What’s more feasible: to build a self-sustaining colony on Mars or as a self-sustaining space station? The 1976-book The High Frontier advocates the latter while Elon Musk seems pretty confident that Mars is the better option. Why? Are they taking other factors into account? Me, I’m not asking where it would be nicer to have a station but where it’s easier to get it done.

Things in favour of a station on Mars:

  • You already have some gravity.
  • Less radiation (0.7 vs 1.8 mSv/day)
  • Some sort of raw materials on your front door
  • Lots of space to put stuffs without it floating away.

Things in favour of a space station, say a design similar to an updated O'Neil cylinder like the Kalpana One (or is there something more up-to-date?) in GEO or near-earth space:

  • Closer to Earth. I guess if you want to reach Mars within closer to three than the full six months, you’d need a lot more propellant than to get to the space station. (I can find many different Delta-V estimates, probably depending on lots of parameters like how much you aerobrake and what constellation the planets are in. And yes, I'm aware that Delta-V to dollars is exponential.)
  • Little Delta-V needed to get to and fro asteroids, which should make it easier to mine them. You can then cherry-pick (to a certain extent) raw materials from asteroids.
  • A lot more solar power (no atmosphere, no sandstorms, no/less night)
  • Some processes are probably actually easier in 0g.
  • Seems like no heating needed (or is that a minus because it’s hard to get rid of the excess heat?)

Any more factors? How would you weigh them in terms of feasibility?

Edit: More objectively, what do experts think and how do they reach that conclusion? Has our knowledge/assumptions changed between O'Neill's and Musk's approaches or do they just have different opinions?

Edit2: With "self-sustaining" I mean able to sustain itself without major imports of materials from earth which would make the endeavour at current launching costs very expensive. Importing small masses of specialised equipment (like computers and instruments) from earth is allowed, as well as obviously whatever you need to get it all going. More importantly, regularly importing raw materials from asteroids (and/or the Martian surface for the Martian colony) is allowed—for the purpose of this question, I consider those facilities part of the colony.

  • 6
    $\begingroup$ Neither is feasible, we are so far away from having the technology for a self-sustaining colony on another planet or in orbit that there's no way to answer this question. $\endgroup$
    – GdD
    Jun 10, 2014 at 12:15
  • 1
    $\begingroup$ @GdB well, that hasn't stopped people from making educated guesses. I updated the question. $\endgroup$
    – mb21
    Jun 10, 2014 at 12:42
  • 4
    $\begingroup$ This question assumes a self-sustaining colony is feasible. It isn't. We can't even do it on the Earth's surface. Every attempt has died after a short period of time. We don't know how to do it. $\endgroup$ Jun 11, 2014 at 12:58
  • 1
    $\begingroup$ @mb21: They are political, which is 'psychological' on a macro scale. Humanity has no actual 'need' (as per Maslow's hierarchy) to explore or settle Mars. It just wants to. Scientific exploration of Mars for the purpose of understanding Mars' geo-history and current conditions if done by humans, will likely take much less time and less aggregate cost than unmanned approaches discussed so far. The trick is to have it meet with the value systems of those paying the bills for it. That is where the psychology comes in. Beyond science? There is no commercial potential unique to Mars. $\endgroup$ Jun 11, 2014 at 19:07
  • 1
    $\begingroup$ To more directly answer the above question, Elon Musk wants to make money by creating a permanent customer for his rockets, crew capsules etc. Mars is ideal as NASA has historically considered Mars an ultimate objective. $\endgroup$ Jun 11, 2014 at 19:11

3 Answers 3


There is one central question with creating a colony somewhere uninhabited: resources.

Material resources

With a ship-based colony, the material resource is just whatever you put up there, and then whatever solar power you can collect. In order to be self-sustaining, you therefore need to be able to create a closed system which requires no further input. At present, we are very close to that goal technologically, and I think with the addition of some small amount of support from Earth, a colony is within our capacity now.

On Mars (or the Moon), there is an available resource under your feet and above the dome roof, so the limitation is really our technological capacity to reprocess the atmosphere and rock. Thankfully, the atmosphere does contain a small amount of oxygen and the rocks are similar to Earth rocks, so we are quite capable of extracting raw materials from Mars itself. Even if we weren't, we could fall back on a hermetic environment like that of the ship-based colony, so it is at least no worse than the ship-based colony.

Costs of the initial mission

If it is technologically possible, what is the cost to Earth-inhabitants of either path?

A Mars mission to establish an entire colony (not just to send 1 or 2 astronauts) would be more expensive by quite a margin. The equipment and resources for a Mars colony would be no less massive than for an orbital one, and then we are subjected to the tyranny of the rocket equation for the fuel to make the journey. There is a difficult balance here; the slower and more efficient the journey (more gravitational slingshots, using the "interplanetary highways" etc), the greater the material required to support life during the flight. However, the Mars mission could be done much more slowly, by establishing a small base and slowly expanding it using largely local resources.

By comparison, having only succeeded in getting a few small lumps of metal to Mars, we have put a whole space station in orbit. The costs are far lower because the delta-V requirement is lower. It is conceivable that a station could be constructed and populated near Earth and then shifted to a Lagrange point via ion drives and so on, which are largely solar powered, further reducing the mass to be lifted. Inflatable habitations seem a particularly appropriate option, permitting a large structure to require minimal volume and mass on the rocket.

Time to independence from Earth resources

The space habitat wins here. A Mars mission would require a lot more preparation, and a lot more time to build up from the initial payload to a working environment. It would be capable of expansion, unlike the habitat, but it would take much longer.

Potential inhabitants (and thus sponsors)

The space habitat wins here again. An Earth-orbit habitat or a Lagrange point habitat would both be reachable from Earth in days, so a rich family could take a holiday there and older people could retire there (consider the benefits of lower gravity environment for the elderly). Mars is more of an undiscovered country; America was populated largely by the poor because the journey was so long and promise was more of opportunity than quality of life. Yet voyagers to Mars would need to be well-funded by governments, so it would not be within the means of ordinary people's life savings.


I would illustrate the difference as being a choice between creating a holiday resort on an island in the sea versus creating a colony at the South pole. Despite the limits of the island, the tradeoff of proximity, cost and timing would make it both a better initial proposition and a better long-term bet.

One day we will populate Mars, but I suspect it will not be an expanding colony until there are several habitats near Earth.

  • 3
    $\begingroup$ Re At present, we are very close to that goal technologically No, we aren't. Every attempt has failed. What we have learned is that life is messy and incredibly interconnected. Rocket scientists et al use math all the time thanks to the unreasonable effectiveness of mathematics in the natural sciences. We technicians blithely ignore the unreasonable ineffectiveness of mathematics in biology. $\endgroup$ Jun 11, 2014 at 13:04
  • $\begingroup$ @DavidHammen I agree with you that physics and biology are two different pairs of shoes. But it seems we wouldn't have to create an entire biosphere: hydroponics for food and algae for oxygen should get us quite far. $\endgroup$
    – mb21
    Jun 11, 2014 at 20:22
  • $\begingroup$ @PhilH Thanks for your answer, will accept it if there are not more coming in :) I was also leaning towards the space station, that's why I was surprised to hear that Elon Musk, who usually does the proper math, is so focussed on Mars... $\endgroup$
    – mb21
    Jun 11, 2014 at 20:26
  • $\begingroup$ @mb21 - Name one fully closed system that converts your bodily wastes, the unused food parts, the dead plants, etc., into a fully-balanced set of water-soluble fertilizers for the plants in the hydroponics system. You can't because it's never been done. Just as a warp engine is pure sci-fi right now, so is a closed biological system. (Some biologists have told me that this closed biological system is even more sci-fi than a warp engine.) $\endgroup$ Jun 11, 2014 at 20:33
  • 1
    $\begingroup$ @DavidHammen: a habitat in space is supposed to be more like a chunk of the Earth biosphere than a clever hydroponics system. It is really a step away from a small island; largely isolated from the rest of the biosphere. Unless there is something about islands I don't know, like they absorb undefined nutrients from the sea? I would also argue that a Mars mission would require a more closed and clinical system than a habitat, because the journey is so far that you can't be producing a lot of waste or taking a lot of nutrient with you, and you also have to construct a system when you arrive. $\endgroup$
    – Phil H
    Jun 12, 2014 at 8:43

Q: The Gerard K. O'Neill idea was that... he spoke of "planetary chauvanism." Have you given any thought to building space colonies as opposed to building on a planetary surface?

Elon Musk: The problem with space colonies is not that it can't be done, is just that's doing it the hard way. In order to create a substantial space colony you have to transfer mass from a planet or from some asteroid, or something. You have to move mass from one place to another. So why move mass from one place to another instead of just going to where that mass is in the first place? Any sort of orbiting space colony is always, in order to expand, is always going to have to pull mass from somewhere, and why bother doing that? It just seems like a much harder thing to do than just going...

Q: Well the argument there would be to use asteroidal material for the colonies, so you haven't got the gravity well which you have on Mars or a planetary surface.

Elon Musk: It'd actually be harder to travel to the asteroid belt than it would be to travel to Mars. So, if you're talking about people coming from Earth, it's going to be easier to go to Mars. Having the atmosphere, you can use atmospheric braking as well, and you just have an enormous number of resources on Mars. Mars is like, it's not perfect, but it's pretty good. It's got a 24.5 hour rotational period. It's got a CO2 atmosphere, which means if you just had a transparent dome and pump, you could actually grow Earth plants in martian soil. In fact, it's recently turned out that martian soil is non-toxic so you could actually grow Earth plants in martian soil just by heating it up and pressurizing it with CO2... simplifying... [laughter]. You need a little fertilizer, but Mars actually has 2.7% nitrogen in the atmosphere which means that you can synthesize fertilizer as well. So yeah, it's a pretty good option. In fact, it's the only option, I think.


Sustainability on Mars does not require 100% closure of all loops in a life-support system. It only requires economically sustainable levels of closure - 'Good is good enough.' NASA is far too hung up on total system closure, particularly for Mars where there is no question that essential resources are there.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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