The illustrations of O'Neill cylinders I have come across with so far allow unrestricted view through the whole tube. Some are more playful with topography but still, there is a visible end. Are there reasons that forbid to close off parts of the tube, lets say, by a wall of mountain?

I would like to have warmer, moister and colder, dryer climates (possibly even more variations) within one single cylinder, that feel "naturally seperated" instead of a greenhouse within the complex. I also would like to increase the surface area with a lot more topography, probably even hollowed out in some places.

(As a site note Isaac Arthur talked about using aerogel, an ultra light material, to use under such constructions, to make them less heavy. I'm rather talking about an illusion of mountain than a real mountain.)

Each section could have it's own way of lightsetting, with a "sun" in the centre, or indirect light embedded in the environment, a long tube of light, or lightpoles as in street light..

Well, I wonder what reasons could speak against this idea. Could they be worked around, or do they make the idea impossible?

  • 2
    $\begingroup$ I don't see why it shouldn't be possible, though you would probably need to make the barriers between section go all or most of the way to the centre if you want different heat, moisture, and weather patterns. Such barriers may also make sense in case of leakage, though they then have to be strong enough to resist vacuum. $\endgroup$ Commented Sep 27, 2019 at 13:04
  • $\begingroup$ Unrelated to the feasibility, but if the craft is designed for habitation, it would be an incredible waste of space and resources to have dry, cold or daresay hostile climates, given that the entire thing is so heavily engineered. But then again, Hunger Games in space.... $\endgroup$ Commented Sep 27, 2019 at 13:56
  • 1
    $\begingroup$ @VaradMahashabde Not necessarily hostile, I was rather thinking of a variety of climates to comfort a wider range of flora and fauna, more as if I would merely pick "one" climate. Some plants need dormancy (hibernation for animals respectevly), while other plants die by the tiniest amount of frost. $\endgroup$ Commented Sep 27, 2019 at 14:03
  • 1
    $\begingroup$ @VaradMahashabde Not really. Firstly, if there is one thing in space it is, well, a lot of space. The resource question only comes up in the early days of space habitat construction or in Dyson Swarms near their waste heat limits. For a K1 civilisation with massdrivers and industrial parks on the moon construction of cylinders is trivial. Besides, some people and animals like harsh climates for habitation and recreation. Think of winter sports or using the thing a a nature preserve for animals used to harsh climates. $\endgroup$ Commented Oct 1, 2019 at 8:22

3 Answers 3


Are there reasons that forbid to close off parts of the tube, lets say, by a wall of mountain?

Well, you'd spoil the view, for one! Its hard to get 2000km+ uninterrupted vistas on earth, but the theoretical limits for cylinder habitat construction would allow them, and so why not take advantage of that fact?

If you were making something more like a Stanford torus your lines of sight are limited by the curvature of your habitat. The original Stanford design (major radius 1790m, minor radius 65m) means the longest line of sight you'd get would be under 700m, so if you divided your habitat up into 8 regions each region would appear to extend beyond the "horizon" from its midline.

Of course, this is just an aethetic consideration, so there's absolutely nothing to stop you violating it.

I note, however, that a Stanford Torus design is probably more amenable to slicing up into distinct ecoregions, as you could completely seal each section from each other in a way that could not be trivially done in a cylinder habitat (the minor radius of a torus generally being much less than the major radius of the habitat, etc). This would allow you to have radically different biomes, temperatures, pressures or even atmospheric compositions if that's what you wanted (how about entirely water-filled segments, for example?) and they'd be completely isolated from each other.

In a cylinder, you might need to make your dividing walls as high as the scale height of the atmosphere, and that's a tricky thing to calculate. It may well be much higher than the minor radius of a toroidal habitat (which presents engineering challenges) and unless it completely transected the habitat you won't get perfect sealing between the two regions. You may not care, of course.

Each section could have it's own way of lightsetting,

This might warrant more serious barrier walls, as you don't want your cold dreary biome having heat and light leaking into it from tropical world next door.

I also would like to increase the surface area with a lot more topography, probably even hollowed out in some places.

Might be easier just to build more habitats...

  • $\begingroup$ This is great, thanks! Regarding your example of sections filled with water; would this not be a massive difference in weight compared to the other sections? Do they not need to be even in weight? Same question fits the cylinders as well; would the weight need to be evenly restributed troughout the cylinder to ensure stable rotation? $\endgroup$ Commented Sep 27, 2019 at 13:31
  • $\begingroup$ @BackupPlan so long as the water sections were distributed evenly around the rim (eg. 2 seconds could be diametrically opposite each other) you wouldn't unbalance the rotation. You'd need materials strong enough to contain the required amount of water under that rotation, but if you're using fancy things like carbon nanotubes and your habitat isn't thousands of kilometres acros you'll probably be fine. $\endgroup$ Commented Sep 27, 2019 at 14:35

Are there reasons that prevent the closing of the ends? The ends are often closed by some form of transparent material and act as light sources. So mountains might block light, but otherwise could be placed there if windows or other light sources ran the length of the cylinder.

If any arrangement of windows may be used, the cylinder can be orientated in any direction and the cylinder is large enough then it should be possible to create multiple climate zones by a number of different methods.

If one or more large windows were to run along the axis of the cylinder they could be engineered in a number of different ways to control the light and heat entering. The windows could narrow towards one end allowing less light and heat to enter or the material of the windows could be modified along their length with coating and filters to allow light in but to block heat.

If the cylinder was fairly long and the heating difference was very great the circulation of the atmosphere within the cylinder would not be able to transfer sufficient heat from the warm end to the cold end and a temperature gradient would exist.

This effect could be enhanced by blocking air flow with mountains running around the circumference. It could also be enhanced by having a dark cold heat sink section at one end of the cylinder.


It can be done in the requested way, but it would require proper managerial skills

(read: yelling at engineers and ignoring their objections)

Dividing any kilometres long cylinder in to airtight subsection is a very reasonable safety feature. Having it would allow you to have all climate zones you wanted, just in separate cells.

If you haven't divided the cylinder into subsections, then unless the cylinder is insane long, the air flow would try to equalise the temperature, and you would need to pump huge amounts of energy just to maintain temperature gradient.

The only way that I can imagine it to work in semi-reasonable way, is to:

  • have a nuclear reactor that would produce energy for whole colony and would dump its waste heat
  • the waste heat would be heating up this single place, and slowly going all around the station
  • everything would have to be really long and there would be tricks to slow the airflow in form of mountain ranges or lines of very tall trees.

You must log in to answer this question.

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