Imagine that one day in the not-too-distant future, our scientists discover that our world is dying. It doesn't matter how this is happening, maybe magic or unstable core or Gaia herself has finally had enough of our sh*t, but the important thing is that it is happening, and soon.

Our only hope is to escape, and head out to space. One tiny solace is that the Earth isn't going to explode, just become uninhabitable, so we won't have to travel far.

With a time frame of around 50 years, assuming that everyone on the planet managed to pull their heads out of their asses and work together for a change, what would be the best method to save the most people, and how many people would we reasonably be able to save? What if we only had 30 years?

Harder science preferred. If your answer calls for needing more time, state how much extra time would be needed.

Edit: Tech level set 20 minutes in the future. I think I may not have been clear enough.

Edit: Thanks for all the great answers guys! Project ORION is clearly the way to go. I would accept more answers if I could, but Jimmy360 was the first to provide the answer. Thanks :D

  • 10
    $\begingroup$ How much inhabitable? If we're talking about a barren desert, it would still be far more practical to try to stay on earth than say, go to the moon or to mars which is a barren desert, but without the atmostphere and several millions of kilometers away. It would have to be so harsh that we couldn't be saved not even by burrowing underground before going into space would start looking like a good option. $\endgroup$
    – Neil
    May 25, 2015 at 8:07
  • 5
    $\begingroup$ @FeaurieVladskovitz "For the sake of argument" inhabitable then? Pretty sure being overrun by xenomorphs is still a more practical option, believe it or not. $\endgroup$
    – Neil
    May 25, 2015 at 8:21
  • 1
    $\begingroup$ By "everyone on the planet managed to pull their heads out of their asses", do you imply that we don't have a lot of people saying it isn't happening and that we can't spend all that money without knowing that without spending it over these decades, what's suggested will actually happen? $\endgroup$
    – user
    May 25, 2015 at 11:16
  • 2
    $\begingroup$ Related: Why Not Space? on the Do The Math blog, written by an associate professor of physics at the University of California San Diego. $\endgroup$
    – user
    May 25, 2015 at 11:19
  • 1
    $\begingroup$ While it's not directly related, here's a cool documentary similar to this,. It talks about how we could escape Earth with a neutron star headed straight at us, and 70 years to get out. $\endgroup$ May 25, 2015 at 18:19

9 Answers 9


Nuclear Pulse Propulsion Rockets

NPP rockets were (still?) actually developed by the U.S. government. They called it Project Orion. Simply, the design is to detonate a nuclear bomb with a nuclear bomb underneath it. One would expect the shield underneath it to melt or be destroyed also with the rocket, but the rocket gets away so quickly that it is safe.

There are 7.13 billion people on Earth and the average human weighs 62 kg. The minimum amount of weight that we have to lift is 441440000000 kg. It takes about 350703 joules to lift a single kg to geosynchronous orbit. However, one of the problems with rockets is that they have to carry their own fuel making it exponentially harder to lift things with a rocket... but this one doesn't, meaning that we can avoid the tyranny of the rocket equation. 350703 x 441440000000 = 154814330000000000. We need at least 154814330000 MJ to lift everyone to space.

The Castle/Bravo, detonated by the U.S., released 63,000 TJ of energy. This is over 6x the amount of energy that we need. This is also enough energy to bring equipment (like terraforming equipment).

Edit (a counter to Jim2b's answer):

In this situation, something I like to call emergency Communism would come into play. The unified Earth government would cut off all unnecessary business/production and force everything to work for the Orion goal. The world's steel production would be massively increased.

  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$ May 28, 2015 at 2:12

If the Earth is dying, then a revived ORION is the way to go. Polluting the atmosphere with fallout is going to be the least of everyone's worries. Calculation made by the ORION team in the late 1950's suggested they could have gone to Mars in the late 1960's, and Saturn by 1975. Their spaceships would have resembled Winnebago's rather than the tin can's we remember from history; ORION is so efficient in terms of both ISP and deltaV that ORION team members jokingly suggested they could bring barber chairs on board if they wanted.

The NextBigFuture blog has also been rather enthusiastic about the so called "Jules Verne Cannon", which involves firing a nuclear "physics package" in an underground salt dome and channeling the blast through a large pipe to boost large and insensitive payloads into space. One suggested example is coal, so you have carbon to carry out various chemical reactions on the Moon, which suggests just how cheap this could potentially be. That idea was inspired by the real life "Plum Bob" series of tests, where one underground explosion popped the cap from the top of the shaft. Calculations suggest the huge steel cap exited the shaft at 6X Earth escape velocity, although no verified records (the cap only appears in 1 frame of a high speed movie recording the event) exist and the cap itself has never been found (it is most likely it disintegrated inside Earth's atmosphere due to aerodynamic stress and heating).

Obviously, payloads being launched by a Jules Verne Cannon need to be very rugged indeed.

So the basic escape route would be to use a Jules Verne Launcher to fling payloads of heavy, unbreakable "stuff" into orbit or even blast it into the Moon (future astronauts can "mine" the new craters for steel, other metals and minerals) while sending the actual astronauts into space in large Orion craft, which have enough deltaV to pick up payloads in orbit and carry on the far reaches of the Solar System. By lofting large amounts of basic materials via the Jules Vern launcher, even relatively inefficient recycling systems can be made to last for years while better life support loops are designed and built, and new sources of materials from the asteroids and moons of the Solar System are developed.

  • $\begingroup$ The Orion part is mostly the same, but the Jules Verne Cannon is a new and quite important addition, since it allows (future) us to make the main ships smaller and use less propellant, since raw materials don't have to be kept in cargo bays. $\endgroup$
    – zovits
    May 26, 2015 at 8:33

This is amplification on previous answers citing Project Orion (aka Nuclear Pulse Propulsion). Read the provided reference for history and technical background.

Research & testing performed from 1950s - 1970s indicated that using that level of technology we could built a 8,000,000 tonne craft capable of achieving orbit and providing a bit of extra $ \Delta V$, perhaps enough for $V_{esc}$. However, only about 1/3 of this mass is payload mass (2,700,000 tonne).

The problem
Let's assume we need to lift every human being off the Earth and there's zero population growth:

Current population ~ 7,000,000,000 each  
Average mass ~ 100 kg / each  
Total mass ~ 700,000,000,000 kg  
Total tonnes ~ 700,000,000 tonnes  

Assume that we need 20x this mass for equipment & consumables for keeping people alive in transit, colony construction equipment, plus sundry other items. Also assume that the ships are constructed in a modular prefab fashion that allows us to directly use them at the destination or dissemble them and use the parts in existing colonies:

Total lift requirement in tonnes ~ 14,000,000,000 tonnes
Per craft payload mass in tonnes ~      2,700,000 tonnes / craft
Total required craft             ~           5200 craft

Estimating timing
Wikipedia states that world production of container ships was ~11,000,000 tons in 2011 and our Super Orions would require similar (but more stringent) levels of construction difficulties (remember we can use normal construction materials like steel). So without straining we could build on average 1 of these Super Orions per year but each would require multi-year construction (say 5 years like US aircraft carriers or perhaps even 10-15 years). We'd be done in 5,200 years.

With straining, (I would guess) we could build 10x this number. We'd be done in 520 years.

With all out desperation, the upper bounds of what we could make would be determined by our critical resource production (such as steel). Assume our 8,000,000 tonne craft are composed entirely of steel and that this is our limiting resource. This site indicates world steel production is around:

World steel production ~ 150,000,000 tonnes / year
Max Super Orion production ~ 20 ships / year

We'd finish making our Super Orion fleet in 260 years.

Estimating cost
Let's assume that the amount of labor and difficulty of constructing these Super Orions equates on a tonne per tonne basis with the expense of building nuclear aircraft carriers.

US nuclear powered aircraft carriers mass about 100,000 tonnes of displacement and cost 26 billion (USD).

Each Super Orion will cost ~2.1 trillion (USD) or about 2/3 of the 2014 US federal expenditures. The fleet will cost 11,000 trillion (USD).

The Earth is the garden spot of our Solar System. Very few places have both the readily available volatiles, metals, and other materials (soil?) that we'll need to survive. IMO, Mars and perhaps a few of the main body asteroids (Ceres is looking really good right now) might fulfill this role.

But really only Mars has enough volatiles, metals, and room to host a significant portion of the human population. Where are we going to put everyone?

  • $\begingroup$ You can't divert all of the world's steel production to this project. Steel is needed to mine iron, to transport the iron ore and other products from mine to mill, and to transport the steel. Steel is needed to produce and transport food (people need to eat), transport petroleum (people need to get to work),and to make cars, trucks, railcars, railways, boats, and buildings. Your basic answer of 5200 years is short. That said, this is a much better answer than the other answer that proposes to use nuclear propulsion. $\endgroup$ May 25, 2015 at 17:15
  • $\begingroup$ @DavidHammen, I just used that as an upper bounds. In such a scenario, I imagine we'd increase our steel production significantly but could only divert this much to ship construction while the increased production went to maintain the necessary infrastructure. We'd probably switch to a rationing system to reduce consumption of resources critical to ship construction. Plus the reality is our ships are 1/3 propulsion (mostly pusher plate), 1/3 payload, & 1/3 all else (mostly structure & radiation shielding). A significant portion of our payload will be consumables like water. $\endgroup$
    – Jim2B
    May 25, 2015 at 20:43
  • $\begingroup$ And you can't bring them down again and load them up???? $\endgroup$ May 25, 2015 at 22:18
  • $\begingroup$ @LorenPechtel, I've never seen any research that discussed landing with these. I can foresee multiple problems with the maneuver too. I'm not saying we couldn't do it, but I do think it'd be terribly difficult and maybe/probably not worth the effort. Perhaps we'd build them along a modular plan which would allow us to dissemble them at their destinations and use them as prefab colony construction. It'd also mean that destinations like Mars are probably out. Which is a shame because I think only Mars has a hope of supporting billions of people. $\endgroup$
    – Jim2B
    May 26, 2015 at 1:00
  • $\begingroup$ @Jim2B Orion was originally envisioned for manned missions. You think they planned re-entry capsules to come down with? Land over water so you don't need as exact a landing as SpaceX needs and the water can wash away most of the debris from the bombs and isn't destroyed by the last detonations. You're dunking brute force engineering rather than an intricate rocket. $\endgroup$ May 26, 2015 at 2:52

If we had 50 years, we'd be boned. With the most optimistic estimates of fuels costs; SpaceX can put things into Low-Earth Orbit for $1,600/kg.

The average person weighs about 62kg.

This becomes an enormous problem when you try to ship all 7.13 billion people into space.

That's going to cost 704 million million dollars ($704,320,000,000,000).

Now you have to find the spare cash to build enough spaceships for seven billion souls.

At the end of the day, with anything close-to today's technology - it's just not practical to save most people. A lot of people are going to be left behind. We can't even keep a few astronauts in space indefinitely. We manage a few years at best.

But in general, how many people we could potentially save is dependent on how advanced our technology has progressed - the further along we are, the more we can do.

Exactly how far? That question is far too broad because every person can select their own subjective answer.

  • $\begingroup$ Sorry about this, I don't think I was clear enough, but I'm cool with the tech being slightly more advanced than what we currently have. $\endgroup$ May 25, 2015 at 6:43
  • 1
    $\begingroup$ @FeaurieVladskovitz, updated answer. The crux of it is; slightly more advanced isn't going to cut it. Needs to be significantly at the very minimum. $\endgroup$
    – user6511
    May 25, 2015 at 7:06
  • 1
    $\begingroup$ I'm not sure where Wikipedia gets the \$1,600/kg figure from; at \$90M to lift 53,000 kg to LEO, the closest I can find is the Falcon Heavy costing a shade under \$1,700/kg (to LEO) maxed out. Regardless, this answer disregards a pretty major factor in looking at only the biomass, whereas to survive you need a lot more. Food, oxygen, heating, ... all mean that only a fraction of your rocket's payload capacity can be used to actually lift humans. Which means that in practice, the situation is significantly worse than laid out above. $\endgroup$
    – user
    May 25, 2015 at 11:11
  • 1
    $\begingroup$ The gross world product is around $75trillion per year. Over 50 years that's ~4 million million dollars. I think this rounds out the answer as to why the cost is a limiting factor. Even if the entire global production was magically focused on space, it looks like we could save only 0.5% of humanity. But for a scifi story this actually gives you more than enough people to start a space civilization. $\endgroup$ May 25, 2015 at 13:09
  • 1
    $\begingroup$ I believe this answer is correct if we limited ourselves to only chemical propulsion schemes, however, as other answers have pointed out, we don't need to restrict ourselves. We can using nuclear which possesses >1,000,000x as much specific energy as fuel. Plus the NPP type craft can afford to use heavy, bulk, vastly cheaper construction in the ships (it actually smooths out the ride) than the chemical propelled craft can. $\endgroup$
    – Jim2B
    May 25, 2015 at 16:31

There is a "documentary" called Escaping Earth that explores this scenario. And while I disagree with some of the options presented (ORION is IMO impractical and the whole "artificial ecosystem in cylinder" is quite stupid), it does make few points:

  • It is impossible to save every human, but it is possible to save human species and maybe some other species
  • It would take combined effort of whole world to save few thousand chosen people
  • There would be people against it

I believe all issues we have right now with space colonization and exploration are related to money. We already have technology to create a colony on Mars, but there is huge pressure to minimize expenses, so it is quite limiting on amount of stuff we can send along with colonizers. But what if we could send a 1000 ton spaceship to Mars every month? In a few years, there would be enough material there that self-sufficiency could be achieve. Yeah, it would be expensive, but not impossible. We just need to motivate our politician to invest into space colonization instead of military and big corporations to make rockets instead of consumer gadgets.

  • 1
    $\begingroup$ "But what if we could send a 1000 ton spaceship to Mars every month?" Well, we can't. But yes, if money wasn't a concern and people were actually working together, we certainly could send a lot of stuff to Mars. $\endgroup$
    – user
    May 25, 2015 at 11:13
  • 1
    $\begingroup$ Why do you feel NPP (Orion) is impractical? It'd be uncomfortable, yes. But most people would prefer discomfort to death... $\endgroup$
    – Jim2B
    May 25, 2015 at 16:32
  • $\begingroup$ @Jim2B I would say worst problem is creating mechanical dampening system, that is able to withstand 1000s of blasts without need for repair or maintenance. IMO creating engine consisting of hundreds smaller ion engines powered by fusion or antimatter would be much more safer and maintainable solution. $\endgroup$
    – Euphoric
    May 25, 2015 at 18:28
  • $\begingroup$ @Euphoric, those concepts have great $I_{sp}$ but insufficient Thrust to Weight ratio. You need both to do what needs to be done in this case and only nuclear has a sufficient "bang for the buck" (or weight in this case). Fusion might do it for us someday but we don't have that yet - so nuclear bombs are the only practical way to do it right now. $\endgroup$
    – Jim2B
    May 25, 2015 at 20:37
  • $\begingroup$ But I agree, the mechanical damping and radiation (esp, $ \gamma $ back scatter) are real problems that would need to be solved. We do, however, have a reason to be very optimistic that we could solve both problems. We understand them and know how to do it in principal we just don't have working models of the necessary systems yet. $\endgroup$
    – Jim2B
    May 25, 2015 at 20:39

If the Earth is going to be uninhabitable anyway, you can use ORION-type spaceships to get most valuable people (this is usually a great plot device) out. Maybe you can help some of the rest to get out if you use some of the 50 years to build a space elevator.

I think it would be impossible to get everybody out, even if you had more time. I reckon the most likely scenario would be to take "valuable people" (define that as you will for your story), maybe useful animals and plants (or maybe just DNA samples) and after that, anyone/anything else that time andmoney allow for. It won't be nice, it won't be pretty and more likely the people left behind will riot and try to stop you...

  • $\begingroup$ Half the first volume can be about deciding who goes. Can you buy a ticket for your family by donating enough billion dollars? $\endgroup$
    – Zither13
    May 25, 2015 at 16:56
  • 1
    $\begingroup$ How about this? Supposedly, only people under 40 would get a chance to go (in a lottery). But our heroes discover thta the super-rich you can buy a ticket for 500 million euros... $\endgroup$
    – Alex San
    May 25, 2015 at 18:34
  • $\begingroup$ @AlexSan: Would you really use a lottery? Or would you rather carefully analyze what skillsets (and genetic mix) you need and pick those? $\endgroup$ May 25, 2015 at 18:58
  • $\begingroup$ @MatthieuM. Yeah, lottery is stupid. But people that don't have any appreciable skills aren't going to be very happy when you declare that only the best and brightest are going to make it. Even today, many "stupid" people dislike "smart" people - add being "smart" as a criterion to survive, and you'll have massive rebellions on hand. The best would probably be some combination - get enough people to ensure the survival of the civilization and technology, and let the rest have a lottery. Since lottery is (stupidly) perceived as fair, it might very well help the PR enough :) $\endgroup$
    – Luaan
    May 26, 2015 at 8:41
  • 1
    $\begingroup$ @Luaan: Or rig the lottery? (might be detected though, so your option might be best). I was thinking you might get a rebellion anyway when announcing the lottery results, maybe delaying the announce each time to right before take-off would preserve hope and prevent the rebellion. $\endgroup$ May 26, 2015 at 8:57

Right now in order to keep astronauts in space they require a LOT of infrastructure on Earth, all the food and other supplies come from it. It is theoretically possible to grow plants in space, but the amount of land required even for a single family to subsist on is not trivial. Finally, technology breaks occasionally and requires repairs and spare parts, which also come from Earth.

In 50 years it should be possible to setup some long term spaceships with a lot of spare parts and space greenhouses, but they'll provide for a few hundred at best, not sure entire planet will work for the tiny few to survive instead of looting and other fun stuff like that. And even then it won't be sustainable indefinitely, they won't be able to acquire resources or develop or grow, and after a decade or two things on the ship will start to break, their bones will become brittle and so on.

Space presents enormous challenges and for every person staying up there requires thousands working down here to make it happen, creating even a tiny sustainable, self sufficient facility up there in just 50 years is pretty unlikely. Being able to put 7 billion in space and have life support for them would probably require a thousand years if it's even possible.


The main limiting factor is time and to a lesser extent money. If the human race is at risk then a temporary solution could be made that meant money was irrelevant, or at least credit could be extended to defer the problem.

I would advocate using the Moon as a temporary base, to buy time for a longer solution to be found.

We already have the technology to get to the Moon and back and all we would need to do is build something similar but on a much bigger scale. Rockets could take off on a regular basis to make the trip and a Moon base could be built to house the ever increasing number of people arriving. The base would include facilities to turn around the rockets so they could be re-used many times to bring more and more people with more and more rockets being built on Earth so as time went by you would have thousands of rockets bringing in hundreds of people in each trip.

Heavy resources that would be required including water (ice) and metals etc could be sent using the Jules Verne Cannon.

Assuming each rocket could carry 1,000 people you would need 70,000,000 trips to evacuate everyone. Five thousand rockets, each taking a week for the round trip would therefore be able to evacuate 5,000,000 weekly, 250,000,000 annually. It would take 28 years to get everyone to the Moon.

The figures are arbitrary of course, building a rocket capable of carrying 1,000 people is not currently possible but equally the 7 day time scale for the round trip is probably too long.

Once established on the Moon a longer term solution could be found such as moving to Mars etc.

Conditions on the Moon would be very crowded so living quarters, factories etc would be constructed under the surface.

If the numbers didn't stack up then you could force a policy of not allowing people to have children, or 1 child per family as they do in China, this would reduce the figure from 7 billion somewhat over the time periods involved. You could also have something along the lines of Logan's Run where people over a certain age are left behind.


To save the human race does not mean saving EVERY single human. Maybe high government officials and the rich people get to buy tickets to automatically be able to board the trip to the other planet, but the poor people which cannot afford a ticket will be either just not taken with or subjected to a lottery process. Or sort of a lottery where the more you pay, the bigger is your chance of winning a ticket. A person who wins the lottery is allowed to take his family with him if the family is not too big.

Also, I would recommend that you construct several giant ships in orbit so that no fuel would be wasted for the ascent into space. You will bring the people that are to be saved to the ship using spaceplanes; the same applies for landing them on Mars when you arrive there.

Also, you can send expeditions to Mars (this is the planet which I recommend) that do reconnaisance and find good sites for bases before setting up the modules (you can use robots too) and the agricultural domes. Before the main wave of people arrives, you will send crops and livestock to Mars.


You must log in to answer this question.

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