Through some evil scientist man's evil plot to destroy the world, he transported the entire Earth away from the sun into an unknown point in space where there is no sun. Could we recover?


  • assume the transportation was instantaneous and there were no collisions with random space bits.
  • Earth and buildings and such are unharmed by the transportation
  • Earth still has an atmosphere
  • humans were not harmed by the transportation
  • the nearest star is too far away to provide any warmth
  • the Moon is still there. You just can't see it
  • any space junk between earth and the moon is still there (satellites, old rocket engines, passing space rocks, etc.)
  • people have three months heads up to prepare

Since the only thing missing is the sun and the rest of the solar system, could we recover before we all die?

I don't want any 'we're all dead' answers because those make for bad story lines. Try to be optimistic for the next five minutes and post a good answer that will help.

This is not a duplicate of What would happen to the Earth if it were suddenly transported to a completely empty void? or If only the sun's light went out, how long would it take for all living things to die out?. These focus on the effects of such a transportation, but I'm concerned about the recovery process.

  • 1
    $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch
    Sep 24, 2020 at 20:35
  • 6
    $\begingroup$ xkcd addresses some 'pros' in this What-if $\endgroup$
    – Robotnik
    Sep 25, 2020 at 1:33
  • 1
    $\begingroup$ It just occured to me, is the Earth still spining, as normal? Is it still going at the same velocity as it did when orbiting the sun, only in a linear orbit, tangental to its position in its orbit when it was absconded, not a circular one? It does not make a huge difference, in comparison to the loss of the Sun, but somehow the energy of momentum has to be accounted for. Especially the lost momentum in the molten core. That is a lot of energy to have suddenly 'lost', and the Earth's atmosphere would certainly complain. $\endgroup$ Sep 25, 2020 at 1:50
  • 2
    $\begingroup$ "Since the only thing missing is the sun..." Heh. :-) "Other than that, Mrs. Lincoln, how was the play?" $\endgroup$ Sep 25, 2020 at 6:43
  • 1
    $\begingroup$ This is a begging belief. The absolute closest thing that could happen is if space were folded between two points and physically moved through that fold in space which would certainly not be instantaneous. A begging belief is basically impossible cartoon physics that literally only work in the context of comedy. Good story telling does not require excess belief and you should never expect your audience to keep reading/watching after such a stretch unless again, this is being used explicitly in a comedy. $\endgroup$
    – John
    Sep 25, 2020 at 10:29

17 Answers 17


We Can't

Nearly everything you see is powered, directly or indirectly, by the sun. Without it, the Earth cools down quickly and everything will die: vegetation, animals, humans. If you can get to a heated shelter with years of food, you can only last longer. The cold is so immense, any mechanism will fail, the atmosphere does weird stuff under such cold temperatures. What is left of humanity simply doesn't have the manpower or equipment to stay alive, let alone recover.

  • 37
    $\begingroup$ "The atmosphere does weird stuff" = it becomes a layer of solid ice oxygen and nitrogen ice, a few meters thick. $\endgroup$
    – AlexP
    Sep 22, 2020 at 21:07
  • 3
    $\begingroup$ @AlexP Unless I'm somehow mistaken, as the air cools, the component gases would rain (or snow, if they don't have a suitable liquid state available, like CO₂) down one by one, as determined by their boiling points. The liquid ones would then flow (along the rivers and other paths previously taken by water) towards the low ground before eventually freezing. They might also get absorbed by the (deep-frozen) ground before that. $\endgroup$
    – Bass
    Sep 24, 2020 at 10:34
  • 3
    $\begingroup$ @Bass: Yes, the details would be intricate. (I doubt they would be absorbed by the ground, which would most likely be saturated with water ice.) I am too lazy to Google whether liquid oxygen and liquid nitrogen would mix or not. $\endgroup$
    – AlexP
    Sep 24, 2020 at 10:43
  • 1
    $\begingroup$ What about the dinosaur juice? Or going sufficiently underground (the geothermal gradient is about 25–30 °C/km and there is so much insulation by the rocks that it isn't necessary to go that deep in the short term)? $\endgroup$ Sep 24, 2020 at 19:09
  • 1
    $\begingroup$ @JustinThymetheSecond although you're rightabout the Earth not cooling down because of a vacuum insulates, it'll cool down rapidly in any case thanks to many processes. If this wasn't true, where would all the heat build up during the day go to? Or why would seasons matter in relation to the sun? We would live in an ever increasing hot world. Without the sun it's essentially a winter night 24/7 and it'll quickly approach temperatures that are unlivable. $\endgroup$
    – Trioxidane
    Sep 26, 2020 at 20:49

Yes, you can recover, but you must act fast. The only hope is the emerging planet teleportation technology.

Put all human resources to either:

a) find that evil scientist and force them to teleport Earth back, or near some other suitable star; or

b) duplicate their technology and do the same.

Best strategy may be to try both at the same time.

  • 2
    $\begingroup$ You are assuming that evil scientist is still on earth. $\endgroup$ Sep 23, 2020 at 12:47
  • $\begingroup$ @stackoverblown Granted, the word "can" is a bit optimistic; "the only hope" gives a better sense of the chances. We search for the scientist; we might not find him. We also have plan b, which likewise might fail. $\endgroup$
    – David K
    Sep 23, 2020 at 13:12

Survival for a small part of humanity is possible for some time, long-term survival is unlikely, full recovery almost impossible.

Someone calculated that it would take almost a year for the top 100m of Earth (including the oceans) to freeze solid. It would take many, many years for the oceans to freeze through.

So there is some time for the construction of deep underground and submarine shelters, and using geothermal and nuclear power it is in theory possible to maintain these (growing food with artificial lighting) for a long, long time. Sure, it's not renewable energy, but strictly speaking, nothing really is.

The problem is that it's obviously not feasible to do this at a scale that can support all of humanity, or even a significant part. In fact, it's questionable whether it's possible at all to do it both quickly enough and thoroughly enough to last - constructing power plants usually takes years, and this situation would not only give much less time but also add challenges (such as the whole thing having to be underground, and of course freezing cold and everyone on Earth panicking).

And then there's maintenance. You'll have to build these power plants using all the latest technology that's become standard. But you won't have that technology around indefinitely. A lot of modern technology basically requires out entire global civilization to produce. Most significantly, microchips. Producing a current-day CPU requires so much know-how and capital to make the tools to make the machines to make the tools to make the machines to make the machines to make the actual CPU that only a handful of companies on Earth can do it.

A few tens of thousands of humans living in shelters deep under the surface won't be able to maintain that level of technology, not even close. So when the original tech starts to fail, they will almost certainly die.

Then again, the ingenuity of humans is often surprising, and they built nuclear power plants before there were microchips, even before were mass-produced transistors. Maybe they'd manage somehow.

Even then, though, recovering to a billions-strong civilization doesn't seem possible.

  • $\begingroup$ Given the time to build the powerplants up to specification, would we have enough nuclear resources to power our nuclear sunless planet for any stretch of time? $\endgroup$
    – Flater
    Sep 23, 2020 at 12:35
  • 8
    $\begingroup$ @Flater: For a population shrunk down to a few million at most, currently stockpiled Uranium fuel would last decades if not centuries, giving us time to mine more. But geothermal energy might well be more feasible - actually more so than now because the heat differential (which is what really matters when you want to generate electricity from heat) with the surface would be bigger. $\endgroup$ Sep 23, 2020 at 14:00
  • $\begingroup$ 'A few tens of thousands of humans living in shelters deep under the surface won't be able to maintain that level of technology, not even close.' A few million, especially if specifically selected for intelligence and knowledge, would. $\endgroup$ Sep 26, 2020 at 1:55
  • 2
    $\begingroup$ @JustinThymetheSecond no way. Even completely discounting the necessary human expertise, no way we could assemble the necessary supply chain in one location from its disparate pieces all around the globe in that short a time. We'd be lucky to get the artifact, no way would we be able to condense the means to reproduce it. Maybe if we had 5-10 years to get ready you suggest in your answer. Not in a handful of months. $\endgroup$ Sep 26, 2020 at 3:31
  • 1
    $\begingroup$ @JustinThymetheSecond for how long though? And what if anything, anything at all goes wrong at all with a key link in the chain? Military insurrection? Popular rebellion? Pandemic? Nuclear accident? Nuclear "accident"? I'm sorry, I just don't buy the "plants all die but otherwise business as usual" scenario. People are going to freak, without enough food you'll have riots followed by warfare followed by warlords followed by the complete breakdown of society. $\endgroup$ Sep 26, 2020 at 23:02

Most of us die within a year. Some groups can survive for a few thousand years.

We could theoretically all survive it, the oceans and ground would hold enough heat for the better part of a year to keep the atmosphere breathable. To survive it we'd need to build in 6 months:

  • Convert all existing large underground space into a bunker.
    • There is a lot of pre-dug large underground spaces already on the planet. We call them mines. For example. A single salt mine in Pakistan has enough space to hold 110 million people (in bunks). There's enough underground mines in the state of NSW, Australia to give everyone in Australia 50 cubic meters underground each.
    • These would need to be insulated, divided into dorms / small appartments, furnished, connected to utilities, etc.
  • Or, build a dome over an existing city.
  • Install a small nuclear power plant or geothermal plant in each
  • Install hydroponics and water purification system.
  • Install air and CO2 filtration systems.

That's a lot of work. With 0 warning, 7 billion people working with a near-hive mind in perfect unison could probably get it done, but realistic humans, panicking, disconnected, poorly organised, and all thinking of their own personal issues, no chance in hell. 3 Months warning, some organised, functional societies may get it done.

A group who gets a bunker up and running could survive for a few thousand years in sealed underground bunkers, and could stretch that for another few thousand by digging down further.

But, looking at how we've handled existing, much slower climate problems, it shows that many countries won't be able to accomplish this. As a society, we're still trying to build new coal power plants despite pleas from everyone who doesn't get coal money - many parts of the world would have no hope of getting our act together in time.

So, who survives? Pick a disaster and compare how the people of that country responded to it. USA? Covid19 - enough said. Katrina? Yeah they're doomed. Europe not looking too good either. I'd give Australia a 10% chance of getting a small colony up, we've done very well at containing covid (our critical 2nd wave Victorian outbreak was better than business-as-usual elsewhere in the world) and our bushfires had a very low death toll compared to smaller bushfires in Brazil and America. Most of the 3rd world is gone.

But a very industrialised group, with nuclear power already near their cities, with good production and R&D near their cities, and an authoritative state controlling every part of their lives, could dome their cities within 6 months. It's very possible some Chinese mega-cities can dome themselves in this time. They'd need a lot of nuclear fuel for this, and all the Uranium deposit in China are far away from the cities. They'd need to mine very quickly (as the deposits are going to be harder to get once the ground freezes) and set up hydroponics very quickly, but its just within the realm of plausible that a hundred million or so Chinese survive.

I'd want to give some Arab cities a decent chance too, however the logistics of importing parts and materials needed during this time of panic, not good.

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch
    Sep 26, 2020 at 6:41
  • $\begingroup$ You neglect that we dont know how to make closed environments that actually work. Even with preparation time and decent conditions our best efforts are measured in months.. en.wikipedia.org/wiki/Biosphere_2 $\endgroup$
    – Innovine
    Sep 29, 2020 at 13:28
  • $\begingroup$ @Innovine Biosphere 2 was a publicity stunt, nothing more. China has done some very serious research. Their current record is one year, successfully completed. It is called the Lunar Palace 1https://www.space.com/40610-china-mock-moon-mission-lunar-palace-1-photos.html but you will not see it come up in a Western google search, unless you search it by name. $\endgroup$ Sep 30, 2020 at 16:21
  • $\begingroup$ Interesting, but still shows that our biosphere will collapse before we run out of power. $\endgroup$
    – Innovine
    Oct 1, 2020 at 6:26

Since there is speculation that we might survive on geothermal and nuclear power, I decided to run some math.

If we had the technology to mine all the uranium in the crust of the Earth, we could gather enough uranium to extract about the same amount of energy as Chicxulub. My calculations are in Is it possible to kill all life on Earth? We definitely don't have the technology to mine al the uranium in the crust because:

  1. we can barely dig 12 km experimentally, whereas continental crust has a thickness ranging from 30km to 50km according to wikipedia. Some sources cite up to 70km in some points.

  2. that would involve filtering all of the Earth's crust, literally. The surface of the planet would be unrecognizable.

Let's assume that, by magic, we teleport all that uranium to the surface and use it. According to this handy table, we have about 5 $\times$ 1023 joules available. That is almost 10% of the total solar energy that we get in a year. Let's say we have one month worthy of global solar energy (the next item on the table).

If we could also magically convert all that energy into sunlight, we could keep the planet running for a month. Let's say we divert it all to agriculture, though. According to the Food and Agriculture Organization of the United Nations, about 11% of the world's surface were farmlands in 2003 (see section 4.3). So we might keep farming viable for 10 months.

But you see, there is a lot of handwaving and magic being done to achieve those numbers. In reality we wouldn't be able to keep the world going as is. Even the logistics to keep a regular nuclear power plant going involve chains of logistics that span areas much beyond the actual power plant.

Suppose we decide to build a self-sufficient plant on top of an uranium mine. With forewarning, we could build a bunker that could sustain thousands of people with hydroponic farming, right?

Well, no, because:

  1. We could barely keep eight people in a "self-sufficient" space designed to provide sustainable farming. We have research on that, both Biosphere missions were textbook fiascos.
  2. A few thousand people would probably provide for very poor genetic variability. This would not kill humanity directly. But over time the population would be so homogeneous genetically that, combined with the small population, it can easily and quickly be wiped out by a new germ.
  • 1
    $\begingroup$ '...both Biosphere missions were textbook fiascos.' BOTH? As if only the Americans are capable of doing this, and as if only their results are valid. The Chinese have run a successful experiment for a year, and it was highly successful. If we depended on the Americans, Earth would be doomed for sure. Fortunately, there are the Chinese. $\endgroup$ Sep 25, 2020 at 4:41
  • 2
    $\begingroup$ @JustinThymetheSecond by years you mean a record 108 days or much less time than even the first failed biosphere project. $\endgroup$
    – John
    Sep 28, 2020 at 21:01

To my understanding, very little of life on Earth can be subjected to deep interstellar temperatures and radiations and survive. Some 'extremophile' bacteria and a very few species like tardigrades can last for a short time, but in the end, no. We're all toast. Very Frozen permanent toast.

  • 8
    $\begingroup$ The life that clusters around deep-sea geothermal vents will last for a long, long time. It might even last for longer than the billion or so years that it has under the current circumstances. That’s no good to human beings, of course. $\endgroup$
    – Mike Scott
    Sep 23, 2020 at 7:25
  • 1
    $\begingroup$ Worth noting that even the geothermal vents wouldn't get away with things unscathed - the oxygen content of the sea at atmosphere depends on photosynthesis, so any food chain that requires oxygen would die, and apparently black smokers run off of oxygenic chemical reactions. Alkaline hydrothermal vents might make it, but would they persist as they do in an otherwise-frozen ocean? Would they adapt to shifting chemical composition in time? Plausible enough but I don't think it's a sure bet. $\endgroup$
    – Oosaka
    Sep 25, 2020 at 14:03
  • $\begingroup$ The Earth would NOT be subhected to 'deep intersteller temperatures'. Ther is no convection nor conduction of heat. In fact, thee is no such thing as 'intersteller teperatures', because unless one has matter, there can be no temperature. Like a thermos on earth, a perfect vacuum is the perfect insulator. If your coffee does not cool down in a thermos, the Earth will not cool down in a vacuum. $\endgroup$ Sep 26, 2020 at 20:36
  • $\begingroup$ @JustinThymetheSecond do you know what therma radiation is? I'll give you a hint its the same way we get heat from the sun in the first place. a perfect vacuum is not a perfect insulator, blackbody radiation is a thing. How do you think space craft cool themselves? the earth emits about 112 Petawatts of heat into space. $\endgroup$
    – John
    Sep 27, 2020 at 22:58
  • $\begingroup$ @ John And most of that is just reflected solar radiation, which will completely stop when there is no Sun. You can't re-radiate energy that does not hit the Earth in the first place. A vacuum IS a perfect insulator. Insulation protects against convection and conduction heat loss. Radiation shielding protects against radiated heat loss. They are very different forms of heat loss. Insulation is useless against radiated heat loss. That thin layer of highly reflective film inside your thermos works well against radiated heat loss, but it is not considered insulation, it is considered shielding. $\endgroup$ Sep 27, 2020 at 23:10

I'm going to be a dissenting optimistic opinion (to a degree).

A major chunk of Humanity will die, and it's very possible that the knock on effects will cripple our ability to react to to the crisis.


If short term survival can be achieved then there's cause to believe we would be able to recover in time.

Firstly, Earth's supply of fissile materials is actually quite extensive - especially once you factor Thorium into it. Fission power in this setting becomes even more attractive than in real life because the excess heat can be used to heat homes and cities, and the Carnot efficiency of the power generation systems increases as the exterior temperatures drop. I don't know what people are talking about in terms of hurdles with nuclear fuel, it's all bollocks from what I can tell. Why would processing, disposal, or handling of nuclear fuel be show stoppers? Secondly, we can burn as much oil as we want in the short term with no ill effect to tide us over until we can build up enough nuclear capacity. (the ecosystem is dead anyways and we actually want as many greenhouse gasses as we can to slow down the cooling of the planet).

Now, you don't need bunkers to survive either, it's much easier to insulate surface structures. It would become a lot more expensive to feed a person because agriculture would have to move into greenhouses with artificial lighting, so population growth would be slower, but there's no physical reason I can think of that we couldn't recover if we could survive the initial upheaval.

Weather would stop after a while, so we'd see far fewer natural disasters as well (except for earthquakes).

edit: Also, a clear and present danger such as this event has a way of bringing people together and focusing the on the problem at hand, so I feel like there's be less infighting than one would expect.

  • 1
    $\begingroup$ there is a lot of fissile material, but they require advanced technology (and the billions of people supporting it) and lots of labor to gather and process. there is not enough fissile material on the planet to replace the sun for this many people for any significant amount of time. There is no recover just prolonging the inevitable. $\endgroup$
    – John
    Sep 23, 2020 at 16:04
  • $\begingroup$ We don't need enough fissile material to replace the sun, we certainly don't use even a fraction of all the solar energy that hits the Earth at a given moment. I'm also not sure why you think fissile material is so difficult to extract - given that it's sold for about $20 per lb right now. This is even more true because in this case we could forego all environmental considerations. $\endgroup$
    – Algebraist
    Sep 23, 2020 at 16:14
  • $\begingroup$ that is the price of ore, not the usable product, and the existing production facilities will not work, you need facilities that can work in what is essentially vacuum. $\endgroup$
    – John
    Sep 23, 2020 at 16:34
  • $\begingroup$ You have a point, that's the price of ore, refining it is more intensive, but fuel grade uranium is still cheaper per unit of power produced than natural gas - nuclear plants are more expensive because of the greater up front capital costs of construction and certification (many of which don't apply to this scenario), not the price of the fuel. $\endgroup$
    – Algebraist
    Sep 23, 2020 at 19:14
  • $\begingroup$ Why would they need to work in vacuum? It will take years (if not decades) before the atmosphere will cool off enough to liquefy, and after that things can be done in pressurized containers. In fact, the liquefaction of the atmosphere will likely ease the temperature issues by providing a ready source of insulative vacuum. $\endgroup$
    – Algebraist
    Sep 23, 2020 at 19:14

You want to be optimistic? You need a generation ship (or the earthbound equivalent of it) - this science fiction mainstay is a spaceship that's self-sufficient in power, food, population, medicine, spare parts, and everything else needed to sustain a population for thousands of years. Of course, they're designed to survive in the vast gulfs between stars.

These show up in Sci-Fi pretty frequently - although perhaps that's because it's a good mechanism for social critique, rather than because it's particularly realistic.

You could choose whether to make it a spaceship or an underground biosphere - not needing to go to space keeps things simple, but some would see an eternity trapped in frozen caves as a bit of a pessimistic outcome.

Could we build such a thing at all? Could we do it with only a year or two until the planet froze over? Could all the world's great powers unite and pool their resources, or does their nature preclude that? Would there only be space for a tiny fraction of earth's population? If so, would the 99.99% destined to die be helpful to the efforts? All questions you can explore in your writings.

If you're interested in this subject, you might enjoy the first half of Neal Stephenson's book Seveneves which has a premise with a lot of overlap.

  • $\begingroup$ The world's great powers do not need to pool their resources. It can be done by one country - China - singulalry. The other nations could come along for the ride, if they wanted to. $\endgroup$ Sep 25, 2020 at 5:06
  • $\begingroup$ The planet will NOT 'freeze over' in a year. It will get bitterly cold, but humans will still be able to function. All of our buildings, manufacturing plants, roads, ships, tunnels, power plants, electric grid, oil wells, gas wells, airports, will still be there, probably useable, most likely still functional if they can be kept heated. Earth will not be a wasteland 'void;' like Mars or the Moon. Everything we built, is still here. Except for batteries, unless they are warmed up. $\endgroup$ Sep 26, 2020 at 21:36
  • $\begingroup$ @JustinThymetheSecond if you define "freeze over" as the oceans freezing over, it takes around a year or two for that to happen. $\endgroup$
    – John
    Sep 27, 2020 at 4:06
  • $\begingroup$ @JustinThymetheSecond China does have a lot of manufacturing infrastructure, but we might only have 1 or 2 attempts to build something that works at large scale, and previous biospheres have only housed a few people and haven't been self-sufficient in sunlight or power - let alone self-sufficient in medicines and microchips. And we'd be doing it all in the dark. So it won't be easy, even for China. $\endgroup$
    – mjt
    Sep 27, 2020 at 9:28
  • $\begingroup$ @John We have devades. No credoble scientiffic source claims that the oceans will freeze over in a year, If youave one, I woul dlike to see it. That is pure 'conspiracy theory' science, $\endgroup$ Sep 27, 2020 at 21:14

Yes, but you need technology and raw materials...

Phase 1

  1. Pump huge amounts of greenhouse gasses into the atmosphere to buy time.
  2. Build as many geothermal power & heating plants as you can.
  3. Put up inflated foil domes (bubble-wrap) over each city to keep heat and air in.
  4. Use huge arc lights to shine up towards the foil dome and reflect the light down.
  5. Pump water from the ocean to heat the land surface, the ocean re-heats from the earth's core.
  6. Build many nuclear fission and fusion reactors near every city to power the lights and heating, shipping container sized plants currently take 5 years to build.

Each city becomes a self contained life support system, outside the cities nature eventually dies away. The heat from the cities keep the atmosphere from complete collapse but is very wasteful of energy emitting to space.

Phase 2

  1. Build cities under the ocean, near geothermal sources if your technology can handle the earthquakes.
  2. Fleets of space ships search out and mine resources on asteroids and further away planets.
  3. Build 1m thick concrete domes over cities for radiation shielding and insulation as the atmosphere freezes.
  4. Eventually connect the cities together so that the concrete dome spans the whole planet.
  5. After a long time of mining, the planet may be turned into a miniature dyson sphere but instead of being powered by a star, it's powered by nuclear fission and fusion.


  • In 2015 humans consumed 5 × 10^17 btu energy, which equates to consuming 6750 kg of relativistic mass m from E = mc^2.
  • The earth weighs 5×10^24 kg which can last 8×10^20 years, ignoring practicality and inefficiencies.
  • The energy needed from matter conversion to run the planet is equivalent to the amount of energy lost to space, better insulation means less energy needed.

Why not stick to coal power? The following calculations indicate there's ample oxygen in the atmosphere to run the planet on coal power for years.

Weight of earth atmosphere
= 5×10^18kg
Atmospheric oxygen as percentage
= 20vol%
Weight of Atmospheric oxygen 
= 5×10^18kg * 20vol%
= 1×10^18kg

Allocate 33% of oxygen to human & animal breathing, 
33% to electricity generation and leave 33% as unextractable.

Human and animal breathing:
Weight of Atmospheric oxygen budgeted
= 1×10^18kg / 3
Oxygen consumed per person sized animal per year
= 740kg
Number of humans and animals saved
= 7 billion humans + 14 billion animals
= 21 billion
Max person years of oxygen extractable from atmospheric oxygen
= 1×10^18kg / 3 / 740kg/year / 21 billion animals
= 21,450 human & animal years of oxygen for breathing.

Electricity generation:
Weight of Atmospheric oxygen budgeted
= 1×10^18kg / 3
Weight of CO2 per kWh electricity from Coal 
= 1kg
Oxygen percentage weight of co2 
= 72.7%
Weight of O2 per kWh electricity from Coal 
= 0.727kg/kWh
Annual electricity generation
= 2.2×10^13 kWh/year
Max years of electricity supply extractable from atmospheric oxygen 
= 1×10^18kg / 3 / 0.727kg/kWh / 2.2×10^13kWh/year
= 20,841 years of coal based electricity available


  • $\begingroup$ It takes years to build a nuclear plant, after the first year internal combustions engines don't work. $\endgroup$
    – John
    Sep 27, 2020 at 4:11
  • $\begingroup$ Thanks John, you got me intrigued so I ran the calculations. It seems we don't need nuclear power for thousands of years. Do you concur with these calculations? $\endgroup$ Sep 28, 2020 at 11:23
  • $\begingroup$ one problem with coal is after about 20 years the atmosphere starts to liquify, which makes combustion an issue. also building a concrete dome over a city is basically impossible with earths gravity. Also concrete won't set in the cold so concrete in general is largely useless after a year, also building things out of concrete means you are losing oxygen to the concrete for the next few decades. your biggest issue is how long it takes to build any of this, you have a very short time in which to build things before current machinery just doesn't work. $\endgroup$
    – John
    Sep 28, 2020 at 20:42
  • $\begingroup$ Do you mind showing how you got to the number 20 years and not 20,000 years as in my calculations? Since earth can only loose heat through electromagnetic radiation it may be difficult to cool it below 100K, meaning Nitrogen will remain a gas indefinitely. sciencedaily.com/releases/2018/09/180924153430.htm sciencedirect.com/topics/chemistry/blackbody-radiation $\endgroup$ Sep 30, 2020 at 0:10
  • $\begingroup$ What calculation, you show no calculation, I used the numbers from experts, I would not even try such a calculation myself. the earth sheds as much heat as it receives from the sun, (it has too to have a stable temprature) ~340 watts per square meter or 1.73 × 10^11 total. earthscience.stackexchange.com/questions/9210/… and the many times this question has been asked before (to follow) $\endgroup$
    – John
    Sep 30, 2020 at 2:09

Interesting question but as per the other answers, this would lead to the death of almost all species within a few months. Humans would fair a bit longer, well the few who have access to underground bunkers with large reserves of food and fuel. Some species of bacteria would survive for eons, as if the earth had not moved at all.

Alternate Interpretation

But you do not mention if we have warning (although the question does imply almost no time). Given enough time to prepare, we could survive and even expand our civilization to the stars beyond.


With enough time, we could shift all of our manufacturing and agricultural infrastructure below ground, granted we have a united and coherent populace where cost is of no concern.

Underground dwellings would offer insulation from the sub zero temperatures of the surface and agriculture would produce not only food but oxygen for the inhabitants. Manufacturing could continue as oil and the various metals/minerals we rely on would surround us. I'm sure there would be a lot of details we would need to iron out and technologies that would need to be researched, but with warning we could move this underground. Obviously as this infrastructure will need to be powered!


Geothermal activity would still be present for billions of years and could be used much as we do today. Uranium and Thorium are abundant in the crust and both could last us around 100,000 years depending on how efficient we can get. With the large deposits of solid oxygen at the surface, we can also make use of bio-fuels in combustion for remote work that is off the grid (if battery tech does not advance).


How society as a whole would fair is a different story as the human mind is complex. Would we all become manic-depressives without sunlight? Would society lose it's cohesion leading to tribal warfare? It's hard to say but an interesting topic for another question.

  • $\begingroup$ the question is do we have enough time to prepare, you have less than a year before petroleum fuels begin to freeze, and a decade or two before the atmosphere starts to liquify. $\endgroup$
    – John
    Sep 27, 2020 at 4:09

TL:DR It is plausible, probable, and in fact very likely, that at least a small portion of humanity will survive and indeed prosper if that event actually occured today. Further, if this question were asked, or this event occurs, in as little as ten years from now, the answer would be an even more certain 'yes'. Change that 'three month' lead time into a 'five year' lead time, and our chances are almost completely assured.

We already have technologies that would allow us to survive, and even more are in the pipeline, and are being ramped up. We are developing technologies that will allow us to survive in the depths of space, on very long journeys, on alien planets, and in very hostile environments. In fact, surviving under such conditions would be equivalent to surviving on the moon, during the lunar night, and even to some extent on Mars. Certainly, the first food production in these colonies will be biodome hydroponic based, using synthetic solar power (non-biological) technology. Replace solar power with nuclear power, and all of the other technologies we are developing for survival on these space habitats are relevant.

With full apologies to the answer provided by @Ash, unfortunately for America and the West, however, the most likely country to have any survival chances will be China, if this event happened today, with only a three month lead time. They are now the undisputed world leaders in advancing the necessary technology. (It is also worthy of note, that the Western news media is unfortunately deliberately obfuscating and downright censoring exactly how advanced China and Asia are in this area, and in fact are deliberately throwing roadblocks in the way to prevent China from developing the necessary technology. See this and this, for instance. Even Internet searches are being filtered. Legitimate sites highlighting Chinese technology advances that came up as late as last year, no longer show up in Google or Duck-duck-go.)

Basically, if Sci Fi writers can argue that Generation Ships are plausible, and people can survive hundreds of years on them, certainly humans can build the equivalent here on earth, and equaly survive for hundreds of years, probably longer. I do not see why @mjt was down-voted, as the answer just needed to be tweaked from humans building a Generation Ship, to humans creating the same concept here on Earth. That is, turn the Earth into a Generation Ship.

We already have aircraft carriers and submarines powered by nuclear rectors that can go years on their own power. They have been doing so safely for almost 50 years. But now, the technologies to do so are no longer military secrets, and small scale nuclear reactors are going into commercial production. Last year, a google search could get you the names of major corporations in China that were leading in the field. In fact, China is already well into commercial production of community-sized nuclear reactors, so the production facilities already exist. China and Asia, are the world leaders in small nuclear reactor technology, and the US are Johny-come-lately to the party. If anyone will survive, it will be because of these Small Modular Reactors (SMR's)that power our survival in local, even remote, enclaves.

We have biodomes in Canada's North, and in the sub-Arctic that can produce enough food to provide a small community, based entirely on artificial light. They have been built quickly, with limited resources. Combine one of these with a commercial community SMR nuclear reactor (at least one is being proposed already in Canada's North). Given the incentive of assured destruction, and the complete evaporation of any fiscal or economic barrier (money will become meaningless) humans can be mobilized into churning them out in mass in three months. A government that can quickly set policies in place, centralize administration, and mobiize the entire manufacturing and production of the country, will be best positioned to survive. Some countries, however, will be rendered impotent through partisan and ideological in-fighting, back-stabbing, political posturing, and blame-gaming while the citizenry dies off.

We already have in Canada underground salt caverns sufficiently large enough for a Generation Ship type environment, as one will realize if they take a tour of such caverns. Here are some pictures depicting how massive salt mines are. It would take a Herculean effort, but not beyond the capabilities of a determined socially responsive nation, to collectively outfit and stock these caves into shelters that could provide refuge in the short term. Incidentally, the interior temperature of the Gooderich salt mine is a balmy 70 degrees F, and this does not come from the sun, but from the Earth's core. Deeper mines have temperatures upwards of 130 degrees F or higher. Definitely not a deep freeze environment. Add hydroponics for food, and insects for protein, and you have a habitat that will be warm and cozzy for millenia. China, of course, has an abundance of such large cavernous mines, all at downright tropical temperatures.

Oh, and that oxygen thing? Here we have a distinct advantage over Generation Ships. We have an almost unlimited supply of water for electrolysis, given that we no longer have to share it with other life forms. The energy to drive the electrolysis? Did you read that 'mini nuclear reactor' thing? Not a practical solution, you might say. Yet American nuclear submarines get all the oxygen they need, for staying under water for perhaps years at a time, from the electrolysis of water using energy from their nuclear reactor. Except for food, a nuclear submarine is entirely self sustaining for every human need, completely independent of the sun, for years if necessary. Just up-scale the level to that of a large cave, and it is evident that human survival on a Generation Spaceship Earth is entirely feasible.

As for a practical, real-world example, again I have to turn to China. They have already completed (2018) and tested such a habitat, with inhabitants surviving for over a year in a completely enclosed, self-sustaiing module, as a proof of concept.

With all of these technologies currently in development, it is certainly feasible today, and when they are perfected within the next ten years, highly likely, that at least enough humans could survive such a catastrophe to maintain our existence.

After all, all humans arose from a population of only 5,000 or so after an extinction event some 50,000 t0 100,000 years ago. Humans with far less technology than we have today have survived ice ages, with very limited resources. They found a way, they adapted.

  • $\begingroup$ Thawing and splitting ice eats a tremendous amount of energy, you will need nearly a nuclear reactor per household. The real issue is the time constraint, you have less than a year of working time on the surface, and most of the existing machinery won't even last that long. we just don't have much equipment designed to operate below the freezing point of oil. $\endgroup$
    – John
    Sep 27, 2020 at 4:19

The obvious thing to do is to position the moon (which we still have) at the right distance, and persuade it to function as a star. (This would give us something to orbit around, which I presume helps, but that would require a particular combination of distance and orbital speed.) I have no idea whether or not this is feasible. I am hoping other people can fill in that gap.

  • 1
    $\begingroup$ Hi, Candace, welcome to Worldbuilding! I've edited the information from your comment into your answer; feel free to edit it yourself additionally if need be. $\endgroup$
    – HDE 226868
    Sep 25, 2020 at 19:01
  • $\begingroup$ So can you help me with this I have no idea what I'm doing? $\endgroup$ Sep 25, 2020 at 19:26
  • $\begingroup$ If you want to if not I understand I'm just asking. $\endgroup$ Sep 25, 2020 at 19:38
  • $\begingroup$ Sure. What do you need help with? $\endgroup$
    – HDE 226868
    Sep 25, 2020 at 19:40
  • 1
    $\begingroup$ Well, I'd recommend starting by taking the tour and checking out some of the pages in the help center. If you have any questions after that, you can ask the community on our meta site. $\endgroup$
    – HDE 226868
    Sep 25, 2020 at 19:43

This is not an answer directly to 'How would we recover?', but it is an answer to the implied question 'How bad would it get?'

I have not found a single credible scientific reference anywhere that even suggests the atmosphere of earth would in any way disappear, as in 'blown off into space'. In fact, the absence of the solar wind seems to make this less likely. There is nothing to 'blow it off', and since gravity is not affected, it would still be gravitationally bound to the Earth. I think it is safe to assume, in light of the absence of any credible theoretical mechanism to the contrary, that Earth's atmosphere would remain.

Many posters are suggesting the atmosphere would quickly freeze, but again I have found no supporting scientific evidence for that, just a lot of baseless opinions. From everything I have read in credible science refrences, the earth would not 'instantly freeze'. The article with the most credence indicates that it would quickly fall below zero degrees F, probably within a week, but that is just a mild winter day. In fact, it WOULD be like a dark winter day, all over the earth, so it is entirely reasonable that the earth would cool to this level. But even if, as the article suggess, if it dips to -100 degrees F, within a year, that is still surviveable, with enough energy and insulation. And according to the article, it would take millions of years for the surface temperature to drop to a stable -400 degrees. At that temperature, the heat from the Earth itself would balance the heat lost to space. But this heat from the Earth would still be insulated by whatever atmosphere is present, so I posit the lower atmosphere would certainly be warmer. I would accept that there would be huge convection currents in the air, as the cold air at the top freezes and drops, but the earth heats it up and it rises, just like it currently does.

Add in volcanoes, thermal vents, and such, along with all of the heat that us humans would put into the atmosphere from trying to stay warm, and trying to keep our manufacturing base going, it leaves us a very long time, perhaps a hundred thousand years or so, before the Earth reached a point where at the surface all of the gasses'froze' out of the atmosphere. And even at that, the oceans would still have liquid water below the thick insulating layer of ice for hundreds of thousands of years. That is a very long time for humans to stick around.

Ice is a poor conductor of heat, so the heat from the water would have to travel completely through the ice layer before it was given off. But here is the thing - it is my understanding that until ALL the water froze, the ice temperature would not drop below freezing. A bucket of water left outside in the winter will stay near the freezing point until substantially ALL the water is turned to ice. If this happened above our oceans, it seems to me that this would keep the lower levels of the atmosphere above the oceans at least around the freezing point. Areas above the land could certainnly get much colder, but enough for the entire atmosphere to freeze?

The primary way that heat would be lost from the Earth is radiation (there is no conduction into space), and most of the heat currently given off by the Earth is reflected solar radiation, from the sun. Absent this reflected solar radiation, what is left to radiate heat from the Earth? I can not understand why the earth would cool down substabtially at all, except by radiation over a very long time. Where would the heat in the atmosphere go, and more omportantly, how would it go? It seems to me it would be the 'heat dissipation from a spaceship' problem on steroids. If a spaceship can not 'cool down' and rid itself from all the heat produced by its weapons, except by radiation, how would the Earth quickly cool down, except by ejecting material into space?

I have found no credible science-based theory or evidence that indicates the Earth would quickly cool down, below -100 degrees or so, as it does on a very, very cold winter day. In fact, there is lots of scientific evidence that the earth would NOT rapidly cool down beyond a really cold winter day. There is no large scale way for it to do so. It would not 'warm up' like it does in a sunny day, but why would it cool down below its inherent residual temperature? I am not saying the atmospherre would NEVER cool down sufficiently to 'freeze out' the atmosphere, I am saying that it would take a very long time, measured at least in hundreds of thousands of years, for the atmosphere to reach a temperature where oxygen would turn liquid.

Yes, we would lose our food supply that comes directly from the sun, but I can not see why the temperature of the earth would quickly descend into levels that we could not compensate for, by supplying enough nuclear energy, or even fossil fuels. We have hundreds of years of coal resources, for instance, since we no longer have to worry about greenhouse gases. And electrolysis of water on a mass scale could replenish our oxygen.

As for our food disappearing? Maybe many living things would freeze, but the absence of the sun would not immediately rid the Earth from all of the organic material. It might not be replenished, but there would be lots of dead organic material around for scavangers like cockroaches and rodents to survive. They do not need the sun at all, just a lot of organic matter - even if it has been dead for hundreds of years. And cockroaches are quite nutritious. Lots of protein. Squeamish societies like prissy Americans might not have a food source, but there are many cultures that would live quite well on whatever food sources remained, even without photosynthesis. This is not like a long distance space journey, with limited organic material available, this is the Earth, with an abundace of organic material already laid down. Millions of years of organic material, in fact.

And it would not be completely dark, either. It is estimated that light from the stars would be about 1/300ths of moonlight, but there would still be light. Enough, in fact, for night scopes to still work. Cats would be fine. They could even survive on all of the scavanger mice that would still be around.

It would not be all doom and gloom.

  • $\begingroup$ it takes millions of years for the planet to reach -400 degrees, true but it only takes 10-20 years for the planet to reach the dew point of air. It only takes a few months before the surface is cold enough most steel becomes too brittle to use, it takes about a year before all steel on the surface is brittle which makes mining incredibly difficult. $\endgroup$
    – John
    Sep 27, 2020 at 4:26
  • $\begingroup$ @John You still don't get it. The center of the earth is very, very hot. It will take billions of years for it to cool down, even without the sun. Below a hundred feet, the Earth is warmed by its interior, not the sun. Salt mines at 70 degrees F will stay 70 degrees F, even without the Sun. Even subway tunnels have to be cooled, because it is so hot from the heat of the earth. Mining will be like it is now - incredibly hot. That isn't ever going to change. Even today, cooling things down in office buildings and manufacturing plants, even in the dead of winter, takes more energy than heating. $\endgroup$ Sep 27, 2020 at 21:44
  • $\begingroup$ Blast furnaces reach temperatures of 3,000 degrees F. It is completely within our technology today to keep our buidings heated, even at -100 degrees F. There is absolutely nothing that prevents us from keeping hundreds of millions of people warm, if not billions. Heat, and the temperature, is not a problem, just an inconvenience. It's all about setting priorities. Bill Gates, and all of his family, and all of his heirs, will be alive for generations and generations, there is not even the slightest controversy over that. Biomeat, artificially grown in a vat, needs no sunlight. $\endgroup$ Sep 27, 2020 at 21:53
  • $\begingroup$ buildings aren't the issue, you have to heat the ore to mine it and the mining equipment for it to not shatter and to keep the fuel from freezing, and you have only a few months to build it all, a few months in which a global war of unprecedented scale will be waged. Can you keep a few alive, sure, but they are on borrowed time from the get go, because they can't keep enough technology running to build, and feed, and repair everything else they need. keeping people warm is nothing keeping all your technology working is the hard part. you have to build an entirely new infrastructure. $\endgroup$
    – John
    Sep 27, 2020 at 22:03
  • $\begingroup$ @John You present like you have a pessimistic, no-hope viewpoint of your fellow countrymen, like you have no faith in your country, or your government, or the ingenuity of your professionals, your engineers, or your scientists and physicists. Dispite what you claim, we currently have the knowledge, the expertise, the leadership (in some countries), the scientific and engineering achievement, and even the current massive infrastructure upgrading that will allow some countries to survive. Some have alredy given uo any hope for the future as it is, and have already degenerated into anarchy, $\endgroup$ Sep 27, 2020 at 22:15

I agree with Michael Borgwardt's answer: there is a very small chance a few small clusters of humans will survive in places with high geothermal activity, specially if there are already geothermal power plants nearby. As he said, nuclear energy is an option, but I think the hurdles with nuclear fuel (processing, disposal, handling, etc) will make it a worse in the long run.

But suppose mankind is able to survive in sustainable society for a long time. In the long run, hope is not all lost for the planet: far from the violent radiation bombardment of the sun, Earth's atmosphere will eventually grow much thicker (both from capture of interstellar gas or from volcanic emissions) which will make the greenhouse effects many times more intense. Maybe even enough that the meager 47 TW of Earth's internal heat energy that flows to the surface will be able to sustain temperatures high enough for water to be liquid in some places.

With unfrozen atmosphere, the range of isolated human communities greatly increases, which might make nuclear energy and fossil fuels feasible again.

  • $\begingroup$ so little gas will be captured the effect is more or less irrelevant. $\endgroup$
    – John
    Sep 23, 2020 at 15:56
  • 1
    $\begingroup$ Without a sun, what greenhouse effect could there be that would make any difference? $\endgroup$
    – rek
    Sep 24, 2020 at 2:44
  • $\begingroup$ @rek Earth has its own internal heat. $\endgroup$
    – lvella
    Sep 24, 2020 at 13:40
  • $\begingroup$ @lvella Emphasis on internal. $\endgroup$
    – rek
    Sep 24, 2020 at 14:13
  • $\begingroup$ As I said, it is 47 TW of heat that reaches the surface. $\endgroup$
    – lvella
    Sep 24, 2020 at 14:16

The obvious thing to do is to position the moon (which we still have) at the right distance, and persuade it to function as a star. (This would give us something to orbit around, which I presume helps, but that would require a particular combination of distance and orbital speed.) I have no idea whether or not this is feasible. I am hoping other people can fill in that gap. I do not know whether to make this an answer or a comment; here it is. Some admin person might like to move it or whatever.

Separately… As I understand it, bodies in space give off all the heat they receive, once they have reached an equilibrium state. Arguably then, all we need is a Dyson Sphere? (I believe they are on special at the moment.)

  • $\begingroup$ We already orbit around the moon. We just don't orbit as much because the earth is bigger. Also, don't forget: the Dyson sphere would emit blackbody radiation not just back at earth, but outward into space as well. So, no, the whole scheme would still be constantly losing heat to the void of space. Slower, yes, since there'd effectively be a vacuum-sealed layer of insulation around the planet. $\endgroup$
    – Kevin
    Sep 25, 2020 at 14:01
  • $\begingroup$ You can't make the moon into a star, at least not without throwing a star's mass of hydrogen at it - and where would that come from? $\endgroup$ Sep 25, 2020 at 19:15

We all die.

There is no easy way to say it. Discussion of underground habitats and power requirements are moot... we don't know how to make closed environment systems which actually work. Even with planning and preparation and decent conditions, our best sealed biosphere experiments have lasted months, not years. See https://en.wikipedia.org/wiki/Biosphere_2

So, questions of long term power and heat retention are irrelevant. The biosphere will collapse and we will go extinct due to starvation, not freezing.


Unfortunately, within seconds of having no sun, the earth would:

  • Be flung into the universe ( with nothing to hold it or the solar system in orbit, the earth would hurdle into space at around 66 - 67,000 mph,
  • Freeze (Every day that passes would have a drop in temperature. I believe four to five days would bring us to -273 degrees Celsius or absolute zero (this is where even the molecules of an object cease to move)
  • Become absolutely dark ( Absolute dark would be darker than your eyes shut in a dark room under a deep cave lol)
  • and I believe the tectonic plates would shift dramatically causing volcanic eruptions, earthquakes and mega tsunamis world wide (very scary to think about all that happening in the dark.)

It is a cool premise, but unless the sun was replaced with another star, we would be dead. Also, even one second of our suns disappearance would cause our planet to react in negative ways. Hope this sheds some light.

  • 1
    $\begingroup$ For what it's worth, volcanic activity would provide isolated sources of heat and light, at least for short periods of time... $\endgroup$
    – Doktor J
    Sep 24, 2020 at 17:43
  • 2
    $\begingroup$ The sun provides 4.3x10^20 joules per hour. Water has a specific heat of 4.18 joules per Gram x DegreeC. There are 1.4x10^24 grams of water in the oceans. Which means, the sun disappearing would only cool the oceans roughly a ten-thousandth of a degree per hour. And that ignores Enthalpy, and is just the Oceans - which is only about a thousandth of the total mass of earth. $\endgroup$
    – Kevin
    Sep 25, 2020 at 14:13
  • $\begingroup$ I'm not sure what the significance of the first point is - motion is relative, does it even matter how "fast" the planet is moving if it isn't near any other massive objects? You're seriously overestimating how fast the planet would cool down, and it would never actually reach absolute zero (even if you ignore geothermal heating and such, the ambient temperature in deep space is ~3 K). Darkness is manageable as long as we have functioning electronics, so that isn't an issue unless we've already hit other likely insurmountable hurdles. And I'm curious if you have a source for the last point. $\endgroup$ Sep 25, 2020 at 19:28
  • $\begingroup$ Here is ONE article on the hypothetical situation for the commentators. Just type in the title of the article below into your google search engine. TECHNOLOGY If The Sun Went Out, How Long Would Life On Earth Survive? PopSci provides chilling answers to your burning questions By Holly Otterbein October 20, 2008 If The Sun Dies NASA $\endgroup$
    – A.T. Coll
    Sep 26, 2020 at 4:39
  • $\begingroup$ Source for the last point is that big white ball that orbits earth everyday. $\endgroup$
    – A.T. Coll
    Sep 26, 2020 at 4:54

You must log in to answer this question.

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