I'm wondering what natural phenomenon would allow a planet to have a part of its surface always under daylight, a part of its surface never seeing said light, and a third part where day and night cycle "normally".

The context :

The "day area" needs to always be under sunlight and be hot enough to be unsuitable for human life.

The "night area" needs to never be under sunlight and be cold enough to be unsuitable for human life.

The "day & night cycle area" needs to have a cycle of day and night (obviously) and be suitable for human life.

Always and never mean that these areas should stay stable for a minimum of some thousands of years, with no specific maximum in mind. It also should have been in this state for at least a few hundreds of thousands of years.

The edges does not have to be absolutly still, as long as it does not have a significant impact over the course of this timeframe (i.e. it should not appear as a threat for sentient life in other areas by its motion).

The phenomenon needs to be natural (no intervention from humans or aliens). It has to be stable enough for life to have developed and evolved on the planet, and the planet should not be doomed in the near future because of it.

Each area should cover a significant surface (minimum 10–15%) of the planet. They can be positionned anywhere, as long as they do not form multiple small patches.

Some hypotheses

I asked an incomplete question with this problem in mind but merged with the first thought I had, a tidally-locked planet with two axes of rotation, that was answered as impossible (credits to rek).

Some ideas were proposed, but since I was too imprecise I have no idea if it can fills all conditions above (credits to John Feltz for both) :

  • Although the planet is tidally locked, it only got there in the (geologically) recent past, and it still has some wobble, aka nutation. This will provide a day-night cycle near the twilight area.

  • The [tidally-locked] planet has a large, low moon that regularly eclipses part of the surface.

  • 3
    $\begingroup$ How long exactly does the "always" day or night have to actually be? Are we talking about a few years, few human lifetimes, millennia, the rest of the life of the star? The first thought I had was a possible geosynchronous moon constantly blocking a part of the planet like you said. $\endgroup$
    – Virusbomb
    Commented Aug 31, 2016 at 22:32
  • $\begingroup$ Edited : Added the timeframe during which the areas have to be stable. $\endgroup$
    – Yutreza
    Commented Sep 1, 2016 at 19:20

4 Answers 4


Tidally lock the planet so you have one side always facing the sun, and the other always facing away.

Then add a small moon that orbits the planet at a very high inclination (for example parallel with the arctic circle and not the equator). (Add a secondary body in a geocentric orbit over the north pole that pulls the first moons orbit northwards allowing this) It's easily visible anywhere in the northern hemisphere, but completely invisible from the south.

Make the moon not out of rock but out of an incredibly reflective material.

Now you have an always day side, an always night side (dark side southern hemisphere) and an area that has day/night (dark side northern hemisphere), with light reflecting from the moon to create the "day".

  • 3
    $\begingroup$ That's not how orbits work. You can't have a moon orbiting around a specific latitude. The moon's orbit can be tilted but it must still be centered around the barycenter between it and the planet, which probably lies close to the planet's center. $\endgroup$ Commented Sep 1, 2016 at 21:10
  • 1
    $\begingroup$ I agree with @ApproachingDarknessFish . That is not how orbits work. $\endgroup$ Commented Sep 1, 2016 at 21:15
  • $\begingroup$ Add a secondary body in a geocentric orbit over the north pole that pulls the first moons orbit northwards and it can be. $\endgroup$
    – Reach268
    Commented Sep 2, 2016 at 5:42
  • 1
    $\begingroup$ Nice idea, but you need to edit the second paragraph before getting upvoted. Leaving it as is (wrong physics) will get a downvote in a while. $\endgroup$
    – JDługosz
    Commented Sep 3, 2016 at 1:01
  • $\begingroup$ Imagine the Brit accent of “The Doctor’s” companion questioning, “A geocentric orbit over the north pole?” Dr.: “All moons are geocentric. That’s just a fancy way of saying it’s a moon.” companion: “But the North Pole?” Dr.:“I like the North Pole.” comp: “But,…” (making orbital motions with hands and waving arms) “isn’t that against the laws of celectial mechanics?” Dr.: (dismissively) “They got a variance.” $\endgroup$
    – JDługosz
    Commented May 12, 2017 at 9:32

Combine tidal locking with either high eccentricity or high inclination (or both).

You can see it now a little bit on the moon: it appears to rock back and forth and nod up and down a little, so the area at the sides will see the Earth bob up and down, rising and setting.


In a body orbiting a star, this libration will cause the terminator to shift, and anyone there will experience sunrise and sunset.

This effect can be significantly greater.


This effect doesn’t “doom” the planet. However, imagine interactions with other orbits that cause the inclination or eccentricity to change over geologic time. Life adapts to the twilight zone as it’s getting wider. Then it starts shrinking as the orbit circularizes, forcing life into narrower and narrower zones, with baked and frozen remains to show what once was.

  • $\begingroup$ Thank you for this answer. You mention that this effect can be greater, did you mean in term of speed, or angle, or both ? Wouldn't tidal locking and high eccentricity make the planet too hazardous, similar to Io and it's volcanism ? $\endgroup$
    – Yutreza
    Commented Sep 2, 2016 at 18:55
  • $\begingroup$ Greater effect: I was referring to angle. Speed would be the period of the orbit. Re Io and its vulcanism: the tidal forces have to do with the distance to the primary. I’ve not seen that analized in the context of the habital zone of a red dwarf star. Such high tides would cause the orbit to be changed. $\endgroup$
    – JDługosz
    Commented Sep 3, 2016 at 0:59
  • $\begingroup$ The reason Io stays put for so long is because 4 large moons are all locked together. $\endgroup$
    – JDługosz
    Commented Sep 3, 2016 at 0:59

Have a tidally locked planet with an axial tilt of 45 degrees.

The sun(star) will move, back and forth, along a 90 degree path in the sky. Adjustable Animation
One side of the planet will be facing the sun constantly, and the other side will face away from the sun constantly.

Circle with four regions

This circle represents the planet viewed from the side of the tidally locked face and not down the planet's axis of rotation.

The top yellow region receives constant sunlight.
The bottom gray region receives no sunlight.
The two brown regions receives sunlight for one half of the day each.
The poles lie in each of the brown regions.
The equator (shown in red) runs down vertically, from the bright side to the dark side.

The three regions that receives sunlight, are gradually fading into each other, with respect to planet curvature and the amount of sunshine at different locations.


I guess that this phenomenon is stable enough for life to evolve and not be doomed. Life would arguably have less surface area available and probably very strong winds coming from the dramatic temperature differences. I assume that the orbit can handle some eccentricity and other imperfections, without disrupting the tidal lock or the axial tilt.
I don't think this phenomenon is a case of an unstable equilibrium, like Lagrange Points.

But frankly I have no idea how to judge this properly.
Please supplement further judgements on stability.

Thanks for posing such an interesting question. :)

  • $\begingroup$ I'm sorry, but I don't understand how it would works. I can't visualize the planet movement relative to its star, unless using the phenomenon described by JDługosz with a very strong angle. I have no idea if it could be this strong but it seems unlikely to me to be possible. Also, I don't understand how a planet with an axis tilt could be tide-locked. I did some research and could only find answers going against this here and here $\endgroup$
    – Yutreza
    Commented Sep 2, 2016 at 20:06
  • $\begingroup$ Assuming it’s possible to have tidal locking with a significant axial tilt, the axis will precess rapidly. In terms of having the sun rise/set over the pole, I thing it gives the same motion as high orbit inclination. $\endgroup$
    – JDługosz
    Commented Sep 3, 2016 at 1:08
  • $\begingroup$ I wasn't aware tidal locking allowed for axial tilt. $\endgroup$
    – rek
    Commented Sep 3, 2016 at 11:56
  • 1
    $\begingroup$ Ah, I will try to explain. I agree with @JDługosz suggestion. Imagine a basic Tidally locked planet. Then rotate its orbital plane by 45 degrees. Does this link work for you?: Animation I am using Tidal Locking "(...)there is no net transfer of angular momentum(...)", which I think is equivalent to 'A planet's Orbital period is the same as its Rotational period, as long as the Axial Tilt is between -90 and 90 degrees. $\endgroup$
    – No.Manual
    Commented Sep 4, 2016 at 19:26
  • $\begingroup$ I realise now that @JDługosz has a perfectly good answer. I just approached the subject through Axial Tilt instead of Orbit Inclination. $\endgroup$
    – No.Manual
    Commented Sep 4, 2016 at 20:45

Tidally locked planet, so close to its star that even the "twilight zone" is uninhabitable. One glance at that hellish orb and you're literal toast. A super-powerful magnetic field keeps the atmosphere intact despite the intense solar wind.

Highly reflective moon, orbiting at 90$^\circ$ to the plane of the ecliptic. The habitable part of the main planet lies on the dark side, close but not too close to the day side. When the moon is visible - always a crescent, but lit up on that side by an ungodly hot sun - it gives decent light. When it's not, the place is totally dark, except maybe for a red, sinister glow on the horizon where the killer sun lurks.

Diagram: tidally locked planet, super bright moon


You must log in to answer this question.

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