Before we go to Mars, why don't we build a sealed dome self sustaining colony in the Antarctic? One that can heat itself, recycle all materials, water and waste. Grow enough plants to produce oxygen and feed the colonists.

I am aware of Biosphere 2. That failed and it was in Arizona where they have plenty of sunlight and warmth. I think they should build one in the Antarctic. That would more closely replicate conditions on Mars with it being colder and having less sunlight. In the Antarctic, if something goes wrong they can open the door and let in oxygen or melt snow for water. Have a plane fly in supplies or evacuate them. Once we get a 100% self-sufficient colony functioning in the Antarctic, get all the problems figured out, then we could replicate the facility on Mars.

Also, if we build a self-sustainable colony on Mars. Underground or glass domes, or a combination of both. We would need enough plant life to produce oxygen and food for the colony. This is all possible.

Potential disadvantages of an Antarctic location include:

  • Some method would be needed to heat the interior of the dome, so the plants do not freeze.
  • If not enough sunlight was available for the plants it would have to be supplemented with grow lights.
  • A local power source would be needed - probably either nuclear power or a lot of diesel.
  • $\begingroup$ Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on Space Exploration Meta, or in Space Exploration Chat. Comments continuing discussion may be removed. $\endgroup$
    – Rory Alsop
    Commented Oct 27, 2023 at 15:41
  • $\begingroup$ Those who have answered and commented on this question including myself seem to be under the impression that the intent of the exercise is to test the capabilities and techniques of a Mars colony in a sealed environment on Earth, with the ability to survive at least 26 months without any outside air, water, supplies, etc. However in a comment on my answer you indicated that your definition of 100% self-sustainable means no more dependence on Earth, and no reliance on resupply. If that is the case that is quite futuristic and possibly a better question for Worldbuilding. $\endgroup$ Commented Nov 5, 2023 at 12:51

4 Answers 4


The colonization sites under consideration would be in the mid latitudes or equatorial, and would be more like Arizona than Antarctica in terms of sunlight, with a near 24 hour day/night period and the sun following a high arc across the sky. The main difference is intensity, and that can easily be accounted for.

And while Antarctica might be closer to Mars in terms of temperature, it is not in terms of what that temperature means for human activities...the thin atmosphere means EVA suits on Mars are likely to need active cooling, the cold is not a major hazard. In Antarctica, it very much is so, as well as a major burden on supporting a livable environment.

In short, it would not actually be more Mars-like in any of the important respects, but it would be far more expensive and dangerous than a more realistic equivalent built in Arizona.

  • 13
    $\begingroup$ This... the atmosphere of Mars is cold, but it's so thin that it acts more like a thermos bottle or blanket of insulation than anything else, and everything humans do (from lighting to running computers to biological life processes) generates heat. The problem is shedding heat back into the environment, not generating extra. Air conditioning isn't going to be overly difficult on Mars since there is some cold air to move around, but most likely they'll never need to actively generate heat. $\endgroup$ Commented Oct 24, 2023 at 15:36
  • 5
    $\begingroup$ @DarthPseudonym yeah, Mars is different. You might need to plug vehicles in...both to keep them warm at night, and to let them keep their interiors cool during the day, if they have large enough windows for a greenhouse effect. And of course, you'll want insulated gloves and boots because they'll be in contact with solid objects at equilibrium temperature. That's not necessarily going to mean they're cold, especially if they're generating internal heat. A radiator could easily get quite dangerously hot. $\endgroup$ Commented Oct 24, 2023 at 16:56
  • 2
    $\begingroup$ it's been a while since I've read anything about the Mars rover design, but I know for sure those include heating units in addition to the RTGs. I don't think they have any kind of active cooling. $\endgroup$
    – Erin Anne
    Commented Oct 24, 2023 at 22:00
  • 2
    $\begingroup$ Maybe they could build some kind of self-sustaining greenhouse structure in Arizona to try it out... $\endgroup$
    – adam.baker
    Commented Oct 25, 2023 at 5:24
  • 1
    $\begingroup$ @ErinAnne Electronics is pretty flexible about what temperature range it can operate over. Military-grade components are rated to -55degC to +125degC. Chimps in a can, not so much - we need extra clothing to survive below about 10degC, and we tend to die of heatstroke above about 40degC. $\endgroup$
    – Graham
    Commented Oct 25, 2023 at 14:08

One reason for not starting a new one is that we have already built several:

Flashline Arctic

Devon Island

Utah Dessert

Sealed 500 day missions



And the actual antarctic stations, one of the general themes of the last couple of decades of work there has been the human factors element of keeping people sane in the hostile and very artificial environment.

None of these have had 100 people in for a full Mars mission replica as of 2023, but there are no serious (as in money allocated) plans for a 100 person Mars mission either.


The conditions in Antarctica are very different from Mars so building a base there is not a good analogue. There is a near a vacuum on Mars and the amount of sun light received on Mars would be much less. It is very unlikely that the first Mars base will be in the Martian Arctic or Antarctic so they would not suffer from 6 months of darkness like an Antarctic base on Earth would. The gravity on Mars is also much less than on Earth.

No doubt some of the first human missions will study the technology required for a permenent base.


There are three main location categories for analog tests for Mars colonization:

  1. Lunar base
  2. Space station
  3. Biosphere on Earth

A lunar base will have the highest fidelity, providing an environment that is in a vacuum, has no radiation shielding, which is moderately isolated from Earth's resources, and which has reduced gravity. While not an exact analog of Mars, many of the techniques and lessons learned can be applied to Mars colonization.

Currently NASA along with international partners is planning a return trip to the Moon, with a long term goal of establishing a lunar base. However this will take many years just to develop the capabilities of building a lunar base, so in the meantime other analog tests could be considered.

Long terms stays on ISS have already been done, with part of the purpose being to build up experience for long term Mars missions. Longer duration space station stays are possible, ideally on a space station with artificial gravity to simulate the Mars environment. However there are no current plans for such a project, which would also take several years to develop.

You are suggesting a biosphere project on Earth. Besides the Biosphere 2 project that you mentioned, there have been other smaller scale Mars analog tests. A larger scale and duration biosphere type of project could have some benefits in gaining knowledge. However the location is not as important from an analog viewpoint since the facility would be completely enclosed. While prior projects required sunlight, modern lighting technology can now provide a better simulation of Mars lighting and day/night cycles than trying to find a location with similar lighting as Mars. And likely for a fraction of the cost of building, maintaining and supplying a test facility in Antarctica. Also the "resupply missions" to the test facility would likely be timed with the Mars 26 month cycle, which would not necessarily line up with the winter/summer cycles in Antarctica which can limit the schedule for supply missions depending on the location.

However the location does matter in terms of EVA simulations, which would be part of the analog since regular EVA's are expected on Mars. Similar to the ISS analog training that NASA has conducted for many years in the NEEMO project in an underwater laboratory off the coast of Florida. These training simulations include diving activities to simulate doing EVA's.


Desert locations have always been preferred for EVA training for the Moon or Mars, and I would expect that this would likely determine the location for a Mars biosphere analog project. Temperature would not really matter since for better fidelity the "bionauts" would likely be wearing temperature controlled suits like they would be on Mars, unlike the coverall and sneaker attire in this lower fidelity Mars analog test:

Mars Desert Research Station Mars Desert Research Station (The Mars Society, via Wikimedia Commons, CC-BY-SA-4.0)

  • $\begingroup$ The ISS and the many already existing Antarctic bases are not self sufficient. They get supply drops. They don't produce their own food, oxygen, fuel, and heat. I clearly said sealed 100% self sufficient.We could do it in Northern Canada near the arctic circle so we would still have a day night cycle. It would be sealed so no oxygen from the earth. Replying listing the ISS and Antarctic bases already existing that rely on supplies sent in does not answer my question at all. It missed my point. None are self sufficient. $\endgroup$ Commented Nov 4, 2023 at 22:21
  • $\begingroup$ Maybe a better way to phrase my question: Why don't we figure out self sufficiency first, then we can transport that knowledge to the Moon or Mars? Again I'm talking about a self sufficient sealed colony, get it to work on earth, before we try it on Mars. Saying Mars can get supplies every 28 months doesn't answer my question. That isn't self sufficient. One rocket fails on the way to Mars and everybody suffocates or starves. $\endgroup$ Commented Nov 4, 2023 at 22:30
  • 1
    $\begingroup$ A Mars colony would have enough supplies to survive a failure of a scheduled resupply, just like ISS currently does. More likely there would be redundant supply missions, so the failure of one or even two would not be fatal. If the technology by then is still so untested that they can't even guarantee one out of three successful supply deliveries then it would also be unsafe to try and land humans on Mars. Full sustainability is Elon's concept for the future in case Earth gets wiped out by an asteroid or whatever. That is very futuristic and possibly more suitable for Worldbuilding $\endgroup$ Commented Nov 5, 2023 at 12:41

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.