There are several questions similar to the one I'm asking, but I'm not sure this fits exactly with them.

Basically, I have two habitable planets, for the sake of this, let's just assume they are clones of Earth. Can they basically stick together as they orbit the star, or at least have the appearance of doing so?

Let's say their distance from each other in this configuration is far enough away that they won't just rip each other apart - so farther than the Moon and such.

Is this a physics impossibility, is it possible but through some alteration in this scenario (say a binary or trinary star system) or is it only possibly through magic handwaving?

  • 1
    $\begingroup$ What do you mean by "stick together"? How close do they have to be, and how consistently? $\endgroup$ Jan 7, 2022 at 19:24
  • 1
    $\begingroup$ @LoganR.Kearsley I lack the understanding of orbital mechanics to articulate this. Basically I need two planets close enough to each other to allow a budding spacefaring civilization to travel easier between the two worlds than something more difficult like Earth and Mars. $\endgroup$ Jan 7, 2022 at 19:31
  • 5
    $\begingroup$ By that definition, you're talking about a binary planet. Those are real, and by some definitions Pluto and Charon are an example. $\endgroup$
    – KeizerHarm
    Jan 7, 2022 at 19:33
  • $\begingroup$ I think you are asking if a 3 body system be stable? Take a look at: astronomy.stackexchange.com/questions/22032/…. $\endgroup$
    – JonSG
    Jan 7, 2022 at 19:34
  • 1
    $\begingroup$ Nothing physically precludes a binary planet AFAIK but the $64,000 question is how would those planets have formed? It seems extremely improbable for two nearly-equal sized accretions form in such a way that they orbit each other for the eons it would take for planets to condense from them both. $\endgroup$ Jan 7, 2022 at 19:53

3 Answers 3


Double stars are quite common in real life, so double planets are quite common in science fiction.

For example in "Competition" by E.Mayne Hull, Astounding Science Ficiton June 1943 http://www.isfdb.org/cgi-bin/title.cgi?46719 Evana finds that the planet Delfi II has a large moon like companion:

Memory came that the educational talks on the space freighter had proved it wasn't a moon at all, but a dead companion planet as big as Delfi II; and that once, long before man came, there had been life on it--of which obscene remnants remained.

According to Wikipedia, astronomers believe that double planets should exist, but would probably be rather rate.


The two planets in a double planet would not have to be exactly the same mass to both be habitable for humans.

Stephen H. Dole discussed the requirements for a human habitable planet in Habitable Planets for Man, 1964.


Onpages 53 to 58 Dole decided that a human habitable planet would have to have a surface gravity of less than 1.5 g, and thus would have to have a mass of 2.35 Earth mass, a radius of 1.25 Earth radius, and an escape velocity of 15.3 kilometers per second,vor less.

Dole decided that the smallest planet that could retain an oxygen rich atmsphere for geological periods of time would have an escape velocity of 6.25 kilometers per second, and thus a mass of 0.195 Earth mass and a radius of 0.63 Earth radius, and a surface gravity of 0.49 g.

Thus human habitable worlds would have a mass range of 12.05 times and a range of radius of 1.984 times.

But Dole believed that a planet would have to be much more massive than 0.195 Earth mass to produce a breathable oxygen rich atmosphere. Dole decided that a minimum mass of 0.4 Earth mass would be needed, corresponding to a radius of 0.78 Earth radius, and a surface gravity of 0.68 g.

That would give a mass range of 5.785 times, and a radius range of 1.6025 times, between the largest and smallest human habitable worlds.

Whatever the exact figures, a double planet doesn't have to be a twin planet with planets exactly the same size for both of the planets to be habitable for humans. There is a degree of variation in the possible sizes of human habitable planets.

So two worlds which some people might consider to be similar enough in size to count as a double planet, and that other people might consider to be different enough in size to count as a planet and its satellite, might both be within the size range where they could be habitable for humans.

Dole discussed the relationship between a habitable world and its natural satellite(s) on pages 72-75. He discussed the case where a habitable world is a planet and has a natural satellite much smaller than it. He discussed the case where a habitable world is a very large natural satellite or moon orbiting around a much larger planet.

And he discussed the in between case of two planets of similar size orbiting around each other, a double planet.

A planet and its natural satellite will have tidal interactions which slow down their rotation, in some cases slowing down the rotation of a world so much that it becomes tidally locked to the other world, with one side perpetually facing the world.

On page 58 to 61 Dole concluded that days and night that are too long would make a planet uninhabitable for humans. Dole decided that a planet habitable for humans should have a rotation rate that gives one complete rotation longer than 2 or 3 hours and shorter than 96 hours (4 Earth days).

Of course the upper limit of rotation period for a human habitable planet might be longer or shorter than Dole believed.

So a possible planet which was part of a double planet would not be habitable for humans - or lifeforms with similar requirements - if its rotation rate was longer than 96 hours -or hwatever the actual limit is.

Thus a double planet where one or both of the planets have had their rotation periods slowed to take longer than 96 hours, or whatever the real limit is, would be uninhabitable for humans and beings with similar requirements.

So somone designing a ficitonal habitable binary planet would have to calculate that the rotation periods of the two planets would not be slowed down to become longer than the limit for habitabiity.

The PlanetPlanet blog has a section called The ultimate Solar System devoted to designing fictional star systems with as many habitable planets as is scientifically possible.

In the the ulitmate solar system part 5 a system is designed with two binary planets (four planets total) in each orbit in the habitable zone, to squeeze in as many habitable planets as possible.


So Sean Raymond, the creator of PlanetPlanet, clearly believes that binary planets can have stable orbits and can be habitable.

  • $\begingroup$ Fantastic answer, and thank you especially for the additional resource of planet planet. This has me wondering now if you could have a habitable contact binary - probably not without some kind of magic or hyper advanced tech. $\endgroup$ Jan 8, 2022 at 18:11
  • $\begingroup$ @WasatchWind The tidal interactions between the two bodies would usually drive them farther apart over time, long before they became habitable. The two planets would also be near their Roche limits if in contact. And it might take hyper advanced tech or magic to solve such problems. If I remember correctly two planets in contact are a setting in "The Devolutionist" by Homer Eon Flint, Argosy, July 1921. hilobrow.com/2013/06/20/the-devolutionist-18 $\endgroup$ Jan 9, 2022 at 18:28

It is entirely possible for the system that you describe to exist, it is not prohibited by physics. However there are a number of caveats to that:

It would seem very unlikely that such a large binary planet system would form naturally, although it is not impossible.

Once formed the long term stability of the pair would be in question. The planets would have to be specifically positioned so that they were far enough away from the central star so as to prevent instabilities building up and significantly changing their orbits over time.

The orbital period of the binary planets around one another would also have to be Just right. At 250,000 miles the tidal forces would be colossal compared to those we see from the Moon (perhaps tides of many tens of metres) and would leach energy from the system causing the orbits to decay over sufficient time.

  • 1
    $\begingroup$ Would there still be tidal forces if they were rotationally locked to each other? $\endgroup$
    – KeizerHarm
    Jan 7, 2022 at 22:45
  • 1
    $\begingroup$ Yes there would, however it would be a static bulge like a frozen tide. For this to happen either they have to be very close to achieve a normal "day" length (so geostationary orbit) or the rotational period must slowed down. For the Earth - Moon system at the Moon's distance the rotational period would be 28 days, with intermediate values at closer approaches. $\endgroup$
    – Slarty
    Jan 14, 2022 at 22:53

Oceans and tide issues

As was put in Slarty's answer, keep in mind the tidal forces will be enormous. These huge forces will cause the tidal lock, eventually. Suppose your planet would have its own rotation, at first: large masses of water would move around, causing floods. Eventually, rotation of each planet is stopped and the two planets will be in tidal lock, with a permanent high tide on one side of the planet.. On the other side, you'd get a land mass.

Maybe inhabitable, but quite uncomfortable

Your planets will both have a very long day-night cycle. When they don't rotate around their own axis, they only rotate around eachother, which will take longer. As a result of their proximity, prolongued solar eclipses will cool down the ocean side, on a daily basis, resulting in strong convection currents and heavy storms along the coasts.

  • 3
    $\begingroup$ Tidal forces aren't enough to keep all the water on a planet on one side. Also, rotational locking won't result in there being a "cold side" to the planet, because the binary pair would still be rotating with regards to the star, just not to each other. $\endgroup$
    – jdunlop
    Jan 7, 2022 at 23:51
  • 1
    $\begingroup$ It's 7 minutes now. I'm still working on this answer text.; Please allow me some editing time before downvoting it. $\endgroup$
    – Goodies
    Jan 7, 2022 at 23:57

You must log in to answer this question.

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