Twenty years ago, in year 2029, World War III happened. Everyone nuked everybody else. More specifically, India and Pakistan nuked each other, the US, England, and France fired everything they had at Iran, Russia, and China, and those countries responded in kind. Iran, who had developed nukes in secret, also launched a nuke at Israel, who responded by nuking all of their neighbors, who had declared war on them in the confusion. North Korea fired a single nuclear missile, but it crashed into the ocean and sank without detonating.

In addition, most countries that didn't have nukes, but were aligned with the nuclear nations contributed their conventional armies to the war as well. Many of these countries were then nuked. Major cities, of course, were the first targets, but fallout was blown by the wind and poisoned much of the surrounding areas. The fallout from thousands of nuclear explosions also entered the atmosphere and ocean, spreading in predictable patters with the currents and tides.

Nobody wins this sort of war, of course, but surely someone somewhere has survived. There are over seven billion people on Earth, and, assuming they aren't all poisoned by fallout, some of them must be in places that are at least somewhat sheltered from a global nuclear war. After counting deaths from radioactive fallout and climate imbalances in the years after the war, how many people have survived, and where are the survivors?

  • 10
    Side comment... major cities would not be the obvious first targets. The first targets would be other countries' military facilities and missile silos, in order to destroy as much of second strike capability as possible. – user11599 Oct 26 '15 at 17:37
  • 8
    The great DPRK's honor has been besmirched by this capitalist running dog! Exterminate! Exterminate! Exterminate!!! – Serban Tanasa Oct 26 '15 at 18:24
  • 2
    If you haven't already, watch the film The Day After. It's a brutally honest look at what the aftermath of an all-out nuclear war would be like for the survivors. Considering that one of the problems you'd end up with is contaminated topsoil, which puts a severe crimp in your ability to grow food to perpetuate life from year to year, it's kind of doubtful that you'd end up with many survivors at all 20 years later. – Mason Wheeler Oct 26 '15 at 20:41
  • 1
    It depends also on the level of nuclear technology involved. A well-designed and well-constructed warhead shouldn't produce a significant amount of fallout. Large amounts of fallout signify crude devices that waste a large amount of their fissionable material. Fissionable material is expensive, so nuclear powers tend to prefer "clean" nukes whenever possible. – Perkins Oct 26 '15 at 21:53
  • 2
    Also threads. Old british mini series that was dreary as hell but told its story powerfully – Journeyman Geek Oct 27 '15 at 2:48
up vote 26 down vote accepted

Projecting worldwide from this document, we can estimate that under a billion people would die as a direct result of the weapons or the fallout. The current arsenals are just too small, and even with full-scale buildups for the next 15 years, it's hard to imagine even a return to peak cold war levels.

It's even unclear whether a full-blown nuclear winter would indeed ensue. There would certainly be massive amounts of dust and soot pushed into the stratosphere, both from the initial blasts and from the massive raging fires that would follow them. Those would lead to short term cooling, but it's unclear that that would be enough to trigger a full nuclear winter as the anti-nuclear scientists (with an understandable and commendable interest in furthering their pro-peace agenda) would have had us believe.

However, the disruptions to the world economy, health facilities and food supply chains would prove far deadlier. With most of the population in the developed world dependent on continuing food supplies, such large scale devastation could bring down supply chains and cause starvation in many areas.

Nonetheless, people are resourceful, especially when hungry, and the survivors would be quick to adapt and, in many cases, rally around the flag rather than decay into mini warlord estates.

All in all, I can't imagine more than a third of the world population being wiped out by a total, all-out nuclear exchange. Unless we bring in anti-matter devices, that is...

  • 3
    All bets are off when anti-matter weapons come into the picture. – Green Oct 26 '15 at 18:37
  • 5
    Indeed, it’s likely that all-out nuclear war wouldn’t kill enough people. With billions of potential survivors and the total collapse of infrastructure, the fatalities in the following chaos would be of an even greater magnitude than the bombs. – Avernium Oct 26 '15 at 18:39
  • 3
    I was wondering when we had lost the capability to thermonuclearly end all life on planet Earth, so I ran the numbers: That paper estimates today's world wide stock pile at 32k warheads. Earth's population is currently 7b. So each warhead would have to kill roughly 218k people. Considering rural populations, it seems like it'd be quite a stretch to get everyone. – Mazura Oct 26 '15 at 19:39
  • 5
    @Mazura - Putting it another way, the average thermonuke will destroy and/or uninhabitably irradiate a radius of about 4.5 miles from the detonation point (the actual radius will vary widely based on yield of the weapon and geography of the detonation site but this is a good overall average). With your estimated stockpile of 32k warheads, evenly distributed, the world's nuclear stockpile could make up to 2.2 million square miles uninhabitable. Earth has a land area of about 57.3 million square miles, so if the estimate's correct our best efforts would only destroy about 3.8% of Earth's land. – KeithS Oct 26 '15 at 20:54
  • 5
    @KeithS, that would not be much comfort to me, living as I do in Washington, DC. It's not just me, most people live inside a relatively small fraction of the planet's surface, but I agree that the general point holds. – Serban Tanasa Oct 26 '15 at 20:57

I would first point out that a completely accurate number is probably impossible. There are too many different kinds of nukes, too many possible locations, and too many possible variables (wind patterns, weather changes, etc) to really give you a great number.

Check out this fun little calculator for a good indication of casualties and deaths. It gives you a listing of all possible explosive types, all possible locations, and even calculates fallout in a radius from the site.

That said, there are a few places to note that will "safe" in regards to direct blasts, and will have minimal fallout damage to worry about:

Arctic and Antarctic Outposts

This site says that between 1-4000 people live on Antarctica during the year varying from season to season. There is no reason to suspect that any country would target scientific research bases to nuke, so these individuals would survive. Same goes for the Arctic. However, these individuals may end up being stranded and cut off from being able to leave the arctic and may die anyway.

Bunkers turned Residences

With the current doomsday mindset in 2015, it doesn't take any great stretch of the imagination to assume that many more wealthy individuals will take advantage of purchasing sheltered real estate. I can't guesstimate the exact number, but it could be upwards of 500+ people, as we have no exact numbers on how many bunkers exist and how many are being turned into residences.

Underground Facilities

Search results for this are overwhelmingly conspiracy theorists claiming there are thousands upon thousands of underground facilities so that they can "cull" the outside population at will. However, there are legitimate military facilities, as well as scientific ones, such as those studying the earth's crust, volcanologists, and drill sites, which would be secure from initial explosion, though the occupants within would probably suffer fallout damage once they left the facility.

Rural Areas

Nuclear attacks will target military installations first, followed by centers of government, followed by population centers. It's true that the death toll from these areas will number in the billions. However, according to The World Bank, almost 70% of the world's population lives in rural areas. These areas will not be targeted directly by blasts, but will likely suffer the effects of fallout.

3rd World Countries and Conflict Zones

Countries denoted by the UN to be 3rd World Countries, or disputed areas that are technically not countries at all, will not have nuclear capabilities. It is possible that nuclear capable countries MIGHT target these areas, especially if they are disputed, but the truth is most of the rest of the world will ignore these. Statistics say that the residences of 3rd world countries number almost 3 billion.

All in all, you could be looking at anywhere from a few hundred million survivors to a shockingly high 3 billion. Deaths in the first few months will be staggering, but it's important to note that the real threat is radiation poisoning, which will end up killing a huge chunk of survivors if they aren't careful. However, you do have the benefit of most of the survivors living in rural agricultural zones. As long as nuclear winter and radiation do not poison the soil they will be able to continue to subsist off the land. Not that this is part of your question, but should be noted in the ultimate death toll.

I had started writing a novel on that theme but then quit it in the middle. When we talk about an exhaustive (no weapons are left unfired), the outcome is more disastrous than you would expect it to be ...

The Subcontinent (India, Pakistan, Bangladesh)

Pakistan would be wiped out of existence with 80% population dead. India would lose nearly 30% of its population in the first wave, but the collapse of government, electricity and gas systems destruction, factories and production units would be rendered dysfunctional. Health services would be primarily overcrowded and then completely destroyed as hysterical people raid hospitals and steal medicines. With winds the fallout would be propagated to far flung areas too. Considering the high poverty and low education rates of Indian suburbs, the deaths in the 2nd phase would be most frightening and horrific. After 1 year, nearly 50% of the population would be dead. After 3 years, only ~5% would survive. These too, would be ghosts of their former self. Ghastly, psychologically destroyed, scared, hysterical and cannibalistic.


Considering that most of the missiles coming to China would be American, and then considering the vast area and population of this country, there are major chances of more survivals here. With all major cities wiped out and government collapsed, the survivors would be found in high altitude rural regions (Shaolin and Wudan mountains) and the Tibetan plateau. The Chinese have an amazing level of unity and discipline, so there are major chances that functional local units would survive together. With lots of game and soil available for agriculture, China would have ~10% survivors after 3 years. However a central government would be missing and China would be thrown back to warring period.


Same as China, but considering its even larger area and much smaller population, there would be considerable more survivors here.


There would be a surprisingly high number of survivors here, considering how many people are already stockpiled and ready for such a catastrophy. However, with all cities razed or evaporated, the survivors would mainly be found in rural areas. I cannot say how many people would die in the first phase (considering USA's missile defense program and whatnot) but the second phase would be horrible. Every survivor would be at daggers drawn against every other survivor. Small groups of coordinating survivors might exist, but after a period of 3 years, most of the survivors would have killed each other, instead of having died of war.


Initially, a lot of survivors, considering almost no country would bomb this town. But the 2nd phase effects and the results of massive conventional weapons bombing would be immense. Fallout carried by winds, the (functional) death of USA and the lack of food imports would cause large scale casualties. Furthermore, fallout would be condensing in this country in the form of arctic snowfall and rains. Most of the deaths would be in 2nd phase. These people would die a slow death of hunger and the gradual exhaustion of health services. Almost all of the population would be wiped out after 5 years, except those where climate is warm enough to allow sustainable agriculture.

Australia And New Zealand

Considering how detached they stay from controversial politics, probably nobody would be interested in nuking these pacifists. If these countries go in-fighting however, that is a different story. Even in this case, at most 40% of the population would be erased, with the remainders having enough soil for agriculture and game for hunting. These folks would survive fine, I would say, unless a few "stray missiles" find themselves landed on these lands and detonating.

The Arabian Peninsula

These lot would take the wrath of an infuriated israel and having only fat, clumsy sheikhs and nothing else (except oil wells) in their regions, they would be quick to die of hunger and government collapse. With no import of food, it would be scary to see the fat sheikhs in their lofty palaces, surrounded by gold, dying of hunger ...


All islands with large enough area and small enough population for sustainable agriculture would survive. However, the islands mainly depending on seafood would die a slow, horrible death as they eat infected seafood and get infected themselves.

Africa And South America

Now this is where things start getting interesting! With no country interested in bombing these ghettos, the deaths would all result of conventional warfare. If all countries invade each other with all force, most of the armies would be wiped out and there would be a large number of civilian casualties. But once the warring factions are rubbed out of existence, the survivors would actually thrive. What with extremely fertile land and what with plenty of game, Africa would become the paradise for survivors and all survivors of the globe would dream of reaching this place. I would dare say 80% of African population would survive ... unless somebody out there gets sadistic and flies a few shots at South Africa, in which case fallout would be dispersed continent-wide, through winds and rains.

South America would survive too. Considering that they have immensely fertile soil, very cooperative peoples, lots of game and nobody would be interested in nuking these folks.


There would be survivors everywhere. However, with no electricity, gas or production lines running, they would be pushed to a lifestyle at least 500 years ago, if not pre-civilization times. In-fighting between survivors for food and medicines would be common in all areas where government has collapsed. Third world people would be affected mostly due to lack of supplies while 1st world citizens would be psychologically crippled due to the destruction of their high-end lifestyle and low survivor instincts. Islands, Australia and most notably, Africa would be survivor heavens where humanity and civilization might survive.

  • Can you add up your estimates to provide a range as the final answer? – Solanacea Apr 17 '17 at 14:48
  • @Solanacea: That is too difficult, considering we don't know which cities would be targeted by which countries, using what powered weapons. We also don't know the defensive measures against such weapons, available to different countries. All in all, I think 4.5 to 5.5 billion people would survive the first wave (the direct damage of the war). After a year, this number would have decreased to 3 - 3.5 billion and after two more years, the numbers would settle at somewhere between 2 and 3 billion. The majority of the survivors would be located in the southern hemisphere. In north China and Russia – Youstay Igo Apr 18 '17 at 10:36

The 2029 timeframe makes this difficult to answer.

  • Consider the concept of virtual nuclear arsenals -- if the global situation deteriorates, many industrialized countries could decide to go nuclear.
  • If Iran becomes a nuclear power, Saudi Arabia might want to respond in kind. Can they? They got plenty of money. Who is next?

That being said, the US and Russia have reduced their arsenals well below cold-war levels. What you describe sounds as if the southern hemisphere will get off lightly, including South America, Africa, and Australia. Things to worry about are:

My guesstimate is millions to hundreds of millions of survivors.

A lot of this comes down to how bad the nuclear winter would be.

There's no doubt billions would survive the war itself. I would also expect billions to die in the chaos that follows--with the infrastructure busted people are pretty much thrown back onto local resources--but we live in such an integrated world that that is basically impossible. Consider my own location: If it's summer it's basically impossible for me to get to any location with any meaningful food production under my own power. If the roads won't pass vehicles that's it. (And that's not counting the fact that trying to haul enough water to survive the trip would just get me attacked anyway.)

What really matters is what's left to rebuild with. If the skies are black you have total crop failure & loss of livestock. There won't be a lot left to loot (expect the cities to burn), soon you'll reach a state where only the preppers have food--and they won't be able to hold onto it against the starving masses.

In the US's stockpile of nuclear weapons the amount of radiation released directly by the bombs has become quite small. If detonated carefully it's estimate that 800-1000 detonations might lead to a world wide increase of cancer of about 1 person per year - nuclear bombs do not need to be dirty. However, I imagine that in this scenario we can assume at least one party would try to maximize damage by making the detonations as dirty as possible. They would do this by detonating the bombs close to the ground to increase the amount of fallout.

Several people mentioned that military targets will likely be the first struck. I'd like to refine that a bit:

  • Opposing military nuclear capabilities will be the first struck along with command, control, communications, and intelligence gathering (this includes civilian governments) for the nuclear forces.
  • After attempting to wipe out the opposing country's nuclear second
    strike capabilities, they will likely switch to conventional military forces AND war fighting capabilities - such as manufacturing.
  • If the opposing country is able to launch a retaliatory strike and you have nuclear weapons which survive it, then you may switch to civilian population centers.

Meaning if you live in a city that makes warships or jet engines, you'll be toast whether they get to the population centers or not.

People living in rural areas do not need the population centers as much as the population centers need the rural areas (at least they don't need them as urgently). I imagine many of the rural survivors will do "ok" for several years. The problem of course is that generally speaking population centers also tend to be manufacturing centers. As things wear out, people in rural areas won't have the ability to replace their technological goods.

People in population centers who survive will be the most likely to suffer and die. since the destruction of so much infrastructure will likely stop the transportation of essentials like food and water to them. People who live in areas which can't normally support large numbers people (such as Southern California & Arizona) will also be at tremendous risk.

People in rural areas that are near such locations could also be at risk as the starving swarms of city dwellers spread out and attempt to get what they need. I generally guess that locations within a 3 day hike of a major city should expect casualties in excess of 50% due to starvation, exposure, and thirst (depending upon season, weather, and other factors).

I think the real answer is that no one knows.

Exactly where would bombs be targeted? Would they be concentrated on military targets or would one or more participants be trying to maximize civilian casualties? Would one side or another succeed in preventing an enemy from delivering its weapons, for example by bombing missile bases before they can launch, intercepting missiles in flight (SDI, etc), or through sabotage?

How reliable are the world's nuclear weapons? How many missiles would blow up on the launch pad, fail to detonate on the target, etc? Is that a tiny percentage or would it be a lot? I read once -- don't have a citation on this -- that the U.S. has NEVER successfully launched a missile from a missile silo. Because launching from a silo would destroy the silo, and they're too expensive to destroy for a test.

Only 2 atom bombs have ever been used in actual war, so there's little empirical evidence to go on. (Not that I'm saying that I wish there was more empirical evidence!) Almost all we have is theory and extrapolation. That's all well and good, but real life experience routinely shows that such theoretical predictions are often wrong. That's why scientists do experiments rather than constructing a computer simulation and then declaring the question closed.

For example, after the Chernobyl nuclear accident -- and I'm not saying that that was a nuclear explosion, just that it was empirical evidence on the effects of release of radioactivity -- after Chernobyl many were predicting that there would be huge numbers of dead, horrible mutations, etc, with some serious estimates running in the millions. I remember at the time experts appearing on the U.S. media discussing how long until the "radiation cloud" reached the U.S. and what we could do to protect ourselves. The actual casualty count ended up 31 direct deaths, perhaps as much as 6000 indirect deaths from radioactive contamination. And that last is disputed and hard to prove either way. The reality is that we did not see millions of people falling over dead. Scientists are studying statistics looking for anomalies.

Will there really be a "nuclear winter"? How many cancers, etc, will be caused by fallout? How much damage will there be to the ability of the ecosystem to produce crops? There's lot of theory and speculation but little hard evidence.

Clearly there would be massive casualties in places actually targeted by nuclear missiles. If major cities are targeted, the immediate casualties would surely be at least tens of millions, possibly hundreds of millions.

Some number of people will retreat to well-stocked bomb shelters. Unless the effects literally destroy the world -- all life is wiped out and the Earth is totally poisoned -- they'll wait out the worst and then emerge to rebuild. I don't know if anyone has statistics on the capacity of such bomb shelters. There are a fair number of "survivalists" in the U.S. who live in remote areas and have built shelters and stocked them with supplies. (Most people laugh at them now as a bunch of crazies, but when the bombs start falling, who'll be laughing then, huh?)

(I read an article a while back on how the U.S. government has shelters for government officials so essential services can be resumed as quickly as possible. One of the agencies listed as having space in these shelters was the Consumer Products Safety Commission. And I wondered: would that really be a priority after a nuclear war? Like, the country has been wiped out. Tens or hundreds of millions are dead. The survivors crawl out of the shelters and attempt to rebuild. Someone tries to repair a damaged tractor so he can get some food production going. And immediately an inspector from the Consumer Products Safety Commission shows up. Do these parts that you made out of scrap metal meet U.S. government standards? Do they conform to the original manufacturer's specifications? Do you have a certificate from an authorized testing lab showing that this equipment has been de-radiated? No? I'm sorry, sir, we cannot allow you to operate such unsafe machinery.)

People living in rural areas, and especially in remote areas, will likely survive. Someone mentioned people at Antarctic research stations as likely survivors. Same could be said for people in northern Canada, most of South America and Africa, Pacific Islands, etc. I'm hard pressed to think of a scenario where anyone is going to fire a nuclear missile at Rwanda or Tuvalu.

I think that as a large-scale global nuclear war starts to develop, many or the majority of the military commanders in operational control will refuse to fire. They have children and grandchildren. They have long and well-informed study of the real effects, both primary and secondary, of a nuclear conflict.

Humans form local communities much faster and more easily, with more commitment than they do city-states or nations. These communities will pool knowledge of all kinds to enhance survival, even in the individualistic U.S.A. In fact, I think that will occur most probably and powerfully in the United States and similar countries, since they currently have a more well-developed sense of individual empowerment, patriotism and manifest destiny than any other people in the world, even in these latter days. Do not rely on their mass media to gauge the character of the citizens of the United States since it is controlled by very few and the very biased. Get to know the people themselves and then make an assessment. I think that the survival rate in other developed countries will be similar, but I think that the recovery back to a technological society will be much faster for the USA then anywhere else. This is also partially due to the increasingly high level of survival preparedness in America due not to nuclear fears but to fears of a Nazi police state or to economic debt collapse. And they will still have their weapons intact which will not decrease, but increase the likelihood of restoration of local law in the earliest days after the war.

  • 3
    Hi Ian, and welcome to Worldbuilding SE and Stack Exchange. You bring up good points, but in the OP's scenario, the war already happened -- twenty years back from the time under consideration. The question is how many survived, which I don't see this answering, so I have flagged it as not an answer and it may very well be deleted by our experieneced community members. Feel free to edit your answer such that it answers the question that was asked. I encourage you to take the quick site tour, and review the help center, to learn more about how our format works. Hope you will stay around! – Michael Kjörling Aug 10 '16 at 5:24
  • 1
    Welcome to the site, Ian. I have to agree with Michael here. Answers are expected to answer the question as asked. While I agree you make some good points, I also get the feeling you have a secondary agenda beyond answering the question, which detracts from your post. – Frostfyre Aug 10 '16 at 12:13

Your Answer


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.