Is there a scientifically plausible way for a habitable world to have a sun that undergoes a regular, dramatic, and easily observable change? One comparable in visual impact to an event like a solar eclipse or a full moon? Such a change might affect:

  • Color.
  • Light and heat output.
  • Visible texture and shape.
  • Solar effects on the terrestrial climate, winds, magnetic field, etc.

The change doesn't neccesarrily have to come from the sun itself. An atmospheric effect would be fine for example. It's more about what it looks like from below.

My goal here is to create a world in which the sun cycles in a way that promotes feelings of intense religious fixation; inspiring dread, awe, submission, etc. So, just as one example, an ideal effect might be if the sun turned dark red for one month every few years, and somehow caused windstorms on the surface of the world, before fading back to normal.

As I don't have the knowledge to guess at the limits of the plausible, I've intentionally made this question a bit wide. Any kind of solar event that accomplishes the aesthetic goal could be very helpful.

  • 2
    $\begingroup$ Our sun goes through an 11 year cycle. These changes are observable. If this is helpful I can turn it into an answer. $\endgroup$
    – Andrey
    May 11, 2017 at 18:26
  • 4
    $\begingroup$ @Andrey I wouldn't exactly call the sunspot cycle "dramatic". $\endgroup$
    – Phiteros
    May 11, 2017 at 18:29
  • 3
    $\begingroup$ Note: When I read this, I was thinking "new sun" "quarter sun" "half sun" three-quarters sun" "full sun" and so forth, along the lines of lunar phases. Clearly what I was thinking is much more difficult than what you asked $\endgroup$
    – Cort Ammon
    May 11, 2017 at 18:40
  • 4
    $\begingroup$ The sun undergoes a daily eclipse when the earth interposes between the sun and the earth... 😁 $\endgroup$ May 12, 2017 at 2:18
  • 5
    $\begingroup$ Surprised nobody has suggested simply giving the planet an eccentric orbit. $\endgroup$
    – Qwerky
    May 12, 2017 at 11:47

11 Answers 11


Your world has a very long rotational cycle

Consider our moon. It rotates about its axis remarkably slowly, taking around 28 days to complete one full day/night cycle. That means 336-hour days and 336-hour nights; dawn becomes a big event, a time of celebration and renewal, a time for weddings and preparations for the day ahead. Sunset becomes a dark and forboding time, a time to batten down the hatches, gather in your people, and eke out your resources until the light can return...

Your world orbits a gas giant

For a less extreme result, your world could orbit a gas giant planet like Jupiter, and experience regular, very long eclipses. Give your planet a fairly wide orbit and offset it from the ecliptic to make the eclipses of varying frequency and duration.

The population would see the heavens as a perpetual struggle between the glory of the light and the domination of the dark, with the gas giant clearly visible even in the daytime sky. The gas giant might be seen as the devil holding the people in thrall.

There is an active supervolcano

A supervolcano that regularly burped out a few hundred square kilometers of ash could throw clouds of debris up into the atmosphere, changing the colour and strength of the sun until they settle out. The sacrifice of virgins would, of course, be required to return the sun to its normal colour and brightness.

A comet swarm

There's a star out there in the galaxy called KIC 8462852, and it's weird. Observation of the star by the Kepler space telescope has detected a series of sudden and dramatic drops in the star's brightness - much greater than would be caused by a planetary transit. One of the hypotheses is that there is a 'swarm' of comets passing between us and the star, blocking out a huge amount of the light.

For our purposes, a swarm of comets is a possibility, or there could have been a recent collision between two planets in the system, leaving a huge cloud of debris sunward of our planet. Either of these possibilities would result in semi-regular dimming of the star as our planet passed through the shadow of the swarm/debris. The colour of the star could change as well, depending on the density and makeup of the debris cloud. As a bonus effect, your planet would have a lot of meteor activity, allowing for fun things like cities destroyed by sudden impacts, regular fireballs streaking across the sky, and other fun stellar fireworks.


The sun is alive. It is sentient, and it is angry. It watches these pathetic ants crawling on the surface of the world, and when it sees something it doesn't like, it does something about it. It may deprive the world of life-giving light until the appropriate sacrifices have been made; it may burn blazingly hot until the one that offended it has been identified and staked out in a desert somewhere to be burned to a crisp.

  • $\begingroup$ This is great, thanks. I have a few follow up questions re: the supervolcano, for you or anyone else who cares to answer any of them. 1. Is there a plausible composition of ash thin enough to turn the sunlight red across a landmass approaching the size of Eurasia, but not blot out the sky completely or cause a Permian-style extinction? 2. Is it also plausible for a volcano of that size to erupt like clockwork, as with some geysers? 3. Would the ash eventually settle, or snow, and what general effect might that have on the ecosystems below? $\endgroup$
    – Random
    May 11, 2017 at 21:17
  • 2
    $\begingroup$ "Is it also plausible for a volcano of that size to erupt like clockwork" yes, but those 1816-level explosions only happen when (1) there is a lot of rock over the magma chamber, and (2) enough liquid water is percolating down into the hard rock to make it eventually go boom instead of steam seeping up through different cracks and then venting out via geysers. $\endgroup$
    – RonJohn
    May 11, 2017 at 21:49
  • $\begingroup$ @Era 1) Yes - the 1883 eruption of Krakatoa caused such vivid sunsets in the Americas that people thought there was a fire and called out fire engines. That continued for years. 2) Plausible, but it would take a very precise set of circumstances that RonJon described. 3) yes, the ash would eventually settle out of the atmosphere. It would probably come down mixed with rain. Volcanic ash leaves very fertile soil behind, so it could even be a net positive. $\endgroup$
    – Werrf
    May 12, 2017 at 10:37
  • $\begingroup$ Now I'm wondering what's the longest eclipse possible for a body in circular orbit around any known planet. $\endgroup$ May 13, 2017 at 22:11

Have a companion star that orbits the same star

The easiest way to accomplish something like this is to simply have a binary companion star. Here is an answer where I show a stable orbit for a binary companion inside the orbit of a habitable planet.

In this scenario, there would be one main sun with a relatively fixed strength. However, there would be a companion star which would vary in brightness. For that question, the companion star would vary from brightest to dimmest by several orders of magnitude every 2 months or so.

The key impact of this is that when the companion star is closest, it will significantly extend the length of the day into a long twilight in either morning or night; while at its farthest it would have almost no effect on daylight time.

Have the planet orbit two stars

Here is an answer where I show the seasonal solar intensity of a planet that has a figure eight orbit around two stars. The result would be that the planet would move from orbiting one sun to the other every year or so.

In this situation, the stars themselves could have different characteristics (different colors, for example). Also, depending on the luminosity of the the suns, one could turn the planet into a sweltering jungle for part of the year, and the other, cooler star could turn the planet into a frozen taiga...sort of mega-seasons.

Also, during certain parts of the year, both suns will be almost equidistant from the planet...if they are equally bright then there is no more nighttime on the planet. In fact by altering the planet's rotational parameters, you could cause a year to be a complex cycle of all-day and all-night at different points of the year.

This is probably the better explanation for something truly awe-inspiring.

  • 6
    $\begingroup$ In the two star example, there would be a simple cycle of "12hrs of sunlight" and "24hs of sunlight", it would never be "all-night". Unless one sun was a black hole. Now that would be interesting. $\endgroup$ May 11, 2017 at 23:07
  • $\begingroup$ @MooingDuck black holes usually have some disks around them. This can get pretty bright. $\endgroup$
    – Mołot
    May 12, 2017 at 10:05
  • $\begingroup$ @MooingDuck Unless your rotational period was close to the planet's orbital period, a la Venus. In that case, it might line up such that one side of the planet is facing away from both stars for days, or even months at a time. $\endgroup$
    – kingledion
    May 12, 2017 at 12:05

You could put your world around a variable star. In variable stars, the actual radiation output changes in quantity and sometimes in type as well. There is a layman's article that discusses some of the implications with respect to habitation, written from the point of view of using variables in Traveller, at http://www.freelancetraveller.com/features/science/variables.html

  • $\begingroup$ No wonder you gave a Traveller-related answer. variable stars was the first thing I thought of & when I checked the other answers it seemed no-one else had too -- until I discovered your answer. Plus one because great minds think alike. $\endgroup$
    – a4android
    May 12, 2017 at 5:12

It isn't necessarily very regular, but small (between 0.09 and 0.40 solar masses) red dwarf stars can have very strong flares (which periodically really increase luminosity) and huge starspots that can cut the energy output dramatically for long periods of time. From solstation.com:

Many dim, red (M) dwarf stars exhibit unusually violent flare activity for their size and brightness. These flare stars are actually common because red dwarfs appear to make up more than two-thirds of all stars in our galaxy. Although flares do occur on our Sun every so often, the amount of energy released in a Solar flare is small compared to the total amount of energy that Sol produces. However, a flare the size of a Solar flare occurring on a red dwarf star (such as UV Ceti) that is more than ten thousand times dimmer than our Sun, Sol, would emit about as much or more light as the red dwarf does normally.

Proxima Centuri seems to have a 6.5 year cycle involving both flare activity and (dark) starspots.

From the Habitability of Red Dwarf Systems, starspots can lower light output by 40% for long periods of time.

Note that this behavior tends to be worse in dimmer red dwarves (< 0.2 solar mass) and not as pronounced in larger ones.


One of the possibilities is that some extremely advanced interplanetary civilization is harvesting the energy from your sun.

Thousands, possibly millions, years ago they've built a stellar engine around your sun, most probably Dyson swarm.

Example image from Wikipedia:

enter image description here

Some of the components are closer to the sun than your planet, some of them are much further away. This creates a seemingly random pattern of astronomical events. Sun can be fully or partially blocked for large periods of time. Some parts of the night sky will be affected as well.

The orbiting patterns of the power-harvesting components are under control of their creators. Hopefully, they take into account the life on one of the planets.


An aurora is caused by solar radiation interfering with Earth's magnetosphere, exciting charged particles in the atmosphere that then give off photons as they relax to their equilibrium state. Cycles or phases in the output of electrically-charged solar particles (solar wind) could create a suitably dramatic visual effect in the sky, although you may not be able to see it during the day (unless it was very intense).


Your planet could have one or more moons that regular eclipse the sun.

As each moon passes between the planet and the sun, the sun will go through phases.

Of course we get this phenomenon on Earth, as the Moon occasionally eclipses the Sun. It's pretty rare on Earth, but one could imagine an orbit for the Moon that would make it more common. Namely, an orbit that is in precisely the same plane as the planet's orbit around its sun.

If there are multiple moons in similar orbits, it would be more common still.

  • 1
    $\begingroup$ This is currently in the low quality review queue. You should probably edit to expand on it; otherwise, this might be at risk of being deleted. $\endgroup$
    – user
    May 11, 2017 at 19:36

A an object of 1/4 the mass of their sun (like a smaller star, giant comet, etc.) could have fallen into the sun a hundred years ago. The star might still be undulating irregularly, like a blob of water in microgravity.


(Hopefully some astrophysicist reading this can clarify whether this would completely destroy the planetary system, due to gravitational effects when the foreign object arrived, or whether the planets could survive. I'm sure orbits would be perturbed! Perhaps if the object approached from the "back side" of the sun relative to the planet in the story, the planet would still be able to orbit, though the orbit would be more elliptical, causing more extreme seasons.)



You will want to select answers that don't threaten to kill off the natives of the planet. That means a lack of drastic changes in the amount of radiation the planet receives from the sun.

The planet's axis of rotation will probably be tilted away from exactly right angles to the plane of its orbit. That will give the planet seasons like Earth has - Earth's axis is tilted 23 degrees from exactly a right angle to the plane of its orbit. Thus at different seasons the sunlight hits a spot on Earth at different angles and with different intensities, resulting in different temperatures. Even though the amount of sunlight reaching Earth is almost exactly the same year round, Earth's axial tilt causes drastic seasonal temperature changes in various parts of Earth.

In temperate and arctic regions the temperature changes with the seasons are drastic enough that plant and animals life and humans have to adapt to it in order to survive.

The severity of your planet's seasonal temperature changes will depend on its axial tilt which you have not specified.

If the planet's sun was significantly variable the temperature changes added to the seasonal temperature changes could wipe out all life.

If the planet's orbit around it's sun was significantly elliptical instead of almost circular the temperature changes added to the seasonal temperature changes could wipe out all life.

Volcanic eruptions could change the appearance of the planet's sun drastically as they often have on Earth. They have often caused widespread crop failures and famines and one tens of thousands of years ago might have killed almost every human alive at the time. Even larger and deadlier eruptions are in the geological record.

Stars suitable for having habitable planets are all main sequence stars and have a relatively small range of mass and luminosity.

I suggest that you try something less drastic than most of the suggestions.

The eyes of the natives may be more sensitive to bright light than human eyes. Maybe they are usually active at night, and don't go out in the day more than humans go out a night.

Thus they don't get good looks at the blinding sun even as much as humans do. When they go out in the day, they notice the position of the sun from the length and direction of shadows.

And the sun could "change shape", and thus the shape of its shadow, if it was a double star. The planet could orbit around both stars if it was several times as far from the two stars as the separation between them. The stars would probably orbit each other with a period of a few Earth days, while the orbit of the planet around them might equal a few Earth months to a few Earth years.

So over a period of Earth days or weeks, the stars would appear to move apart to a distance of maybe ten or twenty degrees of arc, and then get closer together until they passed each other and grew apart on the other side, and then came back together. And their shadows, which the natives could see, would do the same, pointing farther apart as the stars grew farther apart, and moving closer together as the stars got closer together.

And maybe the planet's atmosphere would periodically change, and fill with light scattering particles, dimming the apparent brightness of the suns so the natives could look directly at them and see that they were 2 balls of light and how they moved relative to each other.

Or maybe it is the opposite. Maybe the natives are active in the day, and the atmosphere is normally so overcast that they can't see the suns, merely a blob of slightly more intense light that periodically gets wider and narrower. And perhaps there is a periodic changes in the atmosphere that make it less overcast and enables them to see the two stars clearly.

Another possibility is having a star that is slightly - repeat slightly - variable and a planetary orbit that is slightly - repeat slightly - elliptical, and both the variability and the orbit have the same period.

So either 1) the star is brighter when the planet is closer to it, and thus the two causes of temperature differences will reinforce one another, or 2) the star is brighter when the planet is father from it, and thus the two causes of temperature difference should almost cancel each other out.

And if the star's period of variability and the planet's orbit period are not exactly the same, the planet's situation will gradually change from 1) to an intermediate condition and then to 2) and back again.


An atmospheric effect would be fine for example. It's more about what it looks like from below.

Big volcanoes eruptions spew lots of dust and gasses into the atmosphere. Seeing that, and having a year without a summer https://en.wikipedia.org/wiki/Year_Without_a_Summer would cause oodles of dread, awe and submission to the god who did this.

  • $\begingroup$ Thanks, a volcanic eruption is emerging as one of the most likely ideas so far, and that looks like a very handy link. I've added a few relevant follow up questions to the subject in the comments under Werrf's answer if you feel like addressing any of those. $\endgroup$
    – Random
    May 11, 2017 at 21:41
  • 1
    $\begingroup$ Volcanos also have the (religious) advantages of being: Big. Loud. Explosive. Awe-Inspiring. Extremely dangerous. And appearing to erupt (to pre-modern people) for no apparent reason. $\endgroup$
    – Kaz
    May 13, 2017 at 21:54

No to phases as we consider them, because phases are changes in reflation/shadow, and stars are central source of light, broadcasting in all directions.

Regarding light and heat output, that's very possible. Unfortunately, those are based on the energy output of the sun, which would indicate the sun is late in it's life, soon to become a red giant. Life on that habitable world is in serious trouble.

  • $\begingroup$ Ok, to follow up on that: If I had a world that was orbiting a star that was verging on becoming a red giant, but not there yet, would it periodically change color or luminosity and then revert? If so, is there a term for that I can google? And how much trouble is serious trouble? Thanks. $\endgroup$
    – Random
    May 11, 2017 at 21:21
  • $\begingroup$ According to this en.wikipedia.org/wiki/Future_of_Earth#Solar_evolution article, the luminosity of the Sun will keep increasing. Naturally, that's going to eventually boil off all the water and turn the planet into a dead lump. In your story this is happening, but the luminosity rises in a "bumpy" pattern instead of smoothly. How frequently do you want these luminosity changes to happen? $\endgroup$
    – RonJohn
    May 11, 2017 at 21:31
  • $\begingroup$ Ideally, and giving myself a bit of wiggle room for later development, I'd say in a matter of years, somewhere in the between range of once a year (which would be like a new year or a seasonal change) or once a decade or two (marking a longer, more generational span of time). $\endgroup$
    – Random
    May 11, 2017 at 21:37
  • 1
    $\begingroup$ The Sun is already on an 11 year cycle of sun spots, and that does impact it's luminosity (but just not that much). In 100M years, it's fictiony-plausable that the luminosity fluctuates by enough to noticeably impact the climate (Sun-baked to frozen, with "nice" in the middle). $\endgroup$
    – RonJohn
    May 11, 2017 at 21:42
  • $\begingroup$ What term would you use for the variations in output of a variable star, then? $\endgroup$
    – Shalvenay
    May 12, 2017 at 22:22

You must log in to answer this question.

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