This is just an idea I have been toying with.

Taking Earth as an example with 7 days in a week, my idea would that there would be 7 moons in orbit around the planet. They would not necessarily share the same orbit, but they would orbit in such a way that only one is visible each night. That way each night the people see a different moon, on the 8th day it goes back to the first moon as it has orbited close enough to be visible again while the other 6 are currently too far away to be visible.

Would there be any way this could happen in reality? Even if it may be unlikely, is it at all possible?

EDIT: After some further research I have come across information about the Trappist 1 system. It does a similar thing to what I am talking about though with planets instead of legitimate moons. It also ties in with the point made about orbital resonance made below. Might be an interesting starting point for anyone looking to do a similar thing in future

  • 1
    $\begingroup$ short answer, no long answer if they are small and close to the planet, sorta $\endgroup$
    – Topcode
    Sep 3, 2020 at 17:14
  • 1
    $\begingroup$ The "too far away to be visible" doesn't work. Mars isn't all that big, and is easily visible even though it's millions of miles/km away even at closest approach. Mercury is even smaller, and is visible if you have a clear horizon. Plus any object in such a highly eliptical orbit would eventually escape. $\endgroup$
    – jamesqf
    Sep 3, 2020 at 17:21
  • $\begingroup$ @jamesqf - you make a fair point about Mercury and Mars, yes there are 'visible $\endgroup$ Sep 3, 2020 at 22:26
  • $\begingroup$ @jamesqf - I take your point about Mercury and Mars being visible, perhaps to clarify my question it may be more accurate to talk about one predominant moon per night while the others are unremarkable distant points of light $\endgroup$ Sep 3, 2020 at 22:30
  • $\begingroup$ @LearnedStudent: But it would take months for anything on a highly eliptical orbit to recede far enough to become a distant point of light. $\endgroup$
    – jamesqf
    Sep 4, 2020 at 17:14

8 Answers 8


This is not possible because satellites in different orbits will necessarily have different periods. They will all move across the sky in different speeds, so they will catch up to each other every now and then.

Having multiple satellites in the same orbit is complicated - such arrangements are not stable on a geological timescale, and are too improbable to happen naturally.

Things less intuitive and stranger than you propose do happen, though. Orbital resonance and horseshoe orbits are a thing. A long sequence of horseshoe orbits might be close to what you want.

  • 1
    $\begingroup$ Hmm interesting, even though the first part of your answer obviously tells me it's probable not a likely occurrence I do quite like the idea of where some may catch up with each other. I assume that would mean some nights there may be only one moon, next night might have 4, maybe the next 2 and so on - I almost like the sound of this better if I understand your point correctly. Thanks for the links as well by the way, the one on Horseshoe Orbits is very interesting. As you mention as well, satellites in the same orbit is the least likely of all possibilities so I'll avoid that route $\endgroup$ Sep 3, 2020 at 19:44
  • $\begingroup$ Satellites in different orbits can have the same period. Some may orbit in a different plane and some may have eccentric orbits $\endgroup$
    – Slarty
    Sep 3, 2020 at 20:16
  • 1
    $\begingroup$ @Slarty satellites in different planes or with different eccentricities will stabilize in resonant orbits too. Over geological scales they modify each other's orbits. $\endgroup$ Sep 3, 2020 at 20:20
  • 1
    $\begingroup$ It can take a very long time indeed. The hypothetical situation being proposed here might well not be at equilibrium. $\endgroup$
    – Slarty
    Sep 3, 2020 at 20:33
  • $\begingroup$ @Slarty - what do you mean by it being in equilibrium here? What effect or difference would it have on the proposed system? $\endgroup$ Sep 3, 2020 at 22:33


It will not work from every spot on the planets surface, and you may catch glimpses of the other moons, but at least at the equator and tropics it should be possible.

The moon I've used for this is smaller than our current moon, at 5000 metric tons. It comes much closer than our current moon, so should be about as visually noticable.

Assuming Earth-Centered-Earth-Fixed coordinate system, this orbit should work for you:

a = 154600km +/- 0.1% (Semi major axis)
b = 80000km (Semi minor axis)
With phases of 2n/7 * pi for v.

The apoptosis is 286,892km - which is a bit further out than our current moon. The periapsis is 22,308km, which is an altitude of 15937km. This is above the ISS, above space shuttle, but below GPS satellites and geostationary satelites.

enter image description here

You'll be able to see at least 5 moons during the day, forming a ring from east, up to the midday sun, and down to west. They will start to disappear in the late afternoon to the west, with the last setting just after sunset, with that nights moon rising from the east at some point during the night. The moon will follow east to west rapidly, and then set in the west. A few other moons will start to rise just before sunrise, and then sunrise will occur in the east. These sunrise and sunset moons will be much dimmer than the main moon at night.

This would be stable for hundreds possibly thousands of years, but not millions.

The reason for the +/-0.1% is I haven't factored in the advance necessary to keep the apoptosis pointed towards the sun so the closest approach is always at night as the planet goes round the sun (I just used ECEF coordinates cause they're easier) - Somewhere between 153000 and 156000 will be your sweet spot, but I suspect you don't need it to this accuracy.

(edited - actually run the equations)

  • 1
    $\begingroup$ "Closest approach to earth and furthest distance from sun occurring at the same time": how does this work? Let's say this is true in June, how would you arrange it so that it is also true in December? $\endgroup$
    – AlexP
    Sep 3, 2020 at 19:08
  • $\begingroup$ Would this even be a gravitationally stable configuration? $\endgroup$
    – cowlinator
    Sep 3, 2020 at 21:19
  • $\begingroup$ But if I understand your orbit correctly, the people on the planet really aren't going to be able to see the moons during much of the night, as they'll be eclipsed by the planet. $\endgroup$
    – jamesqf
    Sep 4, 2020 at 17:21
  • $\begingroup$ @jamesqf The orbit is designed so that a single different moon is visible each night. The other 6 moons are eclipsed by the planet by design. $\endgroup$
    – Ash
    Sep 4, 2020 at 17:24
  • $\begingroup$ @Ash: I think you misunderstood. 6 of the moons would be on the day side (again, if I'm interpreting the picture correctly). The 7th will be on the night side, and will be visible at dawn & dusk, but otherwise will be in the planet's shadow. This is similar to the way the ISS is only visible low in the sky, then enters the Earth's shadow. $\endgroup$
    – jamesqf
    Sep 5, 2020 at 21:07

You're going to need:

  • small, very faint moons that cannot be seen during the daytime. What we're counting on is that these small, dim moons tend to get lost in the morning/evening glare in their crescent phases.
  • Some sort of regular orbital perturbation that advances the line of apsides of the moons precesses with the same period as the planet, so that the semimajor axis points at the sun.
  • Some sort of super-technological or supernatural mechanism to place these moons and keep them in synch; I don't think this setup is going to be stable on historical timescales, much less astronomical ones.
  • This won't work for the entire planet. You'll get the best results for an observing site directly under the periapsis.

GeoGebra Graph: 7 Moons in 7-day elliptical Orbit

7 moons in 7-day orbit, with mean a

In the above image, we have 7 moons in the same orbit, simulated in GeoGebra. I've used Earth's mass and Earth's radius. An orbit of semimajor axis 154,473 km and an eccentricity of a shade under 0.752 results in an orbit with a period of 168 hours, where each moon spends 6.5 days on the Sunward side, and 12 hours on the night side. The moons are separated by 24 hours on this orbit.

The horizon for the for a chosen observer, who is on the Equator directly under the periapsis also shown, turning red when the observer is sunside, and green when the observer is nightside.

Schematic of Sky View from observer Location - Link to Animation

Schematic of Sky View from observer location.

For this observer at sunset, he has line of sight to up to three of the moons; the Current Night's moon for will be directly overhead, with the next two moons in the orbit setting.

The Current Moon will appear to drift westward initially, but will start moving Eastward, as the next two moons in the rotation slip below the western horizon.

As Midnight occurs, the Current Night's moon will be the only one visible, and it will be at the meridian again.

As the night progresses, the Current Night's moon will overshoot the meridian, and then drift westward again to be overhead at sunrise, at which point there will be line of sight to the three previous moons in the rotation.

As midday approaches, all of the moons are in the daytime sky,

I don't know of a mechanism that would allow this to happen naturally, and your standard non-keplerian perturbations are probably going to pull these moons out of their perfect positions in short order, and then it's orbital Thunderdome.

  • $\begingroup$ +1 for the animation. :-) $\endgroup$ Sep 7, 2020 at 7:32
  • $\begingroup$ As I wrote in my own answer below: Something that would make such a thing much easier is if your week has just 6 days. Then you could have just three moons sharing the same orbit, which is three times longer than your planet's rotation. If the moons all rotate in a 2:1 spin-orbit resonance, they would show alternately one and the other face towards your planet, making it look as if there are six different moons. Such a configuration could just possibly - if highly unlikely - happen by accident and might even be stable for a few hundred thousand years. $\endgroup$ Sep 7, 2020 at 7:34

Well first of all it depends on the viewpoint. Is this just from one side of the planet, or is it from every point? If it is just one side we and the moons have an elliptical orbit then yes. Imagine 7 moons that all have a very elliptical orbit and circle the planet once every week. On the side of the ellipse that is furthest from the planet they move slowly. Then, as they approach the planet, they speed up, and once they reach the other side they zoom by in a single night. They could even share the same orbit. I don’t think it would be possible from viewpoints though.

  • $\begingroup$ This is actually very close to what I had in mind actually, they would only be visible from one side of the of the planet, specifically the side at which it is night time. It is unlikely I will actually try to use 7 (I plan on my world running on a much different calendar) so hopefully this should simplify it further. As you mention above though, I imagine the moons would require a very elongated elliptical orbit for it to work properly? $\endgroup$ Sep 3, 2020 at 19:33

I do not believe this is possible. Assuming a 24 hour orbital period for the planet and 7 day period for the Moons.

In a single night an observer would be able to see roughly half a hemisphere, so on average half of the moons should be visible.


Suppose the orbital period of the moon were exactly equal to the orbital spin of the planet. In such a case, the moon would always orbit exactly over the same spot on the planet, day and night. It would always be seen on one side of the planet, never seen on the other side of the planet, and in the same place both 'day' and 'night'. Now, slightly slow down or speed up the orbital period of just the planet, not the moon. The moon would slowly disappear from sight, and remain out of sight, until the orbital spin of the planet 'caught up' to the orbital period of the moon again.

Now, place seven moons spaced equally around the planet, in the same orbit, so that as one disappears and 'sets', another one 'rises' in the other side of the sky. It would not be 'every night another moon', but it would be 'only one moon in the sky during any particular night', never two moons appearing in the same sky during any one night, and always cycling through the same sequence of moons.

That is, it might be the same moon day and night for, say, a week, then another moon for the next week, and so on as each moon 'creeps' across the sky in turn.


Put all of the moons in a polar orbit, so the orbit of the moon is not on the same axis as the spin of the planet. Like many satelites in polar orbit, each moon would cover a different 'slice of longitude' on each orbit. Seven moons, in seven polar orbits, no moon ever visible at the same time as another moon.



It would require all seven moons to share the same orbit, which probably isn't a stable configuration and would have to be artificially made - there's practically zero chance that it would occur naturally. Let's suppose, for the sake of argument, that some advanced alien race made it so in the very near past, geologically speaking, and maybe added some mechanisms to stabilize the orbits.

What you need is for the moons to be in an orbit that is exactly 7 times as long as the planet's rotation period - let's call it a septogeosynchronous orbit. During the night, one moon will move more or less from one horizon to the next, and the next night, the next moon will do the same. During daytime, all the moons will be hidden behind the planet. This means that the configuration only works on one side of the planet - on the other, they will see multiple moons in the sky.

Something that would make such a thing much easier is if your week has just 6 days. Then you could have just three moons sharing the same orbit, which is three times longer than your planet's rotation. If the moons all rotate in a 2:1 spin-orbit resonance, they would show alternately one and the other face towards your planet, making it look as if there are six different moons. Such a configuration could just possibly - if highly unlikely - happen by accident and might even be stable for a few hundred thousand years.


My answer to this question:


May help:

Another possibility is having a planet with many moons of almost totally identical size which are equally spaced in a ring around the planet. The moons all share the same orbit, equally spaced.

The PlanetPlanet blog has a post:


It references a scientific paper showing that a number of equal mass and equally spaced objects can share the same orbit and be stable.


According to their calculations seven to forty two objects could share the same orbit.

Thus your planet could have seven to forty two equal moons equally spaced along a single orbit.

Of course it seems extrmeely improbable that such a system could form naturally, so perhaps an advanced civilization assembled those mooons and placed them in orbit around the planet.

How will the people of the planet tell the difference between one moon and another if they have similar masses and thus diameters?

If the moons are less than a few hundred kilometers in radius they can be irregularly shaped and perhaps each moon has a different irregular shape. They will have to have very low orbits for their shapes to be clearly seen from the surface.

If the moons are larger than a few hundred kilometers in radius they will probably be roughly spherical and look round to the eye. But they might have maria as on Earth's moon and thus different patterns of light and dark on each moon.

And possibly the hypothetical civilization which hypothetically created that moon system helpfully coated each moon with materials of a different color.

And possibly the hopothetical ciilization might have built giant flashing beacons on each moon, with each moon's beacon flashing a different color or a different pattern.

if there are 7 to 42 moons equally spaced along a circle 360 degress around the planet, the average angle between moons will be be about 8.57 degrees for 42 moons to 51.42 degrees for 7 moons.

Thus if an arc of more than 8.57 to 51.42 degress along the orbit of the moons is visible in the sky of a location on the planet, more than one moon should be above the horizon at any one time. On the sea or a plain, the visible sky should cover 180 degrees, and so about 3 or 4 moons if there are 7 total moons, to 21 moons if there are 42 total moons, should be visible at any one time.

Possibly the story could be set in a deep valley or an urban canyon with only a limited view of the sky.

Making the moons as few as possible would make them farther apart along their shared obit and reduce the problem of having more than one moon visible at a time.

If the moons are very small they will have to have very low orbits and thus they will not be visible from a full hemisphere of the planet at any one time.

Phobos, the inner moon of Mars, is so low it cannot be seen from an entire hemisphere of Mars at once.

Because it is close to the surface and in an equatorial orbit, it cannot be seen above the horizon from latitudes greater than 70.4°


So phobos is visible from a circle 140.8 degrees wide, intead of 180 degrees wide, on the Martian surface.

I note that such a low moon would probably have an orbital period much less than a day of the planet.

And it is possible that the atmosphere of the planet is thicker than that of Earth (but not too thick for humans or natives of the planet with similar requirements. See Habitable Planets for Man Stephen H. Dole, 1964 https://www.rand.org/content/dam/rand/pubs/commercial_books/2007/RAND_CB179-1.pdf[5]).

You may need to increase the total density of the atmosphere, and/or the amounts of various gases and atmospheric dust, and/or air pollution, to make it scatter a lot more light, while still being breathable for humans or natives with similar requirements.

If the atomosphere scatters enough light, the moons may not be visible during the day, and maybe even visible during only part of the night. Possbily even at night the atmospheric haze hides the moons near the horizon and they have to be high in the sky near the zenith, to be visible. And of course at night the Zenith may include the planet's shadow.

So possibly the area of night sky where the haze is thin enough to see the moons is less than 51 degress wide and there are only 7 moons spaced 51.42 degrees apart. So there will never bee two moons within the visibility zone around the zenith at the same time. And possibly the moons orbit at such a distance that the planet's shadow fills a large part of that region around the zenith near midnight, so that a moon wiil be visible for much less than all 51 or fewer degrees.

I think for this to work the moons will have to orbit at near, but not exactly at geosynchronous orbit, so that the 51 degree cone of the night sky where they can be seen will not sweep over two or more moons during the night. But the orbit of the moons has to be far enough from geosynchronous orbit that the 51 degree or less cone of visibilty will sweep over a different moon each night.

The length of the planet's orbit around the star may also have to be adjusted to make this work.

The mass of the planet, and the distance of the moons from the planet, will determine their orbital periods. And you will need to work out a day length and an orbital period which will result in a different moon appearing in the night sky each night for a period of about a week or whatever.


You must log in to answer this question.

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