We have a fleet of sub-light speed generation ships and associated engineering devices/ships. Their original world can no longer support their species, and they must move to survive. The fleet of course cannot support its inhabitants forever. There is therefore a need to populate a new, self sufficient, home world.

Concept drawing of giant space ship

Image Source

The only feasibly reachable planetary system they've identified is very much like our own Solar System, but there's no Earth, no Mars, no Venus - just a wide gap where the Goldilocks zone is.

For the sake of simplicity, I'll call these planets after their Earth solar-system equivalents.

There's a larger asteroid belt, but it's close to the Jupiter whose gravitational forces has prevented the formation of a planet.

What would it take to create a new home planet out of the resources available in the other planets and whatever engineering the fleet has brought with them? This should be earth-like, capable of sustaining a breathable atmosphere and plant-life to support life as we know it.

Salient facts: Population of the generation-ships - One million homo sapiens equivalents. There is an ark of breeding stock for plants and animals.

Resources - We make the assumption that these people understand the mechanics of doing what they want and have brought some necessary equipment - some of the ships are modular in construction so that engines/power generators/whatever else can be separated and combined/used in different ways.

Limit of resources - the fleet can support the species for the next 1,500 years (counting from when they reach their new home system).

I couldn't find any questions that actually built a world, so I'm hoping that we can make this one work.

Scope Guidance

So, I'm looking for a high scale plan with some time estimates for each stage and roughly what would be involved for each stage. Alternative solutions are also welcome.

The end result should be a self-sufficient environment which would support and nurture this species until they decide to kill each other off or otherwise ruin yet another world....

  • 1
    $\begingroup$ Like I said in the question, they're aware of how to do this and have appropriate plans. They certainly haven't assumed that they'll find a perfect "Earth" analogue. $\endgroup$
    – user10945
    Feb 20, 2017 at 10:35
  • 1
    $\begingroup$ @Pᴇᴛᴇ That leaves a lot of room. If you are actually speaking of collecting materials to build a terrestrial planet, the fleet will need to use an awful lot of energy to move them out of their orbits, and most of it will be just to collect "useless" mass into a sphere. $\endgroup$
    – pablodf76
    Feb 20, 2017 at 10:51
  • 1
    $\begingroup$ While it's physically possible I'll leave the discussion on whether it's viable or not for someone with more time to answer than I! $\endgroup$
    – Joe Bloggs
    Feb 20, 2017 at 11:03
  • 31
    $\begingroup$ If they have the power to build a planet, why can't they fix their current one? It seems easier to scrub the atmosphere or filter the oceans or even shift the orbit than to start from nothing. $\endgroup$
    – Zxyrra
    Feb 20, 2017 at 16:52
  • 5
    $\begingroup$ I upvoted this question just for the WorldBuilding joke. $\endgroup$
    – M i ech
    Feb 20, 2017 at 18:31

12 Answers 12



Difficulty considerations aside, if you took a lot of asteroids and merged them together into an earth-mass planet, it would be entirely molten just from the gravitational binding energy. You'd need a few million years minimum just to get a stable crust, then you'd have to deliver water and volatiles..

Then, once all this had settled down, oxygenate the atmosphere.

So no, unless you have quite incredible technology levels. We could do it with human+ tech now, but the time requirements would be huge; millions of years for gradually shifting asteroid orbits and then dealing wit the results of the engineered collisions.

Much easier would be hollowing out asteroids, spinning them up for interior gravity and turning them into habitats. This would still require a huge amount of work - not least making them strong enough to hold together, but could be squeezed into 1000 years. It's also a much more efficient use of mass.

  • 2
    $\begingroup$ Note also that even if you had a more-or-less suitable planet just outside the Goldilocks zone it would be extremely costly to move it. In-story, maybe there could be a conflict between those who want to live on a planet (even though it would mean spending their last energy and time into terraforming it) and those who'd rather take the realistic route of living on artificial habitats. $\endgroup$
    – pablodf76
    Feb 20, 2017 at 11:57
  • 1
    $\begingroup$ Brainstorming here... what if they didn't bother making the world solid, just dock asteroids together like an old Roman arch with precisely fitted stones. No mortar, no welding... no heat. Do you think that might work? $\endgroup$
    – SRM
    Feb 20, 2017 at 20:06
  • 1
    $\begingroup$ Honestly, a 'Jupiter' style planet is very likely to have a bunch of moons. Trying one of them first might be the best plan. $\endgroup$
    – Turksarama
    Feb 20, 2017 at 23:10
  • 3
    $\begingroup$ "If you took a lot of asteroids and merged them together into an earth-mass planet" there's a basic problem there: finding enough mass. Our own Asteroid Belt only contains 3e21 kg of material or not even 5% of the Moon, and half of that is in just four asteroids. The Jupiter Trojans only have a fifth of that. Large amounts of material in the inner solar system, free of the influence of a gas giant, would have coalesced into planets, but those planets aren't there. $\endgroup$
    – Schwern
    Feb 21, 2017 at 0:06
  • 2
    $\begingroup$ The planets were molten because they started out molten. As the remnants of novas, they were quite hot to begin with. This isn't the case for the asteroids in the belt. Now, crashing all of the asteroids into each other to form a planet is going to create a lot of heat, but I'm not sure that it will generate "the whole thing will be molten for millions of years" levels of heat. $\endgroup$
    – Shane
    Feb 21, 2017 at 19:49

What would it take? Answer: Nothing. They're already living in it. If there's a Goldilocks zone, they can park their Space Ark there. That basically sorts out their energy requirements forever, and they can mine asteroids. There's no reason they can't live in their Space Ark indefinitely.

But more likely, like any family, they're going to start wanting to build extensions on their house as their family gets bigger...

Andrew Dodds's answer is perfectly correct - they can built habitats out of asteroids. Hollowing them out is probably not going to be successful, because space debris tends to be fairly porous, but at the very least it'll provide the raw materials for making concrete. (Not the kind with limestone, of course, but as the Romans proved, there are other ways of skinning that particular cat.) The various metals and silicates in the asteroids can also be processed for other raw materials, and water-rich asteroids will start building up atmosphere and terraforming in those habitats.

The best part with this is that it's evolutionary. You start with the Space Ark ship, and over time it'll accrete more and more habitats attached to it. What you'll end up with won't be an "Earth" as such, it'll be more of a 3D zero-gravity rabbit warren.

  • $\begingroup$ what about food? Wont they run out of that if they don't have space for crops? $\endgroup$
    – J.Todd
    Feb 21, 2017 at 0:25
  • 2
    $\begingroup$ Yes, but they could make more room for crops quite quickly (probably a few years, max, if they have a high tech level) in comparison to the time taken to build a planet. They probably would have made this room before departure, though. $\endgroup$ Feb 21, 2017 at 1:06
  • 2
    $\begingroup$ @Viziionary Why don't you think they'd use some of those habitats for growing crops? We're on a generation ship here, so by definition that's already got all the hydroponics tech you could want. And a generation ship has to be self-sufficient pretty much indefinitely, so they won't just have spares, they'll have the means to manufacture spares from raw materials. You're not going to have acres of rolling grain under an alien Sun, but you'll certainly have algae tanks, tunnels filled with rows on rows of salad leaves, "balls" of fruit trees tethered in the centre of huge round rooms, etc.. $\endgroup$
    – Graham
    Feb 21, 2017 at 10:24
  • 1
    $\begingroup$ @JarredAllen if they have enough crop space before departure it seems like there would be no need for a planet. What's wrong with living on a high tech space ship if you have all the food and resources you need, and can just refine metal from asteroids to expand your ship as needed? $\endgroup$
    – J.Todd
    Feb 21, 2017 at 10:44
  • 1
    $\begingroup$ @Viziionary One word: babies. As your population increases (and you want it to increase - remember that you're setting up a new human colony here), you need some way to feed and house them. The Ark is going to have some spare capacity so it won't be immediately urgent, but it's going to be needed sooner rather than later. $\endgroup$
    – Graham
    Feb 21, 2017 at 13:22

No, you cannot build a world in 1000 years

As already mentioned by Andrew Dodds, the time span you have at your hands is orders of magnitude less than what is simply required by a newly formed planet to cool off and get stabilized in its orbit.

Of course we are talking about moving asteroids from their orbit and combining/colliding them together to form a planet in the habitable zone. Here are some of the major concerns. There are many, many more, but I am confining myself to only the primary ones here.

  1. Moving mile-wide and larger asteroids out of orbit is extremely difficult Yes, it is. With a density of no more than 3 g/cm$^3$, (normal for metal/rock mixture) an asteroid of diameter 1 km would have a mass of 1.57 x 10$^{12}$ kg (i.e. 3 billion tons). Asteroids in the asteroid belt have a speed of more or less 25 km/s. One asteroid of 1 km diameter would have momentum equal to 3.92 x 10$^{16}$ kgm/s.

    It is going to slightly hard for you (the people in the spaceships) to change the direction of motion of hundreds of millions of asteroids of this much size twice. Once changing their direction towards the central star (to bring them inwards into the goldillocks belt) and then, at tangential direction (to put them into orbit around the sun and stop them from going on and falling into the sun).

    I haven't done the detailed math, but I can tell you already it will be practically impossible to meet the energy requirements for this, unless you are using antimatter engines and have tens of thousands of tons of antimatter available to you.

  2. It takes really long time at conventional rocket speeds to move things at interplanetary scale

    A one-way journey from Earth to Mars takes 150 days (shorter estimate). And that is when you have everything planned and a lot of fuel at your disposal. For pulling in individual asteroids one by one from the asteroid belt to NEO distance, you would require at least 5 years for each round trip. Travelling to the asteroid, attaching to it and then travelling back to NEO.

    If you work on individual asteroids one by one (and I see no other method, even theoretical, at current technological levels), it will take you a couple million years at least just to bring in the chunks together in the same NEO orbit, let alone combining them to form a planet!

  3. You need the right materials. And lots of them

    Earth has a large content of silicates and metallic minerals. The sodium content of our planet alone is astounding. Then there is a lot of iron and nickel, too. Beyond the freeze line, a lot of the asteroids are made of ices (frozen carbon dioxide, ammonia, methane etc). Bringing them closer to the sun into the goldillocks zone will quickly melt and then boil them away, removing all hopes of incorporating them into your dream-Earth.

    While water is available in more than excess quantities throughout the galaxy, the correct ratios of rock, water and ferromagnetic metals (for the core) is far too difficult to find or create. If you pick up asteroids randomly, you would probably end up with a terrestrial planet with a too high metallic content.

  4. The cooling will take tens of millions of years

    That has already been discussed by Andrew Dodds, so I will not get into the details here. It should be sufficient to say that your newly formed planet will take orders of magnitude more time to just cool off, than what you have available at your hands.

  5. And the other little issues…

    Like oceans formation… the right chemical mix in the oceans… setting up the water cycle… setting up the atmosphere with correct composition and thickness… continent formation… setting up and maintaining the orbital tilt… finding and capturing the right sized moon (of course you wouldn't want to go the giant impact path)…

To Sum It Up

…it means something to say that our Earth is 4.56 billion years old.

  • 1
    $\begingroup$ I'm rather curious now if you could play some serious games with the N-body problem, and try to perturb thousands of orbits slightly to get the asteroids to manage some of the energy problems themselves. Certainly an absurd idea, but hey, that's world building, right? $\endgroup$
    – Cort Ammon
    Feb 20, 2017 at 19:48
  • $\begingroup$ O.o Hmm. But won't a large number of them go the other way right into the Big Brother or out of the solar system altogether? I mean, some will come falling inwards, but many would also go to the big, bad, sinister guy standing out there. It will take much, much more energy to bring those escapees back :D :p @CortAmmon $\endgroup$ Feb 20, 2017 at 20:58
  • $\begingroup$ You'd have to make them do something akin to a gravity slingshot or two on the way out, so that you maximize the escaping energy and minimize the lost mass. As for the guys standing out there, I think you could sell this as a defensive approach: porcupine teraforming! $\endgroup$
    – Cort Ammon
    Feb 20, 2017 at 21:18
  • $\begingroup$ You don't need very many trips to get enough mass for your planet: asteroid sizes follow a roughly power-law distribution. Grabbing the half-dozen or so largest bodies should get you the majority of the available mass. $\endgroup$
    – Mark
    Feb 20, 2017 at 23:02
  • 1
    $\begingroup$ @Mark - Even with larger bodies they would take inversely more energy (and time) to move. If moving a mile-wide asteroid is difficult (as in point 1), imagine a several-thousand mile 'moon'. $\endgroup$
    – Robotnik
    Feb 21, 2017 at 0:04

Build a fusion candle.

You have a large gas giant? Perfect, you have a world mover with its own reaction mass/power supply.

I assume this gas giant has a few largish moons? If not, move it around the system to capture some other suitable planet.

Forget the asteroids, they contribute nothing of value, at least concerning raw mass. (Wikipedia: "The total mass of the asteroid belt is approximately 4% that of the Moon.") Maybe use a quick sweep with your gas giant to clear it.

Once you have your mobile gas giant with one or more nice rocky satellites, move the assemblage in a cozy orbit and start to terraform the moons. If you did it just right, those icy asteroids you swiped earlier, or maybe a saturnian ring, can provide the water for your oceans.

The biggest hurdle is your time plan. I don't know how long the astroengineering will take, but 1000 years isn't even enough to build habitable biospheres from scratch.

You'll really want to start (while all the big moving goes on) by constructing closed ecologies to resettle your tribe, with mining and farming, which will extend the time left to you. No need to restrict yourself to an all-or-nothing / ship-or-planet choice.

  • 5
    $\begingroup$ Okay, settle down Larry Niven. $\endgroup$
    – alessandro
    Feb 21, 2017 at 11:14

I would like to post an answer which expands on Andrew Dodds' excellent post. The colonists, upon arrival, would most likely not WANT to settle on a planet.

1) They have lived their whole lives in space and that's probably true of their parents and grandparents as well. People accustomed for generations to living aboard ship will likely view the lifestyle of living on a planet as akin to how most of us would view the Amish. I'm sure they are generally happy with their lifestyle, but I'm not going to do it myself. ("You grow your crops in DIRT? Actual dirt? And you eat real animals? And you go outside and expose yourself directly to the raw environment?")

2) Asteroids (or space stations built from them) can easily be moved, eventually forming a Dyson swarm. This is much more efficient and a Dyson swarm is largely the standard model nowadays for a K-2 civilization.

3) There is a LOT more living space. A planet can support a few billion inhabitants, a Dyson swarm can support trillions.

4) Asteroid mining is better than planetary mining. The reason we don't do it now is because of how hard it is to get there, and minor lifestyle inconveniences like not being able to breathe. But if you already are there, and have the equipment to survive, the resources available are far greater than what we have on Earth.

5) Earth-size planets have inconveniently large gravity wells, hampering space travel. These colonists are likely to want to continue to travel in space and constructing a massive gravity well just so you can climb up and down it is not helpful at all.

6) The colonists may have genetically engineered themselves to adapt to the space environment, for example adding radiation resistance or tolerance for low gravity. Regardless of cultural and lifestyle issues, they might find an actual planet not especially welcoming biologically, or at least not more so than their regular space environment.

7) Energy availability on a planet is bad. You have three plausible choices for energy: solar, fusion, and fission, and all of them favor asteroids or space stations. Solar collection is optimized by constructing a Dyson swarm, for maximum collection area. Fusion power requires lots of volatiles (likely hydrogen), which will be most easily found in either cometary ice or, secondarily, the atmospheres of gas giants, depending on how good your rockets are and how much mass the gas giants have. This requires lots of travel to the outer, or at least middle, solar system to collect it. The low gravity of asteroids simplifies the launching and landing (as well as less energy and time needed for travel due to starting farther away from the star in the first place). Finally, fission power requires lots of uranium, which needs to be mined, and mining is easier on the asteroids. And if you're using fission, disposing of the radioactive waste is a lot easier if you can just toss it out into space (although I would expect colonists at this technology level to be quite a bit better at recycling, or at least concentrating, the stuff than we are).

8) Planets have pollution problems that you just don't have on space stations. While simply throwing waste overboard is not necessarily suitable in general (because it is still in your orbit and sooner or later it will come around again and might hit you) you can recycle what you can and easily move most of your really troublesome garbage into a non-intersecting orbit. If some disaster contaminates an entire habitat, you can just build a replacement. Actually there would probably be so many habitats that the civilization as a whole would barely even notice a Chernobyl or Fukushima.

9) Politics are easier when you have millions of more or less independent but cooperating habitats. When your only practical choices for war are "we will stop trading with you" and "we will blow up your habitat, killing everyone" it is more advantageous to just pack up and move away or ignore neighbors you dislike, discouraging violence and war. It also means everyone can find a political system that suits them personally. I think it's unlikely that everyone in the colony fleet will want the same political system. Some might want a libertarian utopia, a communist utopia, a socialist utopia, a military dictatorship utopia, or any other form of utopia they can think of, and all of these systems, including many I can't think of, will likely exist. "Asteroidica, love it or leave it" becomes a real option and people can move to any of the thousands or millions of competing political and social systems.

I think interstellar colonists will be more interested in the number, variety, and positioning of asteroids in a target star system rather than caring about planets at all, with the possible exception of small gas giants for hydrogen collection. For further insights I recommend Isaac Arthur's excellent Youtube series, including an episode that specifically talks about this issue: Life in a Space Colony, ep3: Early Interstellar Colonies

  • 1
    $\begingroup$ "radioactive waste is a lot easier if you can just toss it out into space" - just make asteroid from them and let them decay if they are lazy to burn them - they will get some energy as by-product, there is no need to toss them anywhere. They are not limited nor with space for that not with volume. " it is more advantageous to just pack up and move away or ignore neighbors you dislike", " thousands or millions of competing political and social systems." - nice, that's true. $\endgroup$
    – MolbOrg
    Feb 21, 2017 at 6:17
  • 1
    $\begingroup$ @MolbOrg Thanks for fixing the Youtube link :) $\endgroup$ Feb 21, 2017 at 6:32

We need an approximation of their technology level.

Suppose the trip took 1500 years. The ships are the size of US aircraft carriers. And the distance was about 20 light years.

The ship would then be travelling at 0.013c, which is pretty fast. We'll use Newtonian energy approximations (it is relatively accurate at 1% c) and we get roughly 0.5% of the mass of the ship is kinetic energy.

An aircraft carrier is about 100 kilotonnes. So the KE of the ship is 5e22 J, or 1e23 if you count coming to a stop at the end. Slightly more if you want constant accelleration (as that leads to a top speed 2x higher; this is reasonable to not assume, as at 0.01c the interstellar medium isn't frictionless).

Plowing through the interstellar medium is another issue. At 0.01 c each square meter of the ship impacts 158 micrograms of hydrogen/year. Over the 1500 year journey, that sums to 237 milligrams. A cylindrical aircraft carrier has a cross section of 5000 m^2, so that comes to about 1 kg of hydrogen.

In comparison to the ship mass that is trivial; thus the interstellar medium is mainly dangerous as it is a form of hard radiation.

In comparison, the binding energy of Earth is about 1E32.

The big problem with assembling a planet is dissipating that binding energy. The binding energy of an earth-sized planet is enough to send 1 billion supercarrier sized generation ships to nearby stars.

So the task you have for the ship is 1 K-type civilization higher in effort than the ship itself.

Another way to look at it is that it would take a K-2 type civilization 6 years to process enough energy to disassemble or assemble a reasonable tempurature earth-like planet. And if you are anywhere near a K-2 type civilization, why do you want a planet?

But we can approach it that way.

Assume these generation ships where sent from a K-1 type civilization. The trip might have begun with the plan to aim for somewhere else, or get resupply or something, but a solar-system wide war that ended badly for everyone wiped out the civilization behind them.

Over the 1500 years, they managed to avoid a similar conflict, and developed technology sufficient to reach a K-2 level of civilization; they just are missing a star.

They arrive in the destination solar system, and they release von neumann machines. These first disassemble small bodies and reproduce, gathering raw materials.

This provides them with an industrial base. Huge solar collectors are built to lift more mass out of gravity wells and build yet more solar collectors. Matter is fuzed and transmuted on large scales. Low atomic matter is sucked out of gas giants using ridiculous technologies (fusion candles whose only purpose is to shoot gas up the gravity gradient? Orbital fountains? Vacuum energy pumps?)

Computronium is produced and a Matrioshka brain is constructed to aid in further technological development. The star starts to go dark.

Only as they approach a full K-2 civilization does the "lets build an Earth" project start. A full K-2 civilization would take 6 years to pull this off; either they efficiently wait until they are a K-2 civilization (or beyond using exotic technology), or they do it slowly before hand (you know, for practice). Regardless of which they do, if they approach K-2 exponentially almost all of the work will be done when they are almost-K-2.

A 1% K-2 civilization requires 600 odd years to build an Earth.

Of course, the technology level required makes the 1500 year limit a bit of a joke; this civilization has mostly moved out of physical bodies by this point. The building of the duplicate Earth might be a thank you to the ancestors, and a nature reserve for them and a biosphere. Pretty expensive even for a K-2 civilization however.

  • $\begingroup$ If anything, you could say that the binding energy is energy that you get for free when doing this. This is why planets will self assemble in the right situation. All you need to do is fly to the asteroid belt and do something to slow asteroids down. This will cause them to fall towards the sun. If done right, they will assume a new lower orbit at the location you want your planet. If you get them close enough to each other, they will clump together. Keep going and eventually the clump will have enough total gravity to reshape the material into a sphere and, viola, you have a planet. $\endgroup$
    – krowe2
    Feb 21, 2017 at 17:53
  • $\begingroup$ @krowe2 The clumping generates piles of heat, and getting rid of that is the problem. That is free energy, but it is free energy in the wrong place. You have a star nearby, you have lots of free energy -- the hard part is arranging for the energy to be in the right place doing something useful. Hence treating it as a generic energy handling budget problem. $\endgroup$
    – Yakk
    Feb 21, 2017 at 18:19

Well, I have to ask...

The fleet of course cannot support its inhabitants forever

...why not?

We don't quite have the tech to keep people alive in a fully technological environment indefinitely yet, but we're getting there - and if that technology would be an absolute prerequisite for a generation ship. So honestly, if you have generation ships, why would you need to anchor yourselves to a planet?

A planet is a nice place for a species to grow up, but ultimately there's a lot of wasted mass and wasted space that we just can't use, and we're stuck at the bottom of a pretty steep gravity well. A planet, ultimately, could be seen as a dangerously unstable place to live. You can't move it if there's a big rock coming, you can't use the vast majority of its mass, and you can't control its climate.

Rather than trying to build a planet, think instead of staying aboard the fleet and just using the resources available in the solar system. Any raw material you can think of can be got from asteroids or gas giants. If you need more living room, consider building asteroid terreria and living inside them.


I don't know if it can be done in 1000 yrs, but it seem unlikely.

Here are some ideas to help.

First, you need to get a large dense enough group of rock into an earth like orbit. It has to be large enough for gravity to pull in debris from the asteroid belt(well maybe that is too much) and surrounding space.

From what I have learned the planet will start auto-rounding at a round 100mi in diameter with the right density core.

If you could add a significant layer of water, perhaps you could float an island like structure on it.

The water would cool the incoming debris, and allow it to more gradually fall into the center of the planet. There would be giant waves from underneath as the planet rounds itself, but the water would still buffer them.

Hopefully, your island like structure, can move like a ship at sea. As long as it has enough engine power to move of the way of giant waves you could start living on them. Also growing food on the tops decks of the ship(s)/island(s).

After that timing is going to be a critical issue. You need enough debris to fall into to eventually create a usable land mass, but not so much that the impacts evaporate or blast the water into space.

If you don't have gravity assisting you, and you have to actually move 1000's or 10,000's of asteroids manually the project will probably fail.

If you can nudge enough asteroids into our new earth's gravity well it improves your chance of success. All of your ships will have to be nudging debris causing it to change trajectory to our planet.

You will have debris which contains or even mostly water so you don't have to make it all yourself.

The key is having enough land to settle people, but not enough mass to pull in the entire asteroid belt. While pulling in the asteroid belt would certainly create a planet, it would never cool fast enough to live on in a 1000yrs.

  • $\begingroup$ Curious: Would there be enough metal in every asteroid combined for the colony to use swarm technology to mine and refine it all in about 500 years and build it into a massive ring space station with the capability to support the population in a sustainable manner? s-media-cache-ak0.pinimg.com/originals/ac/7b/12/… $\endgroup$
    – J.Todd
    Feb 20, 2017 at 23:40
  • $\begingroup$ enough metal purely by mass probably, but you might not have enough of the hard metals like platinum. Depends on how fast they can churn out metal from their factories. 1,000,000 / 800 per sq mile. You would need 1250+ square miles for a tightly packed city. I would say 2500 to be safe, and much area will require a bunch of infrastructure. Over 6 million pounds would be needed. Over 2 billion yards of cement like material. That would probably produce its own gravity field, and draw stuff in. $\endgroup$
    – cybernard
    Feb 21, 2017 at 0:13
  • $\begingroup$ I would imagine the city in general wouldn't be the important part, but fields for crops right? And that's assuming it's even possible to make dirt from asteroids. $\endgroup$
    – J.Todd
    Feb 21, 2017 at 0:23

Hypothetically you could probably do it... The first thing you'd have to do is get an iron nickel ball that is roughly 1200 km thick. That's not that hard since you could farm it from just about anywhere in the solar system. You set up a massive magnetic spinner thing, then you wrap that with liquid iron-nickel that is has been heated to a very high temperature till the radius is about 3500 km. Again. This shouldn't be very hard because there should be plenty of iron-nickel in any solar system all over the place in fairly easy to get places...

The problem is the mantle and crust which can easily be made, but it's more the getting the proper balance of materials, but in a thousand years that shouldn't be a problem. You just continue heating as you poor the liquid resources onto the planet and then when you get to the right radius you stop heating, poor a layer of silicates more or less over the mantle and build up the continents and such as you like.

The biggest issues is getting the materials hot enough. The giant magnet thing might seem like a problem, but you're building a planet. if you're doing that you can build something like giant magnet thing.

The second issue is the proper material balance...

Lastly there is the terra forming which may make you run into a problem, not in the atmosphere. THat'd be easy, since you're already moving massive resources it's hardly an issue to dump a bit more. The hard part is that dirt is not soil so once you have it all there you have to do a lot of working on making and spreading soil. If people worked diligently it probably could be done, but more than likely they wouldn't. They'd get to "good enough" and stop, which would slow the process down a lot and then it becomes a probably not, but it can probably be done ignoring human laziness.

And while all this is possible hypothetically, it would be a huge amount of work for and you'd be wasting resources and time to do something that is largely just a show of power and not a "needing this for living on" situation.

But whether it can be done or not isn't really an issue so much as if a civ could, they'd not do this. If they have the tech to move the resources to make a planet then they could have done it to their homeworld. If it's their star that is the issue they could simply move the planet. If it's the magneto sphere dying then they can charge it back up.

So you see, the tech needed to do this wouldn't make a whole lot of sense.


If the gas giant is on the smallish side and within the Goldilocks Zone, your ark ship could build very large microwaves (or some other suitable electromagnetic wave machine...eg giant lasers) to photoevaporate the lighter elements (H, He, Li, etc). This should hopefully leave you with the rocky core and the heavier gases (O, N, At, CO2, etc). It'll still be a tad bit warm, but less so than the amount of heat generated by fusing asteroids together.


If the gas giant has at least one moon that is suitable for terraforming, an artificial sun can be built which supplies the energy needs. This will work only if the moon receives less energy than optimal.

If there is a different moon that receives more energy than needed, then a megastructure that covers the moon needs to be built. The purpose of the megastructure is to reduce the intensity of the energy from the star.

Another option is to build a ring world. Similar to the rings in the Halo game.

A seemingly good option will be to build a Globus Cassus.


O'Neill Cylinders - Just going to paste the answer

Assumption the Gas Giant or Moons or both can supply enough "gas" to create an atmosphere.

One distinct (my opinion) benefit here is we know how to make bigger rocks into smaller rocks :)

Ok, as each habitat is constructed it decreases the stress placed on the Generation Ship(s) and with logistical planning may be capable of reducing demand on Ships Stores (build a food/water producing habitat first?).

It is believed that a single O’Neill cylinder could house several Million people.


Assuming the Gas Giant has the required gas’s for creating an Earth Atmosphere, then the colonist may want to construct a ship that can “dip through” the Gas Giant's atmosphere and “Scoop and Compress” the gas. It is equally reasonable to assume there are comets or other sources for acquiring the components of an atmosphere as well as the regolith for soil.

The only Time estimate I have is for the 2nd Island and that is 2 years (L5 Society). So I feel it is reasonable to assume the 3rd Island can be built in less than 100 years.

The following is extracted from


The Island Three design, better known as the O'Neill cylinder, consists of two counter-rotating cylinders, each five miles (8 km) in diameter, and capable of scaling up to twenty miles (32 km) long. Each cylinder has six equal-area stripes that run the length of the cylinder; three are transparent windows, three are habitable "land" surfaces. Furthermore, an outer agricultural ring, 10 miles (16 km) in radius, rotates at a different speed to support farming. The habitat's industrial manufacturing block is located in the middle, to allow for minimized gravity for some manufacturing processes. A NASA lunar base concept with a mass driver (the long structure that extends toward the horizon that is a part of the plan to build O'Neill Cylinders). Substitute Asteroids, Moons etc To save the immense cost of rocketing the materials from Earth, these habitats would be built with materials launched into space from the Moon with a magnetic mass driver.


Artificial gravity

The cylinders rotate to provide artificial gravity on their inner surface. At the radius described by O'Neill, the habitats would have to rotate about twenty-eight times an hour to simulate a standard Earth gravity. Research on human factors in rotating reference frames indicate that, at such low rotation speeds, few people would experience motion sickness due to coriolis forces acting on the inner ear. People would, however, be able to detect spinward and anti-spinward directions by turning their heads, and any dropped items would appear to be deflected by a few centimetres. The central axis of the habitat would be a zero-gravity region, and it was envisaged that recreational facilities could be located there.

Atmosphere and radiation

The habitat was planned to have oxygen at partial pressures roughly similar to terrestrial air, 20% of the Earth's sea-level air pressure. Nitrogen would also be included to add a further 30% of the Earth's pressure. This half-pressure atmosphere would save gas and reduce the needed strength and thickness of the habitat walls.

At this scale, the air within the cylinder and the shell of the cylinder provide adequate shielding against cosmic rays. The internal volume of an O'Neill cylinder is great enough to support its own small weather systems, which may be manipulated by altering the internal atmospheric composition or the amount of reflected sunlight.


Large mirrors are hinged at the back of each stripe of window. The unhinged edge of the windows points toward the Sun. The purpose of the mirrors is to reflect sunlight into the cylinders through the windows. Night is simulated by opening the mirrors, letting the window view empty space; this also permits heat to radiate to space. During the day, the reflected Sun appears to move as the mirrors move, creating a natural progression of Sun angles. Although not visible to the naked eye, the Sun's image might be observed to rotate due to the cylinder's rotation. The light reflected from the mirrors is polarized, which might confuse pollinating bees.

To permit light to enter the habitat, large windows run the length of the cylinder. These would not be single panes, but would be made up of many small sections, to prevent catastrophic damage, and so the aluminum or steel window frames can take most of the stresses of the air pressure of the habitat. Occasionally a meteorite might break one of these panes. This would cause some loss of the atmosphere, but calculations showed that this would not be an emergency, due to the very large volume of the habitat.

Attitude control

The habitat and its mirrors must be perpetually aimed at the Sun to collect solar energy and light the habitat's interior. O'Neill and his students carefully worked out a method of continuously turning the colony 360 degrees per orbit without using rockets (which would shed reaction mass). First, the pair of habitats can be rolled by operating the cylinders as momentum wheels. If one habitat's rotation is slightly off, the two cylinders will rotate about each other. Once the plane formed by the two axes of rotation is perpendicular in the roll axis to the orbit, then the pair of cylinders can be yawed to aim at the Sun by exerting a force between the two sunward bearings. Pushing the cylinders away from each other will cause both cylinders to gyroscopically precess, and the system will yaw in one direction, while pushing them towards each other will cause yaw in the other direction. The counter-rotating habitats have no net gyroscopic effect, and so this slight precession can continue throughout the habitat's orbit, keeping it aimed at the Sun.

  • $\begingroup$ You should use blockquote for the pasted content. $\endgroup$
    – JDługosz
    Mar 11, 2017 at 8:27

You must log in to answer this question.