A terrible virus has made most of the populations and government officials of the largest superpowers on Earth stark, raving mad, without depriving them of their intelligence.

They have decided that they want to alter the orbit of the moon. Would it be be technically possible for humans to alter the moon's orbit if they wanted to?

For example, if 200 copies of the Tsar Bomba were detonated at the same time on one side, would that alter the moon's course significantly?

If significant enough, could it make the moon collide with Earth? (A borderline undesirable scenario.)

  • 1
    $\begingroup$ Re reading Atomic Rockets page on the Casaba Howitzer projectrho.com/public_html/rocket/… suggests that there are ways to transfer momentum and energy far more efficiently than simply "bombing" the moon. Up to 80% of the energy would be directed in a jet at the Moon, essentially making the Moon itself the pusher plate for an ORION type drive. It would still take tens of thousands to millions of bombs to do the job, however. $\endgroup$
    – Thucydides
    Commented Jul 16, 2016 at 0:46
  • 7
    $\begingroup$ Or you can just use laser pointers. $\endgroup$
    – PTwr
    Commented Jul 16, 2016 at 16:20
  • 2
    $\begingroup$ With current technology, the solution with the most hope for success is for all of these mad people to pray to their gods, who, if they are pleased with the prayers, will send a large stellar object to crash into the moon. $\endgroup$
    – EvilSnack
    Commented Jul 16, 2016 at 17:10
  • 1
    $\begingroup$ "...without depriving them of their intelligence." Are you insinuating they actually have some to start with? $\endgroup$
    – jpmc26
    Commented Jul 16, 2016 at 18:30
  • 1
    $\begingroup$ An extremely watered down and simplified version missing some important revelations, but yes. I read the series, and I quite liked it, although it is a little bit dark. I'm probably going to delete my comment, though. $\endgroup$ Commented Aug 3, 2016 at 15:06

10 Answers 10



The Moon is huge and moving very quickly, at least on human terms.

First, let us calculate the kinetic energy of the Moon. The mass of the Moon is 7.3 × 1022 kg, and it is moving at 1 023 m/s. This gives the Moon a momentum of 7.3 x 1025 kgm/s. Using this equation to convert momentum into kinetic energy we get 3.8 x 1028 J. The Tsar Bomba has an output of 210 PJ, or 2.1 x 1017 J.

Even if all of the bomb's energy could be directed in the opposite direction to the Moon's motion, you would need two hundred billion of them to stop the moon. Of course, that's assuming you could direct all of the energy of these bombs in the same direction.

Simply detonating one of those things on the surface of the Moon could probably do some damage, certainly make a pretty new crater and maybe rain some debris down on Earth, but I don't think nuclear bombs are a viable method for changing orbits of celestial bodies.

EDIT: Here's a great video explaining why changing the orbit of celestial bodies is so damn hard, explained by someone who's much better at explaining things than I. It's in the context of a videogame, but it uses real physics.

  • $\begingroup$ The Moon is huge and moving very quickly, at least on human terms. - Are you sure), do you mean I'm not a human? or those Speed of an Artillery Shell or those U.S. Navy railgun with 2.1km/s. Or rockets. Moon just moves with 1.076 km/s max Moon Fact Sheet - not some thins we call fast - jetfighters move faster then that. $\endgroup$
    – MolbOrg
    Commented Jul 15, 2016 at 22:55
  • $\begingroup$ Wrong velocity... not 1023 m/s... it's km/s. 1 023 000 m/s. Editing the answer for you. $\endgroup$
    – MichaelK
    Commented Jul 16, 2016 at 1:52
  • 1
    $\begingroup$ @MichaelKarnerfors: I'm really pretty sure you're mistaken. I don't think you should make that edit, but feel free to post what your source is, so monodokimes can decide whether (s)he's convinced. $\endgroup$
    – ruakh
    Commented Jul 16, 2016 at 1:54
  • $\begingroup$ Ah, I was... misread the decimal point as a separator. My mistake. $\endgroup$
    – MichaelK
    Commented Jul 16, 2016 at 1:55
  • 1
    $\begingroup$ Still going to edit the post though because the momentum of the Moon is wrong. :) $\endgroup$
    – MichaelK
    Commented Jul 16, 2016 at 1:59

Yes, so long as you will accept a very long leadtime.

As others have shown, trying to do it via rockets isn't viable, but that's not the only way to move a moon.

Instead, lets go with impacts. We can let gravity provide most of the energy.

Go out into the Oort cloud with a nuclear powered spaceship that shoves stuff around, using material taken from it's targets for reaction mass. Go out far enough and the orbital velocities are very low--stopping substantial objects won't take insane amounts of energy. Aim very carefully, bring them to almost a stop and let them fall inward.

Pelt it with enough icy bodies, it's going to move.

The empire that launched this will no doubt be long gone before the first iceball falls, though. You'll probably need Orion + generation ship just to get out to where to do the pushing.

  • $\begingroup$ I like this idea, but you can do better: If get your bodies to swing around the moon and come up in the opposite direction to how they came in, you could rob twice the momentum from the moon as you would by just hitting it! Basically, you send in your material for a gravity assist. (Bonus points if you somehow set up a conveyor belt of icy bodies transferring momentum between the moon and some other celestial body - a gravity assist alone might not be so much harder than hitting the moon, but a conveyor belt could be... difficult) $\endgroup$ Commented Jul 15, 2016 at 23:15
  • $\begingroup$ @MiloBrandt I didn't worry about the optimum trajectory, if you're doing something that elaborate you'll figure out the best way to use the energy. I don't think a conveyor is possible without putting engines on each body you throw. $\endgroup$ Commented Jul 15, 2016 at 23:36
  • 1
    $\begingroup$ Spot on with "yes, but it's going to take a long time". My approach would be to build a solar powered mine and mass driver on the moon: Dig up loads of rocks, fire them off at the right angle as fast as possible, and you can (very slowly) change the orbit. Not a quick way to do it, but definitely within the capabilities of current tech. $\endgroup$
    – The Geoff
    Commented Jul 16, 2016 at 16:31
  • $\begingroup$ This answer is underrated, uses very little energy to modify trajectories enough to use Sun’s gravity to do the job. No Tsar bombs needed, achievable by today’s tech. $\endgroup$
    – Szundi
    Commented Apr 28 at 7:22

You can alter the orbit of the Moon or any planet, so long as you are patient and have access to materials in space.

The trick is transferring momentum from one (or ideally a multitude) of small bodies to a larger one. This is thought to explain how some star systems have "roasters" orbiting the sun, Jovian sized planets orbiting closer to their star than Mercury does to our Sun. During the formation of the solar system, planets were moving through a dense cloud of dust, gasses and other protoplanetary materials. As the Roaster moved through the cloud, it is thought that it accelerated materials out into that system's Oort cloud, wile all the while slowing down and spiralling into a closer and closer orbit.

Just like a spacecraft doing a gravitational assist around planet is "stealing" some of the planet's momentum to speed up, the planet itself is slowing down by a corresponding amount. Since the planet is so much large than a space probe, this isn't measurable by our current technology, but the effect is true, never the less. Obligatory XKCD comic to explain the process.

So to change the orbit of the Moon, we would ideally set out to modify the orbits of NEOs and send them spiralling towards Earth to pass by the moon for a momentum exchange. If we want the Moon farther away, we send the NEO on a crossing orbit that ends up with the NEO being slowed down and falling sunward, while if we want the Moon closer, we send the NEO on a crossing orbit that results in the NEO speeding up.

Because the moon is quite massive and NEOs are in limited supply, ideally the NEO deploys a solar sail after the crossing orbit and uses the solar energy to adjust its orbit to cross the Moon's orbit again (and again, and again) until sufficient orbital adjustments are made.

Of course, this does not mean it is going to be easy or cheap, and the residents of Earth will be rightly concerned by large objects passing very close to our own planet. Nothing like making a mistake in your orbital calculations and having a dinosaur killer (or worse) slam into the Earth....


The mass of the Moon is 7.34767309 × 1022 kilograms.

The Moons orbital velocity is 1 022 m/s.

The kinetic energy is calculated thus: E = 1/2 * mv2

For the Moon this becomes: 3.8 * 1028 Joules

A tsar bomb contains 100 megatons of TNT or 8.1 x 1017 Joules

So it would take about 50,000,000,000 Tsar bombs to stop it dead in its tracks so that it plummets into the Earth. This assumes that 100% of the energy of the blast is somehow channeled into moving the Moon via some kind super efficient super large jet engine.

  • 1
    $\begingroup$ You would probably want to drill into the moon and detonate the nukes under ground and hope the resulting explosion pushes a large mass of the moon in one direction amd call the rest of the mass the new moon...but itd be near impossible to calculate what went where with that many nukes. $\endgroup$
    – Twelfth
    Commented Jul 15, 2016 at 21:41
  • $\begingroup$ @Twelfth drilling and calculating reactive stuff and so on, will be far far bigger then this number - this number is just measurement of kinetic energy changes and not nearly close to what may be achievable with reactive propulsion(way much inefficient, and number will be bigger). Things are getting math nasty there - so no digg that if you not ready to be converted in to rocketscience nerd $\endgroup$
    – MolbOrg
    Commented Jul 15, 2016 at 22:47
  • $\begingroup$ "The momentum is calculated thus" — what you calculate there is the kinetic energy. Momentum is calculated as $p=mv$. Of course, energy is also what you need to compare it to the Tsar bomb. $\endgroup$
    – celtschk
    Commented Jul 16, 2016 at 5:27
  • $\begingroup$ I made an edit to the post... what was meant was kinetic energy. Fixed now. $\endgroup$
    – MichaelK
    Commented Jul 16, 2016 at 10:59
  • $\begingroup$ It does not have to stop dead in its tracks, just slow down. It is currently escaping orbit at some few cm per year. It is only required to bring it back to a decaying orbit. $\endgroup$
    – mckenzm
    Commented Jul 17, 2016 at 11:35

Bombing the moon isn't going to do much. What you will want to do is build a facility that constructs a rocket on the moon, using nuclear reactors as power and the moon itself as reaction mass. The rail gun idea could also work. Of course you would need hundreds, if not THOUSANDS of these installations to impart any significant thrust to the moon in order to alter it's orbit in any timeline useful to a specific human. So yes, with our current tech level we COULD alter the moons orbit, but I doubt we have the resources and global will power to do so, this project would be a global version of the Palace of Versailles (I.e., it would consume like half of the PLANETS GDP to carry out).

  • $\begingroup$ That better be one sick virus then. $\endgroup$
    – Fiksdal
    Commented Jul 15, 2016 at 16:09

Nope again.

You can't just push the moon tangentially. Imparting a small tangential velocity, say 1 m/s, would not build up to a large offset over time. Instead, tidal forces would circularise the orbit which would then just be a little higher. 1 m/s is 0.1% of the moon's orbital speed, and would result in the orbital radius increasing by about 0.2%, or 750 km. And even that would require an impractical amount of energy.

You can't use solar power to launch rocks as reaction mass in any effective way. The momentum of the rocks is merely the converted momentum of solar photons. Those have already been pushing on the moon for billions of years with negligible effect. The best you can do is cover the moon in a perfectly reflective material, which would impart twice the momentum of the incident radiation. When you do the sums you end up with a negligible acceleration of approximately 1 \times 10^{-15} \text{ ms}^{-2}}.

And finally, you can't lob Oort cloud comets at the moon. There is certainly enough material to shift the moon -- estimated at perhaps five earth masses. At the distance of the Oort cloud you would only have to impart a velocity of about a third of a centimetre per second to shift a body by 1 AU on its return to the sun (which will take about 1.4 million years). But shifting, say, a moon's worth of mass by that small amount is still the equivalent of ten million tonnes of kerosene burned at 100% efficiency.

Let's also not forget that we can't get a very efficient gravitational slingshot from the moon when our comets arrive in the inner solar system. Ideally we want to be coming in from the opposite direction to the moon relative to the sun. We only get maximum effect at a particular time of the lunar month, and at a particular time of year. That gives us a lot of computational complexity too, especially when we also consider the gravitational effects of the large outer planets which may perturb our incoming material.

But let's even suppose we can overcome all that, and we have a nuclear engine and can use cometary material for reaction mass. We still have an insurmountable problem -- the dispersion of the Oort cloud material. Estimates of the number of comets are as high as a hundred trillion. The cloud could extend as far as half way to the nearest star. But let's suppose that the comets are all conveniently waiting for us on the surface of a sphere with a radius of one light year. That gives an average separation of over three million kilometres between comets. We have to travel a hundred trillion times that distance to visit them all. If we assume a healthy speed of 60 km/s -- much faster than the Voyager spacecraft's Jupiter slingshot speed, it will take nearly two hundred billion years to visit them all, and that's assuming we don't even have to stop at each one. Oh, and we need power to maintain that speed since it is much faster than the sun's orbital speed at that distance.

Alternatively, if we have to stop at each comet, but can achieve a continuous acceleration of 1g to speed up and slow down between comets, we can get the total visit time down closer to one hundred billion years. That's still vastly longer than the moon is going to exist.

In short, when you actually do the numbers, no method of moving the moon by a significant amount is really feasible.


Sure. How many zero's are in your check book? We have had the technical know-how since the 60s.

We had proven the technology to reach the moon in 1958 : Luna 2, and had learned how to split the atom in 1917 : Ernest Rutherford.


At the height of the Cold War, the combined number of US and CCCP nuclear warheads was ~50k. According to another answer here, you'd need about 100 million of the largest weapon ever deployed: the Tsar Bomba (note, which was only 50MT; it's theoretical maximum is 100MT).

And according to those calculations, that's what you need to bring the moon to a dead stop. Having "significantly" altered the moon's orbit happens way before then.

So, I'll reiterate: it's not a question of technology, it's a question of your resolve and MONEY.

Will you take a check?The Core

  • 1
    $\begingroup$ Time, money, and big-dumb-rockets will get the job done; no hand waving required. It's not like you're trying to build a working fusion reactor. $\endgroup$
    – Mazura
    Commented Jul 16, 2016 at 1:22
  • $\begingroup$ Sea Dragon could theoretically launch 550 metric tons into LEO, so bringing several hundred Tsar Bombas into orbit would be a breeze.... $\endgroup$
    – Thucydides
    Commented Jul 16, 2016 at 4:21
  • $\begingroup$ theoretical maximum is 100MT it's not theoretical maximum, this was maximum for that construction. Theoretical maximum is not limited for thermonuclear bomb more there $\endgroup$
    – MolbOrg
    Commented Jul 16, 2016 at 17:48

It is certainly possible, however the "naive" approach is way out of our reach.

Stopping the moon is a crazy, but that isn't even necessary. Just like you wouldn't want to stop an asteroid that is going to collide with Earth (nor would you want to nuke it). It is however perfectly feasible (well, in theory, assuming space travel technology advances uh... yet a tiny bit) to land on, or orbit around it, and use some method of creating a small but continuous force, for example by melting ice on one side with a solar reflector or such to get a repulsor effect. That doesn't stop the asteroid, not does it destroy it. But given some 3-4 years of time, it is enough to deviate the course ever so little that with some luck it will just fly by.

A moon's (or any celestial body's) orbit is in some way an equilibrium between gravitational pull and centrifugal force. Like all equilibriums, it is moderately fragile. While the amount of force that is needed to permanently disturb this equilibrium in "finite time" (read as: before the next ice age) is still huge from our point of view, that's only because humans are so small and weak, comparatively. The force needed is however still many, many orders of magnitude smaller than for "stop the moon".

I have no doubt that if one was crazy enough to try and had a major nation's funds, it would generally be possible to give the moon just enough of an inwards impulse (hardly noticeable, but steady) so it will, maybe over the course of 100-500 years, crash into Earth. A bomb would probably not be that great of an idea, but maybe some kind of huge nuclear-driven rocket motor on the back side, why not. It could vaporize moon dust as propellant, as a crazy idea. There's enough material readily available.

The thing with equilibriums is that once they get off balance, things go by themselves. Throwing a small pebble down the wrong mountain side can create an avalanche that buries an entire town.

  • $\begingroup$ See my comment higher up, it is not in equilibrium, it is moving away very slowly. $\endgroup$
    – mckenzm
    Commented Jul 17, 2016 at 11:38
  • $\begingroup$ I think you're confused about the "equilibrium" involved here, if you perturb the velocity of an object in a circular orbit that just leads to a slightly different elliptical orbit (see the diagrams I posted in this answer illustrating the orbits that result if you fire a cannonball a bit faster or slower than the velocity needed for circular orbit), the orbit might change on verrrry long timescales due to tidal forces but certainly not in 100-500 years. $\endgroup$
    – Hypnosifl
    Commented Jul 24, 2016 at 22:16

The Nope answer is right concerning a bunch of Tsar bombs detonating near the moon; it is completely wrong regarding the capability of nuclear weapons to change its course.

Some misconceptions: 1. There is the assumption that we can't build more powerful nukes and in greater quantities than we could in the 60s. 2. The assumption that you somehow need to completely change the kinetic energy of the moon ( a very "convincing" large number is given by multiplying mass and velocity).

More will follow in a few hours.


The answer of Damon is close to what I have in mind, never mentioned in the "orthodox" , or rather conventional thoughts of the bulk. That was my point 2 in the first answer. If you somehow added a 10cm/sec or 0,1m/sec vector towards earth, you would think this is small but would have devastating effects in the course of a thousand years, which is more familiar to our timescale.

People are used to say huh.. what's the mass of the moon ah ~10^22 kg, wait what it is its speed ahhh about 10^3m/sec, soooo you would have to add about 10^ 28 joules to the kinetic energy of the moon, right?

WRONG! This is what happens when you reiterate answers "taught" and not looking scholastically into the details. (Been caught myself doing exactly that many times) You can put just 10^22 or 10^21 joules to the kinetic energy of the moon in such a direction that the moon would acquire a lateral velocity vector of about say 1m/sec or 0.1 m/sec. This would be enough to move the moon the equivalent of its current distance from the earth towards that lateral direction in the course of 100 to 1000 years.

And some would say: now If I had infinite money and time, sure why not. Pointing sarcastically to the fact that it maybe theoretically possible, but not practically, because it would require much time and money, right?

VERY WRONG AGAIN! Thank you very much. The "megatonnage" of the nuclear arsenal of just 2 countries developed in practically 2 decades by Russia and U.S.A. was about 12 000 megatons or about 5x 10^19 joule. But people this was an arms race half a century ago. Now the whole planet is into building nukes and with a better know-how, meaning much shorter times building a certain amount of destructive power and far more powerful warheads.

If an arms race were to break out again be sure , the "superpowers"(U.S.A., Russia, China, U.K. Canada, Australia France, Korea,Iran, Pakistan , India etc) would build 10^ 22 "joules of destruction" in far less than a century with a hardly noticeable effect on their budgets, or alternatively they could build the same force in 150 or 200 years, seemingly a lot but nothing in the grand scope of things.

So, can we build enough nukes to change the course of the moon or other heavenly bodies. Sure it's a certain stretch but very doable within the capacities of the current technology.

P.S. The release of the 10^22 joules by an explosion would not necessarily mean that this would translate into a velocity to a certain direction. The help of a supercomputer would be needed in calculating distance and other parameters so we would maximally utilize this energy as motion.


You must log in to answer this question.

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