Earth rotates around the sun and around its axis (A North, South axis) giving us days, nights and seasons.

Are there any known planets that rotate with an axis pointing toward its sun so that one half of the planet is always facing it (the other half is always facing away from it)? Like Uranus, but not switching which side is facing the sun every 6 months.

(Would this even be theoretically possible?)

  • $\begingroup$ Earth's seasona are a result of the axus tilted by around 23 degrees. Look at Venus with almost no tilt, there are no seasons. She has another unique behaviour tho. $\endgroup$ – ott-- Sep 19 '16 at 18:16
  • $\begingroup$ @ott-- Venus is very... retro... $\endgroup$ – user2293 Sep 20 '16 at 0:19
  • $\begingroup$ Because you want the pole pointed at the star, the answer is no, Closest thing is referred to as tidally locked, it is a real thing, as a reasonable example, look at the moon relative to the earth. orbital rotation (almost)equal to axial rotation. $\endgroup$ – LaserYeti Sep 20 '16 at 4:11
  • $\begingroup$ There are some unlikely situations where you could make it work, like a planet that got hit by a large Iron rich meteor and as a result, it has more Iron on one side than the other. The heavy side of the planet could point towards the star, while the planet still has a rotation more perpendicular to the star. That's theoretically possible, but only if the mass imbalance from the Iron rich side of the planet was greater than the gravitational imbalance from equatorial bulge. Unlikely but possible. If you want more specifics, I can give an answer with more details. $\endgroup$ – userLTK Sep 20 '16 at 5:37
  • $\begingroup$ Some satellite orbits are sun-synchronous, tuned so that precession due to Earth's equatorial bulge has a period of 1 year. Such orbits are highly inclined but not exactly polar. $\endgroup$ – Mike G Sep 20 '16 at 17:04

Would this even be theoretically possible?

No. The law of conservation of angular momentum prevents this for a planet with reasonable inner structure in a gravitational field like that around the sun.

The angular momentum of the planet points in a certain direction. Changing this direction requires a torque perpendicular to the rotation axis.

If a torque existed that could force the rotation axis of the planet to always point towards the sun, the corresponding force had to have a component perpendicular to the line from the sun to the planet. The force also had to change periodically with the orbit of the planet. Obviously such a force does not exist in the planet-sun-system.

  • 2
    $\begingroup$ "with reasonable inner structure" -- the implication of course is that with unreasonable inner structure, such as a counter-rotating core and some sort of super-science to use one to drive the other in synch with the year, you could rig something up. $\endgroup$ – Steve Jessop Sep 19 '16 at 20:14
  • $\begingroup$ From driveby editor: "I cannot write a comment so I try this way and I hoep original poster wouldn't mind. I in part can understand your reasoning and explain ti to my self like this: The material a planet was made of was circling around the star. $\endgroup$ – called2voyage Sep 19 '16 at 20:20
  • $\begingroup$ Part of that angular momentum is retained in the orbit and part in the rotation of planet. This, logicalyy cannot deviate to much form the plane of rotation of original material. But... this should be in a comment: How about if a star traps a wondering plant. This one should have any axis of rotation it likes. Am I wrong about that. Will some kiind of force, like tidal forces, align the roration of the planet?" $\endgroup$ – called2voyage Sep 19 '16 at 20:20
  • $\begingroup$ From astronomy.stackexchange.com/users/14289/user6694745 $\endgroup$ – called2voyage Sep 19 '16 at 20:21
  • 3
    $\begingroup$ The answer to that question is that it doesn't matter how the system originally came to be. Compare the angular momentum of the planet's rotation at one point in its orbit, with the opposite point in the orbit. It has the same magnitude but opposite direction, since the axis of rotation of the planet has changed 180 degrees over the course of the half year. The force required to achieve this is what aventurin is saying simply doesn't exist in orbital dynamics. $\endgroup$ – Steve Jessop Sep 19 '16 at 20:34

Uranus has an axial tilt of about 98 degrees, so I think this is about as close as you'll get for planets with known tilts. However, you won't find one with a pole that always faces the sun, just because of the geometry of the situation. With Uranus, for about 1/4 of its 84 (Earth year) year, it will have one pole more or less facing the sun, then for the next 1/4 it'll have both poles facing perpendicular to its orbit (away from the sun), then the next 1/4 it'll have the opposite pole facing the sun, then for the final 1/4 it'll have both poles facing away again. This is a bit of an over simplification, but it captures the general idea.

In other words, the two poles point mostly the same direction in space**, and for part of its orbit one pole or the other is pointing more or less at the sun, but for the periods in between it will face away. If you want a planet who always has the same side pointing at the sun, it has to rotate with a North/South pole similar to Earth, and it needs its "day" to exactly equal its "year". None of the planets in our solar system do this, however many moons do. Earth's Moon, as well as all of Jupiter's and Saturn's major moons do this. This is called tidal locking. Note that Mercury is on the list I linked, but it isn't perfectly tidally locked, it is in a 2/3 resonance.

**There is a little bit of wobble due to precession, but this happens over long time scales and doesn't change by a huge amount.

  • $\begingroup$ I think it might be interesting -- for a SciFi story -- to posit a planet which both rotates about one pole and rotates about a second, perpendicular, pole to achieve what the OP's asking about. Gotta be more likely to exist than DiscWorld! $\endgroup$ – Carl Witthoft Sep 19 '16 at 19:02
  • 3
    $\begingroup$ @CarlWitthoft See Euler's rotation theorem, which roughly means you can't have two axes of rotation. $\endgroup$ – James K Sep 19 '16 at 19:17
  • 2
    $\begingroup$ I'm pretty sure you can have 2 axis of rotation in a 4 dimensional space, but Euler's rotation theorem limits us to 1 in 3 dimensions. Good luck writing a SciFi story with that and being able to have it make sense to the average reader. $\endgroup$ – Cody Sep 19 '16 at 20:21
  • 1
    $\begingroup$ @JamesK: doesn't it mean two rotations are really just a single rotation over another axis, meaning that it is in fact possible? $\endgroup$ – RemcoGerlich Sep 20 '16 at 9:59
  • 1
    $\begingroup$ Yes, it means that two rotations about axis A and B are in fact just one rotation about C. en.wikipedia.org/wiki/Rotation_matrix#Multiplication $\endgroup$ – AnoE Sep 20 '16 at 14:09

(Would this even be theoretically possible?)

I very much doubt it. What you're suggesting would involve incredible changes in angular momentum over the course of one orbital period. In the scenario you suggest, the angular momentum vector for the planet (the direction of its north pole) starts off pointing in one direction. Six months later, it's now pointing in the opposite direction. Since angular momentum is a vector quantity, that would require an enormous amount of torque and energy to achieve. And then you have to do the same thing again for the second half of the orbit, and repeat the whole thing each orbit.

  • $\begingroup$ This all reminds me of a question I once asked on the Math site, to which I never received and have never really found a satisfactory answer :/ ... math.stackexchange.com/questions/168172 $\endgroup$ – Fattie Sep 23 '16 at 17:19

In theory, there's no reason why a planet couldn't orbit close to the way you suggest. It's probably extremely rare, but theoretically possible. This gets a little tricky to explain, but an axis of rotation can be divided into perpendicular vectors. There's no real benefit to doing it that way but it can be done, just like direction and velocity can be separated into 3 vectors, rotation can too, and the combination of the perpendicular vectors gives you the rotation of the object and it's axis of rotation. That's discussed in a bit more detail here.

If we use the Moon as an example. The Moon is oriented so it's heavier side permanently points towards the Earth. (the side of the Moon facing the Earth has a thinner crust and because crust is lighter than the mantle, the "Earth side" of the Moon has slightly greater density than the Far side. That greater density is more attracted to the Earth than the less dense far side and over time, the dense side of the Moon was oriented to permanently face the earth, similar to how the dense side of a floating objects orients itself downwards when floating in water.

All planets, not being perfect spheres with perfectly layered density are unbalanced but the Moon is more lopsided than most. Mars more than the rest of the planets. Earth's uneven gravity has been measured in great detail, but in terms of effects on Earth's orbit and rotation, it's uneven gravity is pretty insignificant.

Now some objects experience significant tidal deformation, like Io and Enceladus which are measurably squeezed by their elliptical orbits as they move closer to and further from their planets, but that's something else and you probably wouldn't want that on a planet as the volcanic effects would be too big. But what I'm talking about isn't a tidal bulge but a permanent imbalance of mass where one side of the Moon is denser than the other.

So, if we figure the dense side of the Moon must always point to Earth due to tidal locking, that still leaves one way that the Moon can rotate without affecting the locking and that's along it's 90 degree longitude lines, and that, in theory could happen in addition to it's existing synchronous rotation that keeps the heavy side of the Moon pointed towards earth.

This wouldn't be two axis of rotation, it would be one axis of rotation where the East-West vector of the rotation would still be tidally locked but the rotation around the 90 degree longitude line wouldn't interfere with the tidal locking, so the effect, the familiar face would spin around in a circle but always face the earth.

We could in theory do that artificially if we put a big train on the moon and ran it 24/7 in the same direction around the 90 degree longitude line. Do that long enough and the moon would begin to spin.


But for such a scenario to actually exist, you'd need a lot of luck, because the forces that tend to tidally lock a planet or moon tend to also reduce any other rotations other than the synchronous 1:1 tidally locked rotation.

Another problem is the Equatorial bulge which is a consequence of rotation, tends to have more mass and that would want to orient itself towards the planet, so for this to work you'd need a slow rotation and a small equatorial bulge where the added mass around the equatorial bulge was small enough to not alter the direction of the heavier side of the planet.

Now it's not hard to imagine a planet that's not tidally locked having just the right orbital velocity and position to do what you want, but in such a scenario with no tidal locking, it would be temporary. Planets rotations tends to slow down over time, so perfectly matching without tidal locking would be coincidental and temporary. More likely you'd have a very slow rotation, not a permanent side facing the planet, but a very gradual movement.

There are no planets (or moons) like you describe that we know of. It's probably a hugely unlikely scenario that would probably only be approximated (a bit like Uranus).

  • 1
    $\begingroup$ "Do that long enough and the moon would begin to spin." -- well, it would begin to spin immediately, but only at a rate 1/squillionth of the rate the train is going in the opposite direction. Where 1 squillion is the ratio of angular momentums. If it was a rocket-powered train you could build up speed, like a Catherine wheel, but then you'd be better off just gluing the rocket to the moon, not putting it on rails. $\endgroup$ – Steve Jessop Sep 20 '16 at 9:05
  • $\begingroup$ I'm sorry, but unless I'm completely misunderstanding what you've said, this is completely wrong. Unfortunately its too difficult to rebut all of this in a comment. Effectively it sounds like you're saying the axis of rotation could point towards the Earth, but the tidal forces would keep the "heavy" side of the moon (and thus the axis of rotation) pointing towards us. But that requires two axes of rotation which you admit is impossible. What's more tidal locking can't occur if your axis of rotation is facing the planet/star. $\endgroup$ – zephyr Sep 20 '16 at 12:53
  • $\begingroup$ @zephyr I might not have explained it well, but for this unlikely scenario to work, the axis of rotation would be at an angle. A standard tidal locking the axis of rotation is perpendicular to the plane of orbit and the rate of rotation matches the orbit. But from the planet, where the moon is tidal locked, We don't see the Moon rotate, for example, but it does. If it didn't, we'd see different sides of it. And you could theoretically add a spin to the moon where it would appear to rotate in the with the axis pointing to the planet, but the real axis would be a combination of the two. $\endgroup$ – userLTK Sep 20 '16 at 13:00
  • $\begingroup$ @userLTK For one thing, you can't just combine two disparate rotations into a single axis of rotation. Rotation doesn't work like that. For another thing, even if you could add second axis of rotation which caused a different net axis, that axis would no longer be pointing directly at the planet (or stay pointing at the planet throughout the orbit) which means your scenario doesn't fit LeHill's question. What your proposing just can't happen. $\endgroup$ – zephyr Sep 20 '16 at 13:06
  • $\begingroup$ @zephyr So your saying a 45 degree axis of rotation can't be divided into a 90 degree and a 0 degree of equal magnitude? The math gets pretty complicated, but I'd like verification on that because I don't see why you couldn't. See this question here: math.stackexchange.com/questions/1258051/… It's not a "second" axis of rotation it's one axis, but you're looking at it from an X and a Y axis so to speak. $\endgroup$ – userLTK Sep 20 '16 at 18:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.