I'm a fantasy author. I'm thinking about the setting in my next novel having something to do with a world that has purple skies during the daytime.

I'm wondering how could this be done from a realistic standpoint?

I was potentially thinking about having the planet I'm creating have a moon that revolves around it that emits a dark blue light, but I'm not sure if this would have the desired effect.


The Rayleigh-scattering solution

Your chance to make it work naturally is to accept violet (a pure spectral color) instead of purple (a mix of red and blue).

Since it is a shorter wave length than blue, it will be Rayleigh-scattered stronger than blue. So your problem is to make the violet intensity in the light spectrum of your star higher than the intensity of the blue. Which is quite simple to do: just move the temperature of the star by 700-1000K above the Sun's one.

enter image description here

You'll have some side effects with a star hotter that Sun, so you'll probably need:

  • a thicker atmosphere (good, more medium to scatter that light. Be careful tho' with the greenhouse effect) and
  • a decent planetary magnetic field - good, now you can have some violet at night too in more extended aurorae (you'll have some green too if you have an Earth-like atmosphere composition)enter image description here (image found using image google search, original source, credited to Brad Goldpaint, apparently you can buy it in hi-res or printed from here - I'm not affiliated).

By the way... ummm... "the visual perception is in the eyes of the beholder" - which may be more sensible for green/blue than for violet. Look, some say that the Earth's sky is actually violet, we just see it as blue, because of the spectral sensitivity of the dyes our retina cells use.

But if that's the case, why isn't the sky violet? Sure, blue light is scattered more than red or green, but violet light has an even shorter wavelength, so violet should be scattered more than blue. Shouldn't the sky appear violet, or at least a violet-blue? It turns out our sky is violet, but it appears blue because of the way our eyes work.
We don't see the greenish hue, however, because of the sky's violet light. Violet is scattered most by Earth's atmosphere, but the blue cones in our eyes aren't as sensitive to it. While our red cones aren't good at seeing blue or violet light, they are a bit more sensitive to violet than our green cones. If only violet wavelengths were scattered, then we would see violet light with a reddish tinge. But when you combine the blue and violet light of the sky, the greenish tinge of blue and reddish tinge of violet are about the same, and wash out. So what we see is a pale blue sky.

The colored atmosphere solution

Purple aerosols - not a lot that can stay in the atmosphere for long, at least not in the Earth-like conditions

What can we use/handwave:

  1. higher ambient temperature and high concentration in purplish elements/substances in the atmosphere - for example, iodine (triple point - were vapours can coexist with liquid/solid form - at 113.5C. A tad hot, but with a high atmospheric pressure it still may lead to life of the planet - halo-thermo-philes aren't impossible)

  2. endemic purplish material of biogenetic origin - like some spores or pollen of a life form (or more) that's endemic on the planet. Will probably need close to circular orbit around the star and low planetary tilt to exclude strong seasonal variation (which can put a pause on the life of the purple-something generation)

  3. in an ocean world, they'll call "sky" the surface of the water - there are lots of substances that have a purple color in solution, but the bad new is the light won't penetrate too deep inside the ocean)

  4. I wouldn't bet for volcanos as a source of purple aerosols, tho', at least not for creating a long-lived and planet wide purple haze in the atmosphere

Finally, as tongue-in-cheek alternative, is may be a planet which's population loved so much Prince they geo-engineered their planet's weather in a perpetual Purple Rain

  • $\begingroup$ But if the star is that much hotter than our sun, it won’t last long enough for complex life to evolve on its planets. Hot stars burn out fast. $\endgroup$
    – Mike Scott
    Nov 27 '21 at 14:11
  • $\begingroup$ @MikeScott "if the star is that much hotter than our sun, it won’t last long enough for complex life to evolve on its planets" A 1000K increase is not even a fever, much less a terminal illness. 6500k is bang in the middle the F-type main sequence star at 1.2M☉. The formula for the star lifetime-to-red-giant is $10^{10}(M/M☉)^{-2.5}$ years. For 1.2, the value is $6.34Gy$. The age of Sun is $4.6Gy$ and the age of Earth is $4.54Gy$ - both comfortably fit in the $6.34Gy$ limit. $\endgroup$ Nov 27 '21 at 15:10
  • $\begingroup$ Very nice aurora photo. Where'd you get it? $\endgroup$ Nov 27 '21 at 18:17
  • $\begingroup$ @chrylis-cautiouslyoptimistic- added the source of the image under the photo. $\endgroup$ Nov 28 '21 at 2:19

The colour of the sky on Earth is blue due to Rayleigh scattering. This means that unless the incoming light is mostly purple, you're not going to get a purple sky.

Now... all is not necessarily lost. Mars has a reddish sky because of its load of reddish airborne dust. If you could find a way to have your world's atmosphere contain a significant amount of airborne purple dust, you'd have your purple sky.

  • 1
    $\begingroup$ Purple is not a spectral color. Purple light is necessarily a mixture containing red and blue light. The blue will be scattered away and color the sky... $\endgroup$
    – AlexP
    Nov 26 '21 at 21:35
  • 1
    $\begingroup$ @AlexP Actually, purple is a spectral colour. However, it is not a black-body colour. It can't be generated by stars in isolation. However, as an airborne dust, it could work. $\endgroup$
    – Monty Wild
    Nov 26 '21 at 22:50
  • $\begingroup$ Actually, no, there is no such thing as monochromatic purple light. Purple light is always a mixture. (Or, if you prefer, no monochromatic light at any wavelength will produce the sensation of purple.) Yes, violet is a spectral color, and the question allows violet. Yes, the dust would work; that's why I have upvoted the answer. $\endgroup$
    – AlexP
    Nov 26 '21 at 23:13
  • $\begingroup$ @AlexP "there is no such thing as monochromatic purple light" but you can use the violet color (which is monochromatic in the spectrum) as a substitute. And Rayleight scattering will work just fine with higher frequencies, one just need to give them more intensity (than blue) in the spectrum. $\endgroup$ Nov 26 '21 at 23:35

Short Answer.

It might be possible for stars that emit more violet light than the Sun to have habitable planets, where the daytime sky might be more violet colored than blue.

The star of a habitable palnet might have a ring of dust particles that absorb ultraviolet light from the star and re emit it as violet light, and that might possibly color the day time sky more violet than Earth's sky.

A habitable planet in another star system might possibly have a ring of dust, and that dust might poossibly be heated by the star and emit violet light, which might possibly color the daytime sky of the planet more violet than Earth's sky.

A habitable planet should have some dust in its atmosphere, and if that dust is purple colored it might refelct enough light to possibly make the day sky appear more purple than Earth's does.

The star and planet might be passing through a nebula, and that nebula could reflect starlight from the star onto the planet. But It seems improbable that the nebula could be bright enough to change the color of the night sky, let alone the day sky.

A planet habitable for some lifeforms, possibly even habitable for lifeforms with similar requirement to humans, could possibly have a thinner atmosphere, which like the atmosphere of Earth at hight altitudes, was a much deeper blue, perhaps a purplish blue color.

Long Answer:

Part One of Seven: A Star Emitting More Violet LIght.

Adrian Colomitchi suggested that a violet sky could come from having a star which emitted much more violet light than the Sun does, thus causing violet light to be scattered in every direction by air particles andmaking the skay appear violetinstead of blue.

Since it is a shorter wave length than blue, it will be Rayleigh-scattered stronger than blue. So your problem is to make the violet intensity in the light spectrum of your star higher than the intensity of the blue. Which is quite simple to do: just move the temperature of the star by 700-1000K above the Sun's one.

The Sun is a G2V type star with a surface temperature of 5,772 degrees Kelvin. So making the star 700 to 1,000 degreees K hotter than the Sun would give it a surface temperature in the range of 6,472 to 6,772 degreees K.

Spectral class F6V stars have 1.16 times the mass of the Sun and a surface temperature about 6,400 K, while spectral class F5V stars have 1.20 times the mass of the Sun and a surface temperature of about 6,545 K. Spectral class F4V stars have 1.23 times the mass of the Sun and a surface temperature about 6,690 K, while spectral class F2V stars have 1.31 times the mass of the Sun and a surface temperature of about 7,040 K.


So according to Adrian Colomitchi's suggestion a star in that range should have the right surface temperature to emitt more violet light than blue light, and thus make the skies of the planet appear violet instead of blue.

The Earth is about 4.6 billion years ago, and didn't have an oxygen rich atmosphere that humans and life formes with similar requirements could breath until about 500 to 600 million years ago, and thus 4 billion years after the planet formed.

If the story requires the planet to be habitable for humans or lifeforms with similar requirements, the planet will have taken billions of years to tobecomehabitable for them. And the star willhave had to stay on the main sequence with fairly steady luminosity for those billions of years. Thus the star will have to have a spectral type which can stay on the main sequence for enough billions of yeears.

[Unless, of course, in the story an advanced civilization terraformed the planet some time ago, giving it a breathable atmosphere before the planet wuld have developed one naturally}

Someone objected that a star hot enough to emit so much violet light would not last long enough on the main sequence to become habitable, and Adrian Colomitchi said that a star with 1.2 times the mass of the Sun (an F5V according to the table) would last for 6.34 billion years, which would be long enough.

The only scienctific study about planetary habitability for humans (and bengs with the same requirements) in particular, instead of liquid water using life in general, is Habitable Planets for Man, Stephen H. Dole, 1964.

On page 68 Dole says:

The only stars that conform with the requirement of stability for at least 3 billion years are main sequence stars hving a mass less than about 1.4 solar masses--spectral types F2 and smaller--although the relationship between mass and time of residence on the main sequence is probablyy not known with great accuracy and is subject to futue revision (see figure 25).

I note that Dole says that a star with a lifetime onthe main sequence of 3 billion years would have 1.4 times the mass of the Sun and would be an F2V type star. But the table in Wikipedia (which doesn't give the stellar lifetimes) lists a star with 1.4 times the mass of the Sun as an F0V, and a star with 1.31 times the mass of the Sun is listed as an F2V.

Any writer who wants to make the star of a human habitable world as hot and luminous as possible should investigate that descrepancy, and look up the latest information on how long stars of different spectral classes remain on the main sequence.

So it seems possible that a main sequence spectral class star less massive than F0or F2 might remain on the main sequence for at least 3 billion years and thus possibly have a planet which has already became habitable for humans (or being swiht similar environmental requirements).

But some scientists don't think that it is possible for spectral class F stars to have habitable planets. They think that their increased ultraviolet light could prevent life for developing or kill off life osono afterit develops. And they think that because a F class star spends less time on the main sequence it will grow hotter faster than a class G star, and thus that its circumstellar habitable zone will migrate outwards from it faster, which could mean that planets would not spend enough time in the habitable zones of class F stars.

Here are links to several discussions of the potential habitability of planets orbiting spectral class F stars.





So a writer who wants to give a habitable planet a purple or violet sky will have a problem with making the star hotter and emitting more violet radiation than blue radiation. Some more cautious writers might want to avoid putting habitable planets in orbit around and class F stars, and other might want to only use less massive and cooler class F Stars.

Part Two: A Star With a Ring of Dust.

I note that if the star in the system happens to be surrounded by rings of dust, that dust might absorb ultraviolet radiation and reemitt as violet radiation. That might possibly increase the amount of violet radiation the planet receives from its star.

Part Three: A Planet With a Ring of Dust.

And it is possible that the planet could have a ring of particles around it. And possibly the particles absorb ultraviolet radiation from the star and then emitt violaet radiation, which would increase the amount of violet radiation the planet receives.

Part Four: A Planet With Purple Dust in the Atmosphere.

And possibly the planet has dust in its atmosphere, as all Earthlike planets do, and possibly that dust is all violet colored and thus reflects violet light into the sky. And possibly that might increase the amount of violet light in the sky of the planet.

Part Five: A Planet Travelling Through a Nebula.

And possibly the star system happens to be travelling through a nebula.

And possibly that nebula looks a lot like a nebula in a science fiction movie or tv show had is thick and obapuge and reflects a lot of varicolored light. And maybe this nebula happens to be puble colored. So the sky of the planet might look purple at night, and manybe even in the day if the light reflected from the nebula is bright enough.

But movie and tv nebeulas are not realistic. They are based on astronomical photographs which show dense, opague, colar nebulas. And despite the old saying, those astronomical photographs lie. At least those photographs deceive people who don't realized that they are taken with hours long exposures though telescopes which are ketp constantly turning lsightly to keep the nepublas in the view.

The human eye usually processes 10 to 12 images per second. So each image you see has the brightness level of light which has accumulated for only 0.083333 to 0.1 secondexposure time.

There are 3,600 seconds during an hour. So a photograph exposed for an hour would recieve about 36,000 to 43,200 as much light on each picturee elelement as a human eye lookign at the same picture would receive while sensing one image. So an astronomical photo of a nebula exposed for several hours would receive over a hundred thousand times as much laight as you wuld see in each image looking at the nebula though the same telescope.

Nebulas seen by the human eye through a telescope are very delacate, almost transpartent, and pale. They look nothing like photos of nebulas exposed for hours at a time.

So if a fictional solar system is close to a nebula, the nebula would be visible at night as a pale and delicate wonder, but it probably wouldn't be bright enough to lightened the night sky from black to purple. And since daylight on the planet would be tens or hundreds of thousands of times as bright, the nebula light probably wouldn't be enough to make the color of the day sky change.

Part Six: A Planet With a Thinner Atmosphere Than Earth.

If you look at the sky on a clear day, you will see that it is lighter and paler near the horizon and gets brighter and bluer higher in the sky, until it is very blue at the zenith. That is because when you look toward the horizon you are looking through a greater distance of thick air, wich catters the sunlight more. But the air gets thinner with height. So as you look stright up your a looking through a thin layer of the thickest air, then a thin layer of slightly less thick air, and then a thin layer of even less thick air, and so on. The total amount of air you see scattering sunlight above you is less than when you look horizonatally toward the horizon.

Pictures taken at the peak of Mount Everest show a blue sky low down at the horizon and a darker and darker sky above the horizon. At the peak of Everest more than half o the atmosphere that scatters light is below, not above.

The peak of Mount Everest has an altitude of 8,848.86 meters, or 29,031.7 feet, above sea level. And the skylooks darker and more black from there than from lower altitudes. I guess some people might possibly say that the sky looks more purple from the peak of Everest.

The peaks of Mount Everest and some of the other tallest mountains in the world are in what mountaineers call "the death zone", above about 8,000 meters or 26,246.72 feet.

Most climbers in the death zone breath bottled oxygen, and most of them start using the bottled oxygen way below the death zone. And even using bottled oxygen, many climbers suffer from the thin air on high peaks.

People born at higher altitudes can go higher without bottled oxygen than people born at sea level. People from the high Tibetan plateau and the high Andes have the most tolerance for low oxygen levels. And people can train themselves to breath in thinner and thinner air in preparation for living and working in the higest towns and villages in the world, and for climbing tall mountains.

In fact, some mountaineers have performed the fantastic feet of climbing to the peak of Everest safely without bottled oxygen, and the even more fantastic feat of climbing back down the mountain safely afterwards without bottled oxygen despite their fatigue.

But people who try climbing Mount Everest without bottled oxygen have a survival rate which is much smaller than the survival rate of those who try to climb with bottled oxygen, and the survival rate of those who use bottled oxygen is not a very good one itself.

So that makes me think that if a group of people from Earth settled a planet with a slightly thinner atmosphere, their descendants would gradually adapt over generations to function as well there as we do on Earth. And if after many generations some of their descendants settled a planet with a even less dense atmosphere still, they would adapt after many generations to function just as well there as well we do on Earth.

And so, after settling many successive planets with succesivly slightly less dnsse atmospheres, and spending generations on each planet to adapat, a group of human descendants might possibly be able to settle a planet where the typical atmospheric density was similar to that at the peak of Everest, and where the sky was darker than on Earth, a dark blue or maybe a dark purplish blue color.

Insects can fly and kite at very high altitude. In 2008, a colony of bumble bees was discovered on Mount Everest at more than 5,600 metres (18,400 ft) above sea level, the highest known altitude for an insect[citation needed]. In subsequent tests some of the bees were still able to fly in a flight chamber which recreated the thinner air of 9,000 metres (30,000 ft).[12]

Ballooning is a term used for the mechanical kiting[13][14] that many spiders, especially small species such as Erigone atra,[15] as well as certain mites and some caterpillars use to disperse through the air. Some spiders have been detected in atmospheric data balloons collecting air samples at slightly less than 5 km (16000 ft) above sea level.[16] It is the most common way for spiders to pioneer isolated islands and mountaintops.[17][18]

Some birds have been recorded to fly above 8,000 meters or 26,246.72 feet.

They include the alpine chough at 8,000 meters (26,500 feet) on Everest, whooper swans at 8,200 meters (27,000 feet) over northern Ireland, bar-headed goose at 8,800 meters (29,000 feet), the common crane at 10,000 meters (33,000 feet) over the Himalayas, and Ruppell's Vulture at 11,300 meters (37,100 feet).


So birds, multicelled animals who need oxygen to live, are able to fly at about the peak of Mount Everest, where the sky is much darker than at sea level. So alien lifeforms should be able to adapt to a planet where the atmosphere is a thin as on the summitt of Everest, and the sky is much darker than at sea livel on Earth.

Part Seven: Conclusion.

Obviously someone could try a combinatin of several of the suggested methods to make the daytime sky of the planet more violet or purple than that of Earth.

I note that such a planet could be approximately as habitable for Earth humans as Earth is, except for the method of the lanethaving a much thinner atmosphere than Earth. Such a world could be a lot different from worlds with purplish skies for other reasons.

And I could imagine a story where someone comes from a world with a purplish sky because its star emits more violet light than the Sun does, and they crash on some unexplored pplanet. They see thorugh the viewscreen or porthole that the planet has a purplish sky like their home world, and assume it has a similar atmosphere to home.

So they go out the airlock without testing the atmosphere and gasp for air because this is a world where the sky looks purplish because the air is much thinner than on their home planet.


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