The Emperor (may he live forever) plans to visit the Solar System on a rare royal visit in ten years' time. The Bureau for Interplanetary Tidying have decided that the Asteroid Belt is an eyesore that shouldn't sully the eyes of His Mightiness and needs to be cleaned up. The obvious way to do this is to form them into a single body.

Clearly, moving every single asteroid individually would be incredibly energy expensive. Is there a way to start a domino type reaction so that the asteroids assemble themselves over a period of 10 years or so?

Here is what I have in mind when I talk about a domino reaction. https://youtu.be/5JCm5FY-dEY?t=32

  • 2
    $\begingroup$ You say that moving them individually would be too energy expensive. How much energy do you have in your budget? $\endgroup$
    – Mathaddict
    Commented Jan 25, 2019 at 15:55
  • 3
    $\begingroup$ Eye sore? I think you're overestimating just how close each asteroid is to one another. $\endgroup$
    – chepner
    Commented Jan 25, 2019 at 18:48
  • 2
    $\begingroup$ @chepner - These are bureaucrats. You think they are going to make sense? ;-) $\endgroup$ Commented Jan 25, 2019 at 18:51
  • 2
    $\begingroup$ The asteroids are, on average about 600,000 miles apart, and they're about half a mile wide. If you stand one one, you wouldn't see any others. $\endgroup$ Commented Jan 26, 2019 at 0:27
  • 2
    $\begingroup$ 10 years is a far too short time-frame. Jupiter doesn't even complete a single orbit around the Sun in that time. So chances of getting the process to complete in that time are not good. $\endgroup$
    – kasperd
    Commented Jan 26, 2019 at 16:37

9 Answers 9


How difficult would it be? Unfortunately, your timeline is too tight for any of the easy options to work.

The mass in the asteroid belt is highly concentrated. Between them, Ceres (~30%), Vesta (~10%), Pallas (~8%), and Hygiea (~5%) make up about half the mass of the asteroid belt. Combining these into a single body is (relatively) easy, but slow and energy-intensive: unless you're at least a type II civilization on the Kardashev scale, you're looking at a timeframe of decades or centuries just to get them into the same place through creative use of solar sails. Simply bringing Vesta adjacent to Ceres in under a decade will take on the order of $10^{28}$ joules (equivalent to diverting the total energy output of the Sun for one minute).

Once you've got them together, the only practical way to combine them without scattering debris all over the place is to melt them first. Melting Ceres will require on the order of $2*10^{25}$ joules to heat, and then another $4*10^{24}$ joules for the actual melting. This is on the order of the total solar heating of the Earth in five years. The full asteroid belt will take about three times that (a Kardashev type I civilization couldn't quite do it in the allotted decade). Cooling is out of the question: unless you want to risk thermal shock shattering your new dwarf planet back into an asteroid belt, you're looking at cooling times of millenia.

A domino effect of starting small and letting things work their way up isn't an option. That already happened, and the result was the asteroid belt as you see it today. The clear areas around the orbits of the planets aren't the result of those planets sweeping up the available mass, but the result of those planets ejecting that mass from the solar system. It also takes far too long: current models indicate that the clearing process took around a half-billion years. The asteroid belt exists because Ceres isn't large enough to clear its neighborhood without help.

If you had more time to work with, you could attach an outrageously large solar sail to Ceres and fly it around, using its gravity to throw the rest of the asteroids into unstable resonances with Jupiter, leaving it the only body in the asteroid belt, but this is (a) slow and (b) cheating.

  • 30
    $\begingroup$ I find it amusing that the scale of this question is so great that converting the largest asteroid in the belt into a spacecraft and using it for system-scale gravitational engineering is considered a shortcut here... $\endgroup$ Commented Jan 25, 2019 at 9:10
  • 7
    $\begingroup$ Why melt them? Colliding asteroids is a messy process due to relative speeds, but "combining" two melted asteroids at high speed will be just as messy as colliding two solid asteroids at high speed. The speed is what matters, not the state of matter. $\endgroup$
    – conman
    Commented Jan 25, 2019 at 15:24
  • 12
    $\begingroup$ I don't think the melting part is necessary. Ceres is already large enough that's it's spherical. And once something is that big, the gravitational effects already leads to rock kinda flowing like a liquid. And if the impact speed when you bring to objects together was great enough to send bits flying away at greater than escape velocity, doing it with a liquid won't make any difference. $\endgroup$ Commented Jan 25, 2019 at 16:18
  • 3
    $\begingroup$ @Mark And I think you're misunderstanding how things work at those scales. If you slowly lower Vesta onto Ceres, without them being molten, the rock will still flow and roughly form a sphere in short order. Planets and Ceres aren't spherical because they've been slowly worn into that shape over time or something. They're spherical because at those scales under their own gravity, nothing is strong enough to support any other structure. Sure, you might be left with a decent sized mountain on one side, but it won't be that large compared to the total size. $\endgroup$ Commented Jan 25, 2019 at 22:42
  • 3
    $\begingroup$ If you put Vesta into an orbit around Ceres and slowly lower it below its Roche limit, the debris won't achieve escape velocity ever, simplifying the problem. $\endgroup$
    – Joshua
    Commented Jan 26, 2019 at 0:18

Short answer, no.

The asteroid belt is incredibly sparse, where your chance of seeing another asteroid from the surface of one with the naked eye is pretty low.

There's no simple way to overcome the momentum of every asteroid. They would all require manual adjustments to their orbits, many of them requiring multiple burns to reposition them and collide them with Ceres (the largest asteroid).

  • 14
    $\begingroup$ This is one of Star Wars's greatest crimes against science, turning an asteroid belt into a perilous obstacle course. It's kind of absurd when you think about it; by definition an asteroid belt is one planet's worth of matter spread across an entire orbit. There's just not that much stuff there, for any given location of "there." $\endgroup$ Commented Jan 25, 2019 at 16:49
  • 2
    $\begingroup$ @MasonWheeler More like a small fraction of a planet. The total mass is less then 1/4 the mass of Pluto. $\endgroup$ Commented Jan 25, 2019 at 19:23
  • 1
    $\begingroup$ @TheBlackCat, I guess it's another question (on a different site), but I wonder how they come up with claims like that. In my mind it makes sense that all the stuff large enough to observe might be 1/4 the mass, but what if there's a ton of very small material that we just can't detect? $\endgroup$
    – JPhi1618
    Commented Jan 25, 2019 at 21:31
  • 2
    $\begingroup$ @JPhi1618 I assume that if there was substantially more mass in the belt, we would see it influence the movements of other bodies, particularly Mars. But I'm not an expert. $\endgroup$
    – MJ713
    Commented Jan 25, 2019 at 23:22

Probably not all in one go, but along the way why not have some fun, save your job and give the Emperor something to smile about.

Say you have a sixteen space tugs to work for you - first thing first, order 12 of them to pop out to the Kuiper belt and drag in a dozen matching 100 Km diameter KBO's (essentially comets), whilst this is happening, get the other 4 to grab Ceres, Vesta, Pallas, and Hygiea from the belt , and in the most efficient way place them equidistant around the inner edge of the belt.

Meanwhile the other tugs are draging their loads into equidistant positions between the aforementioned roids.

The tugs can then proceed to push all these objects at optimal speed around the belt to gather mass as they go, gradually increasing the diameter of orbit as they clear debris.

"But, what if the Emperor comes before the job's finished?" I hear you cry in alarm. Worry not, approaching from the Pole of the solar system (on the Sun's axis of rotation) his magnifience will see: - centre of his view, dark save for the sun shining like the twinkel of humour in his eye, then the inner planets, the sharp inner edge of the belt carved and delineated by the 4 planetisimals and the comets then like a great iris with their tales radiating outward.

The wise and benevolent one will recognise (with judicious prompting) the magnificent sight of a likeness of his own eye gazing benignley upon the universe, and revel in his subjects rejoicing and adulation.

  • 4
    $\begingroup$ I like the 'eye' idea. Knowing the Emperor he would probably fall for this. :-) $\endgroup$ Commented Jan 24, 2019 at 21:13

Assumption: There's enough mass in the target asteroid belt to make a planet the desired size.

As amazingly powerful as gravity is, it's quite a bit weaker than kinetic energy unless there's a whomping lot of mass nearby. Consider, for example, the impact of a baseball on a bat. Eventually it falls back to Earth — but for a moment, it's free.

Now let's go to space where we're trying to bring asteroids together and, especially at the beginning, you don't have enough mass to guarantee that the chunks of asteroid that fly off when two come together return to the central mass. What are your options?

You don't have any. This process would be slow.

  • You can use an net the size of your final planet to sweep around the orbit and collect stuff. The amount of energy needed to do this is so great that it begs the question "why are we doing this, again?"

  • You can plop thruster packs on asteroids to speed them up a little, thereby causing them to come together (hopefully gently) with the stuff in front of them. Somewhere around a third the way throughA complete guess you could probably dispense with the thruster packs and wait. The final planet would need to swing around the orbit for a very long time (veeeerrrrrryyyyyy looooooonnnngggggg ttiiiiiiimmmmeeeeee) to sweep up the debris.

  • You can get them moving faster, but this means debris winging off into space. You'll build the basic planet faster, but it'll take longer to sweep up the debris to finish the job (meteors are not your friend, you must clean up).

But, even if you did these, how long before you have a planet? Pressure would build, heat would form... eventually you'd get the right kind of core, etc. You'd likely need to drop a bunch of methane/water asteroids on the thing to kickstart an atmosphere....


Ten years? The uneducated reader would probably never think twice about it, but it's totally unbelievable. I can't see how you could do it in less than centuries, maybe millennia. You're dealing with so much space, so much mass, so much energy... there's reasons why spacey things happen slowly. When they happen fast, they tend to be catastrophic.

Edit: I didn't think to ask Chasly what's to become of this planet(oid)? Derived from our own asteroid belt, it won't be big enough to develop heat or hold an atmosphere and might not be enough mass to stick together as a cohesive ball. In other words, the dirt clods may simply hang together as dirt clods in space until gravitic distortions from the orbits of other planets start spreading them out again. In astronomical lengths of time, they wouldn't stay together very long.

  • 11
    $\begingroup$ +1 re the timeframe it's probably worth mentioning that Ceres orbital period is 4.6 Earth years, so 10 years is only around 2 full orbits for any given asteroid, some of which will be on the opposite side of the sun. $\endgroup$ Commented Jan 24, 2019 at 20:58
  • 1
    $\begingroup$ @KerrAvon2055, that's a very good point. I took a quick look to see if the orbital period of our asteroid belt was mentioned anywhere: it didn't pop up. However, Jupiter's orbital period is 11.9 years and Mars' is 1.9 years. A silly average is 6.9 years. That's a very tight window unless you're really moving things along (bullet #3). $\endgroup$
    – JBH
    Commented Jan 24, 2019 at 21:03
  • 2
    $\begingroup$ "Had I but world enough and time!" Actually, just time alone will do the trick. Use the YORP Effect <en.wikipedia.org/wiki/…> to nudge the asteroids into slowly converging orbits. Essentially, you send out painting crews to paint the asteroids so that the YORP effect slowly moves them into orbits where they will coalesce. (If they are in nearly the same orbit as they make their final approach, the relative speed will be very slow and handleable.) Still not very practical, though. And very, very slow. $\endgroup$
    – Mark Olson
    Commented Jan 24, 2019 at 21:21
  • 3
    $\begingroup$ @JBH Ceres is the largest object in the asteroid belt, so its orbital period (4.6 years) is basically the orbital period of the asteroid belt. $\endgroup$
    – MindS1
    Commented Jan 24, 2019 at 22:28
  • $\begingroup$ @MindS1, DUH! I'm in the middle of my work day and that got completely under my radar. That's what I get for not giving it more than a fraction of my attention. Thanks!. $\endgroup$
    – JBH
    Commented Jan 24, 2019 at 22:57


Self replicating nanites/robots eat the asteroids and make more nanites/robots. The nanites have their own propulsion system to track and eat other asteroids and so on.

See Replicators

Unless you have a exponentially growing system system, you will never clear it in time.

  • 5
    $\begingroup$ You've linked a SG post and still recommend creating replicators. Have you learned nothing from the documentaries? O'Neill disapproves. $\endgroup$
    – McFarlane
    Commented Jan 25, 2019 at 11:37

You've specified that you want a pure domino effect rather than a propelled solution for herding / collecting each asteroid, but this isn't possible without flinging the new planetoid cluster(s) into a new orbit and making an even bigger mess long term. If you just need to sweep some debris under the rug for the duration of the royal visit, this may not be a problem.

Why dominoes won't work

As already mentioned, the deadline for the King of the Cosmos Emporer's visit is only 2* orbits around the sun for a given asteroid. Say you stick a giant butterfly net force field on a medium asteroid so it can gently scoop up all the other asteroids Katamari style and eventually stick to Ceres**, and have a giant space tugboat give it a starting push to a new speed.
* 2-ish, but let's stick with simple math to illustrate the point.
** Saving the biggest for last saves time and energy.

Your snow- asteroid-ball either has to move much faster, completing 3 laps instead of 2 just in time to catch the asteroids originally just behind it. Ignoring acceleration times altogether, you have a 50% increase in orbital velocity, and the orbital radius will therefore increase considerably and start interacting more with Jupiter's gravity. Or you could slow down and complete only 1 orbit (50% drop in speed) and give Mars a new moon (or completely re-texture it*). Except that long before you finished either cycle, the initial snowball mass will have already have moved to its new orbit and be far away from the remaining asteroids at the end of the original orbit.
* Not necessarily bad - another user already mentioned terraforming. Instant habitable planet, just add water and allow to cool 10,000 years.

If you are just worried about tidiness and not creating a single mass, you could possibly pull off domino stacking by instead collecting 8 planetoids (or 6 or 12 or His Majesty's favorite number) of equal size and spacing.

Thinking with Portals!

However, you haven't specified your civilization's exact level of technology beyond some very basic clues - obviously they have space flight and, and enough power to get ships around interstellar distances, but probably not FTL or unlimited energy and fabrication time. The minimum power required for any giant problem is a 9V battery and an SEP field, but I'm assuming that's not an option either.

Therefore, I propose a middle ground solution (which may or may not fit your particular universe) involving a probably reasonable amount of power, and a fairly large orange and blue portal gun pair of wormgates.

A single ship travels to Ceres and leaves one of the gates floating nearby, then slows down a bit with the other one and take the full 10 years zig-zagging backwards around the belt. Each time you reach an approaching asteroid, power up both gates for a few seconds/minutes and lasso it, leaving it parked in formation right next to Ceres***. Power back down, shift the receiver slightly, and move to the next one while the gates recharge. If you need to move faster than 1 crew can handle, send a handful of collector ships with their own gates (either a separate pair or all tuned to the same destination).

*** Crashing everything together would make a debris field that was unsafe to operate your gate in, if not sending pieces back into separate orbits.


With sufficient effort, this could be done. It would require the manual alteration of the orbit of each asteroid.

Bear in mind that the total mass of the asteroid belt is around 4% that of the moon, and that around half of that mass is already contributed by only 4 asteroids. The remainder are tiny and barely significant.

It would be an awful lot of work for basically no gain.


Other posters have focused on two large asteroid belt clearing problems: excessive time and excessive energy are needed to suspend disbelief for the genre of hard science fiction.

Excessive need for space or time seems the lesser issue in a Einsteinian universe.

To suspend excessive space-time, I agree with brichins' idea of using wormgates, a SF technology that harnesses macro wormholes (as opposed to quantum wormholes that seem closer to present physics theory).

I agree, in validly suspended SF, that each end of a sizable wormhole can be placed where needed to move even large planetoids like Ceres. I think all asteroids could be teleported adjacent to each other, with matching momentum for bolting (stone) and welding (iron) into something(s).

Something(s) 4% the size of the moon per Arkenstein XII.

I see no obvious SF reason why wormgates can't teleport other wormgates in a preprogramed pattern around the asteroid belt. What I might call leveraged teleporting, starting with a fleet of cargo starcraft powered by (consensus SF) hyper/warp drive.

Teleport clearing the whole asteroid belt certainly wouldn't be instant, and could take years, because electrons and photons - even in the peripherals of quantum computer circuits needed to solve such an enormous set of calculations - travel at about 95% of light speed (in copper).

One will need a great many individual computers/processors/hardware, and quite at lot of (consensus SF) subspace signaling to project managers in, say, a triangle of bases orbiting the solar system. Debugging committees, even with the help of AI, will have to solve many initial project design issues. That's not going to change in the future.

Mark's delightful analysis of the excessive energetics seems more troublesome.

Einstein seems to allow the cheating of our ordinary calculations of space and time, but not the cheating of mass-energy conservation. Mark probably can calculate how much energy it takes to move objects into higher orbit, if it's 4% of the moon mass, raised from the asteroid belt, to spiral orbits into Jupiter for slow disposal. Jupiter disposal has its own set of problems, but I've read that Jupiter is too small to be set on fusion fire.

I don't know how much recoverable, reusable energy, and lost, heat-dissipated energy it takes to operate a wormgate.

Even if you recover most of the reusable energy after the worm field collapses, to inflate the field, one still has to store a great deal of energy joules in pre-charged capacitors (sized 10^xx farads?).

For story purposes, I think the average reader better comprehends kilowatt-hours (at 20c each) to the equivalent physics joules. Will the Emperor pay for the heat-dissipated energy?

The good news is that your story is premised as tongue-in-cheek SF like Hitchhiker's Guide to the Galaxy. That genre depends more on clever writing than a serious SF suspension.

You've mentioned a bureaucracy, so your story could be one of massively bungling the project, yet in Homer Simpson fashion, all the bungling cancels itself out to result in an unexpectedly successful conclusion.

One way to compress 10 years into a short story is to tell it as a series of dated historic documents, like royal proclamations, emails, radiophone transcripts, corporate invoices, lawsuits, arrest warrants, and news accounts.

You could cover over story holes with puzzled notes from a future historian, writing that he can't find documents to explain how event X occurred improbably - which works with tongue-in-cheek SF.

  • $\begingroup$ A very rough estimate is that, if you use high-efficiency transfer orbits, it'll take on the order of 3*10^27 joules of energy, or about nine seconds of solar output, to throw the asteroid belt into Jupiter. Total flight time for the "main belt" asteroids (those between the 2:1 resonance and 4:1 resonance gaps) is around 9 years. $\endgroup$
    – Mark
    Commented Jan 25, 2019 at 23:03
  • $\begingroup$ Thanks @Mark, very helpful. I estimated the Emperor's lowest cost of energy to do this. 3*10^27 joules equals 8.3^20 kilowatt-hours. Complex anti-matter generators are likely most costly source; solar, the least. 2017 USA solar is 0.046/kWh; cost is declining and more efficient in space. My convenient guess is 0.01/kWh. At 0.01/kWh, 8.3^20 kWh costs 8.3^18 dollars. That's 8,300,000,000,000,000,000, or 8.3 quintillion, or 8.3 million, trillion dollars (USA words, not UK). Sounds like a lot for the Emperor to pay for a tidy view, but in a galactic size budget, maybe that's not much. :) $\endgroup$
    – alstring
    Commented Jan 26, 2019 at 8:23


Since the astreroid belt is exactly that due to the gravity of several planets making sure that any "body" would be pulled apart as it forms, or after it is formed. The whole asteroid "belt" is proof of this effect.

So it is impossible without changing the whole solar system this is in. At which point you can no longer speak of an asteroid belt anyways.

  • 2
    $\begingroup$ Not "after it is formed". A moon-sized planetoid would be far away enough from Jupiter to be outside the roche limit for "a moon-sized planetoid next to jupiter" $\endgroup$ Commented Jan 26, 2019 at 17:44

You must log in to answer this question.

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