Oops! I destroyed the world...again.

November 13, 2006 3:48 PM

So...let's say that some freak nuclear accident (or, in our current political climate, perhaps a purposeful nuclear attack) wipes out most of the world's population. How many people would have to survive to successfully re-populate the world? (more inside)

Taking into account the fact that, at some point, inter-breeding would probably become a problem...how would this be avoided? How can the world be re-populated with a limited stock? Is this eventually not a problem? How does genetics play into this?

posted by AlliKat75 to Science & Nature (15 answers total)

I've heard somewhere that you need a minimum of 250 adults to be a self-sustaining reproductive society.
posted by flibbertigibbet at 3:53 PM on November 13, 2006

A biology major friend of mine once claimed that it would only take 63 people—he didn't specify sex distribution—to eventually repopulate the earth with the same amount of genetic diversity present in humans today.

In theory, we are all descended from something like 24 women, based on partial analysis of our mitochondrial DNA. For example, nearly all Europeans have 1 of 7 strains of mtDNA.
posted by infinitewindow at 4:11 PM on November 13, 2006

Depends on how fast you want to repopulate it. From what sources do you qualify evidence?
posted by vanoakenfold at 4:30 PM on November 13, 2006

A google search on "minimum population genetic diversity" will net you some interesting hits. this paper (which requires some kind of subscription to read) delves into it, and based on the first page (which is shown) shows several factors other than genetic diversity. For instance, you need enough people (or a broadly scattered population) to survive an epidemic or natural disaster.

There's (inevitably) a wikipedia article on Minimum Viable Population. And the above-linked paper is cited in the Wikipedia article. I ran across an abstract that is kind enough to mention that the MVP for mammals is estimated to be in the tens of thousands, and may be as high as a million.

I don't think this is chatfilter, even if the question is being asked in a chatty way—it's a question real scientists really investigate.
posted by adamrice at 4:31 PM on November 13, 2006

Adamrice, that abstract says more than that - it notes the variation in MVP for primates, and suggests that for large primates (ie. orangutangs) you're looking at dozens-of-thousands. Note that this paper relies purely on "what's the smallest population found on an island where the species isn't exitinct". It doesn't go into the genetic and ecological reasons behind the minimum viable population size, nor does it appear to test the idea that these populations (ie. of orangutangs) are actually viable, and not in decline.

A number of factors come into play - lack of genetic diversity, inbreeding, genetic drift, catastrophic events. And estimates for mammals in the ecological literature probably don't directly address your question - they're usually focused on the number of individuals in a population in a given area, not the whole species spread across the planet. 12,000 people spread evenly around the world would have some difficulty forming viable populations in specific centres of concentration. For instance, if the world's populations was reduced from 6 billion to 12,000 evenly, you'd have about 40 people alive in New York.

Anyway, the back-of-the-envelope number that was always given to us was 500 females (males don't count), and that's the number that's often used in extinction-risk models for conservation decisions, but as others have said, this is actually a very active and interesting area of study. Fire up your favourite journal search database and look for "population modelling", "population viability", "minimum viable population size", "genetic drift".
posted by Jimbob at 4:54 PM on November 13, 2006

all you need are 2 guys with no genetic defects at all (it would be like reseting the genome) and oh a few women, assuming no spontaneous abortions, etc and healthy children--cause one buck and fertilize a many a doe; children will have to interbred but like i said--no genetic defects will prevent issues like surfacing recesive mutations, etc.
posted by uncballzer at 5:14 PM on November 13, 2006

Two, jeez, doesn't anybody read the bible around here?

why isn't it two? I thought there were claims that everyone on earth today descended from the same people only a few thousand years back... it seems like just a matter of odds that we would need more, but it would be entirely possible to repopulate with only two so long as there weren't too many nasty recessive genes in their DNA... or are we asking in order to avoid any incest for cultural reasons how many separate couples would be mathematically necessary to avoid everyone ending up cousins within a few generations... well, the 12 tribes of israel descended from 12 brothers, right? who each took a wife from ...somewhere, so that's 24, though you don't end up with as genetically diverse a population.
posted by mdn at 5:26 PM on November 13, 2006

Response by poster: I'm not particularly concerned with the cultural ramifications of incest; I'm more interested in discovering the minimum human requirements for re-establishing the human race.

You have all offered some very interesting points of thought. Thank you - keep 'em coming.

FYI: By "successfully" I meant the re-creation of a sustainable human race. My assumption (though, I could be wrong) is that if there were too many genetic defects created via inter-breeding, humanity would not be able to carry on.
posted by AlliKat75 at 5:37 PM on November 13, 2006

I have a strong but unspecific memory that scientists twenty or thirty years ago thought the minimum viable population for humans was about 3,000. (this question was something they thought about a lot in the 70s and 80s).

Sadly, I learned this long before the Internet, and I'm not sure where to point you to support it.
posted by Malor at 6:05 PM on November 13, 2006

For comparison, you could consider the idea that humans suffered a "population bottleneck" around the time of the Toba supervolcano eruption around 70,000 years ago. Global cooling would have reduced human populations down to a few thousand, according to the theory.

Wikipedia: Toba Catastrophe Theory is as good a starting place as any. The linked Population Bottleneck article mentions some alternate and opposing ideas. The subject pops up on Discovery Channel specials and such, too.
posted by gimonca at 6:09 PM on November 13, 2006

Assuming the remaining humans can immigrate efficiently enough to function as a single genetic population and represent a good sample of the complete genetic diversity of the species. A genetically viable population would consist of a census population size of 5000-50,000. Some gentic diversity would be lost in this scenario and inbreeding would be a minor problem -- assuming random mating and easy immigration. If we want to maintain current genetic diversity within the species at any given neutral loci, it would take an effective population size of 100,000-1,000,000 individuals (which is a census population of approximately 1,000,000-10,000,000 individuals). Genetic diversity would be maintained with this number and inbreeding would not be any more a problem than it is now -- assuming random mating and easy immigration within the remaining human population.

To assure a large enough population to hedge against potential demographic and environmental variation would require populations of similar magnitude -- although it would be better under these conditions if the individuals were divided into separate populations (causing potential genetic problems--or requiring bigger populations to maintain genetic diversity).
posted by Tolerant at 8:15 PM on November 13, 2006

2 only works in optimal/divine conditions. The larger numbers suggested (3,000 - 5,000) are likely more accurate not simply to introduce genetic diversity, but to buffer against any further catastrophes.

So 2 people would work, but you'd have to pray that there weren't any floods, or that the roof doesn't cave in, or that Primary Hunter Adam doesn't break a leg or something. A larger population - say 4,000 - could sustain a major outbreak and or natural disaster, like the ones mentioned in the bible.
posted by littlelebowskiurbanachiever at 9:34 PM on November 13, 2006

You say "genetic problems." I say evolution.
posted by Pollomacho at 9:34 PM on November 13, 2006

Small Population Isolates: a micro-simulation study, N McArthur, IW Saunders, and R Tweedie (Journal of the Polynesian Society) 1976:85 307-326 is a purely stochastic study of reproduction rates using historical Polynesian life and fertility tables, and found that a minimum of 5 couples was the number necessary to ensure a surviving population over the long time. This study did not take genetics or inbreeding into account, and permitted some sense of incest in some of the simulations so it probably shouldn't be taken as a direct correlate to actual population growth.

More recently, Testing Migration Patterns and estimating founding population size in Polynesia using human mtDNA sequences, RP Murray-McIntosh, BJ Scrimshaw, PJ Hatfield and D Penny (Proceedings Nat'l Academy of Sciences, USA) 1998 9047-9052 discusses mtDNA research in New Zealand that suggests that the founding population of females for New Zealand was between 50 and 100 (they say about 70). Of course, there were likely other migrations to New Zealand, but this research shows how a founding population could be genetically and reproductively sound and consist of from 100 to 200 individuals.

Finally, let's not forget the case of Pitcairn Island. Pitcairn Island: fertility and population growth, 1790-1856 WF Refshauge and RJ Walsh (Annals of Human Biology) 1981:8 303-312 shows how a founding population of 28 people (15 men and 12 women of reproductive age) created a viable population, and reached population growth rates of 3%.

I don't have anything to say about your latter questions -- I'm no geneticist! -- but I hope that throws a little historical and theoretical light on your question about founding populations.

Note: some of the links above may not work unless you belong to an institution that subscribes to the electronic journals. Sorry...
posted by barnacles at 9:35 PM on November 13, 2006 [2 favorites]

Humans are much better adapted to survival then most mammals. If a small surviving human population can manage to avoid diseases caused by a large number of dead people lying around, they should be set to rebound quite nicely. Population expansion is usually limited by food supply, and for at least 10 to 20 years (and probably more like 50) there will be an enormous supply of canned and other food lying around. Under these circumstances, I' d be more inclined to go with a number less than 100.

Btw, those links provided by Barnacles are fascinating, I'm lucky to work in a library.
posted by jefeweiss at 10:22 AM on November 14, 2006