assistance that isn't going to work on the small scale you're dealing with. $\endgroup$ Commented Oct 2, 2022 at 5:25
  • $\begingroup$ @LorenPechtel From my answer, Is this realistic? To be honest, in real life you probably need 10X this number of people... $\endgroup$
    – JBH
    Commented Oct 2, 2022 at 16:00
  • $\begingroup$ Your average family in the United States is 2.53 people. - this number does not indicate population decay . Say, you can have 1 family of 7 people and 3 families of 1 person each, the size of the average family will be (7+1+1+1)/4=2.5. But it's 5 adults and 5 kids, so a stable population. $\endgroup$
    – Kostya_I
    Commented Jan 25 at 18:33

As well as the biological, social and genetic diversity issues of maintaining a viable population you also need to consider maintaining technology. Clearly a small colony on another planet is going to very very reliant on technology to survive.

Just for a start it is not difficult to imagine a situation where a new strain of a disease could cut a catastrophic swath through a small colony and even without anything catastrophic day to day illness and injury will be a big strain without adequate medical facilities.

Equally you will need engineers and technicians to keep the basic infrastructure going.

Automation can probably help with this to some extent but it would certainly seem risky not to have a reasonable depth of expertise in how it all works, especially as the generations go by and you want to expand and refurbish the facilities.

So if we hand-wave the technology of actually getting there and assume a level of technology similar to current with reasonably foreseeable developments on an earth-like planet then we can start to get at least some idea.


In the developed world there are hundreds of medical specialities and we probably have to accept that a small colony won't have the same breadth and depth of expertise and facilities. Here a good model might be a military field hospital as they are designed to be portable and versatile and are well equipped to deal with both traumatic injuries and primary healthcare as well as epidemics and natural disasters which seem reasonable priorities for a colony. So you might have :

  • Surgeons
  • Anesthetists
  • Surgical nurses
  • General practitioners
  • Dentists
  • Pharmacists
  • Ward nurses
  • Intensive care nurses
  • Primary care nurses
  • Paramedics
  • Specialist cleaning and logistics staff

For a long duration colony you would probably also want lab technicians and medical researchers although these may be part of the science department. Obviously you would want personnel with a broad a range of expertise and experience as possible.

Similarly if you intend to use working animals and livestock you will want a few vets to look after them.

Government and administration

Clearly this depends on how you want the colony to be run but with a military type model you would have something like a regimental headquarters with a commander, adjutant, office staff and probably also heads of the various departments as technical advisers. Police and security may also come under this heading at least in part.

Regardless of the actual system of government you will need at least a few people with administrative and legal skills.

For obvious social and cultural reasons it may not be desirable not to have a purely military culture with absolute authority over the colony but it does make sense as an administrative structure even if you have other more democratic or decentralised checks and balances in place.

Engineering and Logistics

As mentioned the colony will depend a lot on technology which needs to be maintained and you will also need the technical expertise to develop natural resources, construct buildings and infrastructure and set up manufacturing facilities you could also argue that farming and food production falls broadly into this category. Again you have a huge array of specialisations to choose from and there will be compromises to be made between specialist expertise and more versatile generalists. Also each discipline will require theoretical knowledge, technical experience and management and planing. To start with you might set up :

  • Maintenance : technicians who know how the various basic systems work and how to maintain them
  • Machine and fabrication shop : supporting the other departments in making and repairing parts and eventually setting up a more comprehensive manufacturing base
  • Infrastructure and mining : building roads and buildings, mapping and surveying, setting up power generation and distribution.
  • Electronics and IT : maintaining and developing computer, electronic and communication systems.


Although you would probably want to mechanise a lot of food production if you end up a planet with indigenous life or at least the immediate ability to support it you will want some specific agricultural expertise. Equally if the planet is sufficiently earth-like some experts in more traditional skills of hunting and living off then land may significantly improve quality of life especially in the early stages.

Science and education

For a multi-generational colony education is clearly vital as you will need to educate the next generation to replace the skills you brought with you. Here you have the potential issue that you have a specific set of requirements in terms of skills but no idea what aptitudes and abilities the next generation will have so your education system will need to be flexible, effective and adaptable.

It will also be crucial that you have an effective way of storing and accessing as much accumulated knowledge as possible. The expertise will be thinly spread so libraries and librarians may be at the centre of the society and the job of librarian may end up as a very specific niche with a very broad spectrum of knowledge.

You would probably also want to have academic/research scientists representing the main branches of science. For reasonable completeness you might have representatives from

  • Mathematics
  • Chemistry
  • Materials science
  • Botany
  • Zoology
  • Medicine
  • Physics
  • Astrophysics
  • Geology
  • Meteorology
  • Psychology

Obviously there would be additional expertise from the other technical divisions and they might be supplemented by lab technicians and assistants and eventually take on students as part of the education system.

Culture and Services

As well as the basic services required to keep a society running any permanent colony will need to develop culture and leisure facilities to keep the population effective and sane. It is reasonable to assume that these will develop organically to some extent but a few really good chefs and perhaps some resident artists, writers and musicians could well be a reasonable addition to a colony.

Just as an example typical commercial kitchen which could feed a few hundred people to a decent standard might include

  • Head Chef
  • Sous chef
  • Pastry chef
  • Assistant chefs (perhaps 3 or 4 to allow for 365 day operation)
  • Kitchen assistants (could also be trainee chefs)


I've suggested 5 main divisions of expertise and on average it looks like you might want perhaps a dozen or so lead experts in specific fields as well as at least the same number of assistants, technicians and support staff. You might also have a flexible pool of labour from the general population. This also allows for some cross training so there isn't too much reliance on any one individual.

So with this in mind we could perhaps guess that you might have

  • 60 or so 'officers' with versatile technical expertise and administrative and organisational roles
  • 150+ skilled, specialist technicians
  • 300+ general and adaptable labour force with varying degrees of specialisation.

Again going back to a military model a battle-group is typically the basic unit which is more or less self sufficient and consists of around 500-800 soldiers with its own organic logistics and support. Obviously this isn't a very strong analogy to a space colony but it does represent a known structure which is able to function independently long term in a hostile environment with a good mix of specialist expertise and equipment, organisational structures, logistics and versatile labour.

  • $\begingroup$ A colony call-sheet like this looks familiar. Wasn’t this used in another answer somewhere? $\endgroup$
    – JDługosz
    Commented Nov 9, 2016 at 8:48
  • 3
    $\begingroup$ This, You can't claim low values without either (a) being willing to settle for a early metal age technological basis or (b) assuming a self-sustaining technological infrastructure. $\endgroup$ Commented Jan 23, 2017 at 4:09
  • $\begingroup$ Re; Farming; I would argue for potentially much higher multiples of labor to output if effort was concentrated on modern green house, aquaculture etc technologies that can produce 500% or greater increases in agricultural production per sq meter depending on the crop. Even grains can produce 6 crops a year in enclosed environments under recent research. You also have the option of micro algae farming and the bulk manufacture of animal protein via mass produced cell culture. $\endgroup$
    – Mon
    Commented Feb 8, 2021 at 23:01

It would take quite a few indeed, potentially more than the 80-160 otherwise suggested. This isn't just about genetic diversity, "spares" and the like, but about all the tasks the colonists would have to do. Even travelling a short distance, you have to have someone up there for every specialist task possible.

Even assuming you don't take enough people to perform what we would now consider basic medical treatment (for example), you still need to have -someone- up there capable of diagnosing basic medical ailments, even if you've got no surgeons, anaesthetists, nurses, gynecologists, maternity staff, etc.

While people can be trained to do a job, they'll be sub-par, and you can expect a few more casualties as a result. Jobs like plumbers, engineers, builders and whatnot would all have to be trained on the new colony, or sent up to form it. Every field would need representation to ensure the success of the colony. I'd venture you'd need a lot more to cover all the things you'd need to get a proper colony running.

As you said, we're assuming all the proper training in the people sent up, but even then, I'd venture you'd need 300-odd to cover all the necessary professions of a colony.

  • $\begingroup$ What about having a full copy of human knowledge as part of the travelers inventory? Along with a carefully crafted study plan for each profession that is not initially present in the original crew. Then when the population reach certain number start to study to revive the missing professions. $\endgroup$ Commented Nov 16, 2014 at 5:08
  • $\begingroup$ As a best practice, when answering questions I believe in dividing your answer into several "paragraphs" especially if a single sentence takes up three lines. Just trying to improve the readability of the question as I have dyslexia and I am sure some others on this SE do too. $\endgroup$
    – Jax
    Commented Feb 14, 2015 at 1:22
  • $\begingroup$ @DustinJackson Yeah, there were definitely spaces for paragraphs in there, sorry. $\endgroup$
    – Rowanas
    Commented Feb 14, 2015 at 8:45

What quality of life and technology level do you want your colonists to survive at?

The lower bound here is not the MVP it's the number of specialists required to maintain the society. If a society of 160 survives, ie the size of a small village, they'll have the technology of a small village, including medicine, education, and life span.

If you want to raise the standard, you'll need more individuals to specialise. It takes many (no idea) individuals doing farming and building before you can, for example, introduce a full time teacher. You will likely need several teachers before you can support some research and development.

Same for industry. You'll need a bunch of miners and lots of farmers before its economically possible to have a blacksmith. And you'll need a good thriving metalwork industry (plus all its consumers) before you can support advanced metalurgy. Technology and development only really takes off when people are able to specialize. In the village, everyones working their ass off in multiple roles just to survive, which makes it tough to advance.

If you want them to use near present day technology, and to keep that tech, you'll need a population of hundreds of thousands, probably many millions. you'll need plastics production, which needs oil wells, which needs engineers and steel construction, you'll need vast amounts of specialist material production (how are they gonna make integrated circuits), tools production, etc etc. eve\ if they start with all of this the stuff will break down and they won't have the tools and reserve parts to repair, until they're all back to Mad Max standards.


The American native peoples are believed to all be the descendants of about 70 individuals who crossed a land bridge from Asia during the last Ice Age (Hey, 2005):

Taken together, the analyses in this study suggest a recent founding of the New World Amerind-speaking peoples by a small population of effective size near 70, followed by population growth in the New World.

It's not my field, but I gather from that article that it doesn't mean there were necessarily 70 in one place at one time; these could represent multiple waves of migration, and of course these 70 "founders" may have come with their children, grandchildren, etc, so the original tribe may have been many more than 70.

These founders were enough to create a fairly diverse number of peoples and civilizations from the Eskimo to the Inca and everywhere in between. However, history has seemed to show that the native Americans were particularly vulnerable to diseases transmitted by European explorers (while the reverse was not true) so you might hypothesize that they had too little genetic diversity and a larger set of founders would have produced hardier descendants.

Bottom line, my answer is: you could do it with two people, but I'd send 100 (not closely related) colonists to give myself a better start.


Theoretically, you could drive the number lower if you had artificial gestation technology and REALLY good record-keeping.

For example, 4 men (labeled A-D) and 4 women (W-Z) yields:

AW-Z, BW-BZ, CW-CZ, DW-DZ of each gender (assuming you use an X and a Y spem for each pairing).


Assuming the genetic diversity is random and you want them to eventually populate the planet, you want somewhere between 3-5000 individuals. that's the general rule for vertebrates. you need a wide genetic diversity to deal with disease, reduce unintentional inbreeding, and you want enough redundancy to survive a natural disaster cutting the population. the societal constraints are far easier than the genetic ones. http://www.sciencedirect.com/science/article/pii/S0006320707002534

Technological limits are really just having really good data storage and being able to produce enough food. As long as people know what they need to do they can rebuild anything. You assume a colony is going to have to rebuild infrastructure and produce more specialists over several generations. Now if they hand pick the people for genetic diversity you can get away with ~500 but they will be a high risk of failure if an a unanticipated natural (or artificial) disaster occurs. That is the risk with small island population one new diseases or drought can doom them.


One problem with 10 colonists is you do not have a very deep bench. If a couple of them croak you may not have enough people left to continue farming. Consider Jamestown, one of the early American colonies.


The following winter, disaster once again struck Jamestown. Only 60 of 500 colonists survived the period, now known as “the starving time.” Historians have never determined exactly why so many perished, although disease, famine (spurred by the worst drought in 800 years, as climate records indicate), and Indian attacks took their toll. On June 7, 1610, Jamestown’s residents abandoned the hapless town, but the next day their ships were met by a convoy led by the new governor of Virginia, Thomas West, Lord De La Ware, who ordered the settlers back to the colony.

That is 88% mortality, which would leave 1.2 of your colonists alive. I bet the 0.2 guy would not be great company either. Unless your tech could keep his head alive in a jar, in which case he could tell jokes.

Where was I? Your colony's mortality rate will depend on the circumstances of your colony. Jamestown suffered from cold weather and famine. The missing colonists at Roanoake were probably killed by natives. Disease can be a factor. 10 people with advanced tech can work a farm and sidestep disease but if you lose some people, you may come to a tipping point where even with tractors you cannot produce your own food anymore.

In the short term the size of your colony will depend on how much food your colonists can produce for themselves (which depends on your environment, climate, crops and tech level) and causes for mortality / mortality rate. Jamestown seems pretty miserable. But if you are a boatload of shipwrecked slaves in the tropical Caicos, with no disease and the natives long gone, you can probably live off the ocean and your gardens very nicely.

The other issue with a colony is reproduction and sustainability over generations. This will be tricky in the short term with 10 adults. Colony implies self-sustaining, which means a birthrate and babies and kids. Someone has got to bear these kids and someone has got to keep these kids from falling down wells. You need kids if your colony is not going to die with the original colonists but child care (and no old folks or older kids to do it) will cut into your workforce as well.

The real long term deal killer for your 10 colonists colony: your gene pool is too small. When these kids grow up and have kids, who with? With 10 starting colonists, everyone in the colony will be related within a few generations. You are going to have serious inbreeding problems. If this is science fiction you could assert that these colonists have been made genetically pure, or perhaps there is a vat of frozen sperm to use which artificially increases the size of the gene pool. But if it is just 10 ordinary people their descendants are going to struggle.


20,000 over 50 years

This is pretty hard to answer in a rational way, since the factors governming a space colony are many. Instead of reasoning out a number, lets see what a historical colony saw in immigration to establish it as a successful colony.

Plymouth colony (the Pilgrims, in Massachusetts) is a pretty close approximation to a self-contained colony for about 100 years. The Pilgrims left Europe to form their own society, and had little interest in intermingling either with the natives of Massachusetts or non-Puritan Europeans. 103 pilgrims landed from the Mayflower in 1620 (including one born on the way!), and 58 survived the first winter. Only four of the original 18 adult women made it to the first Thanksgiving in 1621; not an auspicious start.

37 new settlers arrived in 1621, 96 more in 1623. All in all, about 20,000 colonists went to the Plymouth colony and the total population in 1690 was about 7,000. This could probably be considered a stable colony at that point; it had repaid the debt that it owed for the land settled so it was reasonably economically prosperous. After all, the first child born on the Mayflower would have been 70 by this time; time enough for three generations on the colony. After 1690, the colony was annexed to Massachusetts and ceased being so isolated.



There are examples of parthenogenesis and that's what you'd use. Have a genetically problem free female produce a colony of clones. Afterwards you could allow random genetic diversity to recombine sexually a la Komodo dragons. But you'd need a source of males at that point. Which I'd suggest some genetic manipulation for. You'd need facilities anyway at some point in the project to either produce the first parthenogenetic human or to provide self-fertilization facilities.

Now you're only limited by hostility of environment and availability of resources. So you might need more people to rule out chance deaths. But you can definitely go as low as 1. And you can model statistical death pretty well, so that number can be conjured pretty easily. Ex: death by drowning might be 10 per 325 million per day. So I you want a full 100 years of life...

$1 - (1-10/325million)^{(36500)} = .11\%$ chance of dying over your lifetime to drowning.

That seems reasonable. Ascertain all threats, and drive it below a few percent total and I'd say you're probably very well prepared.

But maybe...

0? If you start with no population... but have a robotic nurture-care facility. Then growth via artificial womb isn't a very large stretch. It was achieved in 2013 and the scientist stated the technology involved is as old as Dolly, meaning we could have been doing it since 1996 if not for ethical concerns. Furthermore in 2017 we repeated the experiment and presumably developed it further. So the hardest part in this case is having a robot raise a healthy child. The best part is accidental deaths are mostly inconsequential. As long as the facility remains running you can always try again and hope chance is on your side this time.


If we are worried about genetic diversity, we can always bring with us some frozen sperm. All males then could be made infertile, so only insemination would work. Also, some females could be made to produce babies constantly. So, genetic diversity is not a problem, colony could be kickstarted (and sustained) with only one female and some medical staff. The problem is how to support technical society with such a small amount of people.


The answer to this is a ~40 based on the Sentinelese tribe(s) that are located on an island numbers vary from about 40 to 500 people as far as researchers can tell. This number is estimated based on what researchers can see from a distance due to researchers not being able to get on the island. The group has apparently been around for about 55,000 years and has had no contact with the modern world save for possibly at a distance, seeing planes and helicopters.


I haven't read all of the answers above so if I repeat someone I apologize.

For anyone that has read hard sci-fi the question of epigenetics comes into play and the available equipment/technology/knowledge base aboard this vessel.

What are the genetic "trees" of the people involved, including any recessive diseases etc. Is there the capability of "altering" genomes to increase heterozygosity. What are the dictating scientific social mores at the time and in the location regarding these things? These all affect the answer of the minimum healthy population.

Are we sending genetic samples from other people with them to semi-artificially increase the heterozygosity of the population? The age of the people being sent up (i.e. years of viable gestations). It is much less resource and supply intensive to have the capability to transport supplies, organic or non-organic than people themselves.

How diverse is the information base of available information and the skill sets of the explores themselves? The more "jack of all trades" included with a few "specialists" would be more optimal than many specialists and a few multi-potentialities.

I think with near future technology the number could be under the previously mentioned 160. There would need to be more women than men, preferably younger and all people involved, even specialists would need to be able to be trained and educated in multiple variety of roles.

If the near future has viable artificial wombs then the need for more females than males would no longer be necessary other than the potential for it possibly being a lower likelihood of aggressive behaviour.

I would say with appropriate technology, supplies and ingenuity that as few as 75 genetically diverse people would be more than enough, very likely even less. IMHO

  • 1
    $\begingroup$ "...then the need for more females than males would no longer be necessary other than the potential for it possibly being a lower likelihood of aggressive behavior". Um, ever seen a room full of women fighting?! Ever seen the aggression displayed between two/three women fighting over the same man? Women may not be as physically violent as men, but there is still tons of agro involved! $\endgroup$ Commented Aug 29, 2016 at 18:23
  • 1
    $\begingroup$ I am talking physical not psychological aggression lol. I have primarily male friends (and am female btw) so I have not personally observed "two, three women fighting over a man" lol. So maybe females with non-aggressive male tendencies would be a more accurate statement ;) $\endgroup$
    – RTachoir
    Commented Aug 29, 2016 at 18:33
  • $\begingroup$ This would be far more readable if you break it up into paragraphs and maybe add a little additional formatting. You can edit your answer to do that. See How do I format my posts using Markdown or HTML? for a quick guide. Well-formatted answers tend to receive more attention because they are far easier to read. $\endgroup$
    – user
    Commented Aug 29, 2016 at 19:23
  • $\begingroup$ @RTachoir, haha. I've lived with a large number of girls during high school. While I never actually saw any fist fights I definitely saw a few screaming matches...and the hostility in the air, geeesh! Thankfully, I too have not seen two, three women fighting over a man. But I hear it happens :) $\endgroup$ Commented Aug 29, 2016 at 22:05
  • $\begingroup$ Thanks Michael and EveryBitHelps. Obviously very new but love having found a place where there are other people that love theoretical debates on interesting topics :) $\endgroup$
    – RTachoir
    Commented Aug 31, 2016 at 23:29

A Use Case - might be able to work out some rules.

I am assuming a non-garden of Eden (not Earth like ) environment so requires significant technology to survive for example Mars or in Space.

100,000 general purpose population (they can do any job with minimal training). 1% population expansion and as it takes at least 20 year to mature 20% or 20,000 children extra.

Assume 1 Specialist(Doctor, Mechanical Engineer etc) for each 200 people, therefore 500 people in each specialty. If there are as many as 50 sub-categories, then at least 10 are available in each sub-category.

For 20 specialties (500 * 20) => 10,000 ;

The total is at least 100,000 + 20,000 + 10,000 => 130,000.

If we assume that 3 or 5 in each specialty is the minimum needed then 39,000 or 65,000 respectively would be a minimum sustainable population.

Conclusion: The minimum is at least 10s of thousands.

Note: The more people the better, certainly 1 Million will significantly reduce the risk of a colony dying compared to 130,000.


With genetic editing and modification to circumvent problems with inbreeding it is now possible to give rise to a world with only 2 people. Adam and Eve is possible.


If you have artificial gestation, then the minimum vital number drops significantly, because you can run all viable crosses each generation. The potential number of offspring (of each gender, assuming one of each) in such a case, can be expressed as (f^m)*2, where f is the number of women and m is the number of men. For example, (3^3)*2 = 18 (9 girls, 9 boys). Granted, a population that low wouldn’t survive, since each generation 2 female genome would have only 4 potential non-cousin matches, and a generation 3 female would have none, as they would all be second cousins). Still, this method has the advantage of reaching viability much more quickly the more “new blood” is added.


If time and scale isnt a major factor I reckon 30-40 people even would be enough with a few caveats. You would need a means of storing sufficient knowledge gained from the homeworld and the means to distribute it, for future generations having failed to learn enough from their ancestors.

Too many might cause problems of its own, with a heavy overhead of set-up and leadership structures decaying from people with different ideals forming.

Youve already mention a certain resilience of skilled personnel and a not-to-harsh an environment so as long as that resilience is maintained over time this should be ok.

With such a small population you would probably need to be carrying incubators and a sufficiently genetically diverse and healthy population of foetuses/sperm/eggs which could be bred over time to match population growth. Old school values like monogamy and only bearing the children of the ones you love might have to go on the backburner to be replaced by a generation raised by an entire community, rather than just parents.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .