27
$\begingroup$

I'm an artist and I've been trying to find a comprehensive resource that would help me clearly identify likely sky colors for exoplanets that have atmospheres with a similar chemical composition to Earth. It's trickier than I expected. I've cobbled together several resources, and built what I hope is a decently accurate chart of apparent sun color along with sky color.

How accurate is my chart? Is this a good representation of sky/sun on alien worlds with heavily Nitrogen/Oxygen atmospheres? In what ways could I improve it?

New Sky Chart

UPDATE: I've updated the chart. It's based on my latest research and indicates baseline sky color, cloud color, and apparent sun color for noon/pre-sunset.

Latest science references: Reference A, Reference B

The chart is not meant to account for things like dust, the look of the sky at sunrise/sunset, or other atmospheric effects. This is meant to be a boilerplate for the baseline look of the sky during the day.

Here are the key ideas I've gleamed from my reading. If I'm wrong on any of these points I'd love to know! :)

  • Earth-like atmospheres would tend to be light to dark blue at the zenith, due to the scattering of low-wavelength blue light. If the planet's sun was very hot, the sky would look a deeper blue, while cooler stars would give the sky a lighter blue to almost white look. When the sun gets to 3000k and below, the sky starts to take on an orange/brown tinge.
  • Like on earth, the horizon is the lightest in color and the zenith the deepest.
  • Denser atmospheres would appear brighter (more washed out) and the primary color in the spectrum more "pure" (I'm unsure what the term "pure" means exactly when it comes to optical perception). Likewise, thinner atmospheres would be less bright than earth's and the colors more "pure."
  • With increasing pressure the sky color at the zenith becomes increasingly yellow. In my image this means that an earth-like sky at 10x earth pressure would appear blueish/green near the zenith.
  • I'm assuming the sun would appear tinged by the color listed under "star temperature."
  • I don't know how shadows would be effected, though I assume when you get down into K and M class suns, the surroundings on the planet would take on a progressively redder tinge due to the decreasing prevalence of blue wavelength light?
  • I suspect the gradient of color from horizon to zenith will be steeper/gentler in some atmospheres?

I'd like to account for how the color would shift when you look in the immediate direction of the sun. Would the atmosphere surrounding the sun (like a "halo") appear tinged with the star's color?

Reference #1, Reference #2


Original Image:

$\endgroup$
  • 4
    $\begingroup$ Welcome to WB:SE and thanks for a great question! I'd like to point out that hard-science tag, which restricts answers to ONLY those that can back up the answer with mathematics or authoritative sources. I'm not a fan of it as it tends to heavily restrict the answers. Keep an eye on it. If in a day or two you're not seeing answers that meet your needs, you might consider dropping the tag to allow a wider breadth of answer. $\endgroup$ – JBH Dec 18 '17 at 5:05
  • 3
    $\begingroup$ If you looking for real-life answers based on facts-approved research, I'm afraid this is not the precise place to ask this question. Not to say that members will not research to help out on this, but World Building, as defined in the scope of the genre, is to enable members to design the world as they want it be. If you think (and it's you to decide) that your query is more relevant to astronomy and scientific-imagery (as in, you want to find out what the colors would be, instead of what you want them to be), astronomy.stackexchange.com is the place you want to ask your question. $\endgroup$ – Youstay Igo Dec 18 '17 at 5:25
  • 2
    $\begingroup$ Please remember that what you see on your monitor is different from what I see on mine and from what each other user sees. Only people with monitors calibrated just the way your is, and with software that interprets color profiles the same way your is, can really judge accuracy of your chart presented in graphical form. Also, please, one question per question. $\endgroup$ – Mołot Dec 18 '17 at 6:36
  • 2
    $\begingroup$ Since each type of display renders colors differenly (unless calibrated as Molot says) you could complete your chart adding cells with RGB values of the different colors/gradients for a more scientific analysis from other users. $\endgroup$ – Constantinus Dec 18 '17 at 8:59
  • 2
    $\begingroup$ @user7308301 sadly, RGB is not connected with real-life colors. That's why Pantone earns big bucks on their color standards products. $\endgroup$ – Mołot Dec 18 '17 at 12:46
10
$\begingroup$

I think this is actually a good piece of basic artistic research (I've had professors who have attempted it too).

Sadly, I can't give you an all-encompassing answer to your question. Even though you tried to refine your scope, I'm afraid in order to get what you are looking for you have to account for things that you didn't want to get into. I know I'm going to take a hit for this, but I needed the space to help you, so I will try to address some of the problems you have posed.

(I'm unsure what the term "pure" means

In light, white light is actually an ~impure form of light in the same way that black/brown is an impure form of paint colors. Just like when you mix a bunch of paints together to get a brown-black color, so too if you mix a bunch of light colors you get a white-ish color. So "pure" in this context means closeness to an identifiable color in the spectrum.

Is this a good representation of sky/sun on alien worlds with heavily Nitrogen/Oxygen atmospheres?

It's a start; however:

  • Distance from star is just as important as its solar output, if not more so.
  • Chemical composition (/stage of life) of a star can also have a small impact.
  • Maybe the colors are accurate at the zenith; however, once you get to horizons you enter a whole hell of factors:

An atmosphere isn't some homogeneous blend of gases; actually, it has many layers with different types and concentrations of gases, especially when you factor in organic and volcanic processes. This plays out most notably with horizons, as the sun's light has to travel farther through those more erratic layers, the ones closest to the ground.

(See also: link for why the moon appears orange)

So for your boilerplate approach, the spectrum emitted by your atmosphere should shift red as you approach the horizon from the zenith. The magnitude of this shift depends on the size of the planet/atmosphere, atmospheric density, etc.

How accurate is my chart?

I can't say for sure, but I would expect to see more reddish tones in the lower-energy atmospheres.

I'd like to account for how the color would shift when you look in the immediate direction of the sun.

Looking at the sun, the colors would be closer to the apparent light of the sun (its whitish output) and the zenith color of the sky (impure, as you would be seeing many different wavelengths of light being emitted).

Would the atmosphere surrounding the sun (like a "halo") appear tinged with the star's color?

Halos appear due to the presence of water, much the same way as rainbows appear. A halo would appear tinged with the star's color, but not to any significant degree for a human observer.

Just some useful reading if you haven't already.

I don't know how shadows would be effected, though I assume when you get down into K and M class suns, the surroundings on the planet would take on a progressively redder tinge due to the decreasing prevalence of blue wavelength light?

A shadow is a shadow, being the absence of light. The color here depends a lot on what light the object receives and emits back.

$\endgroup$
  • $\begingroup$ Thanks for the reply. I'm reviewing your links. Thanks! One question...why is distance from the star important? I imagine the only effect distance has is in the luminosity of the sky. Planets further away would (all else being equal) have darker skies. Planets closer in would have brighter skies. Now, the temperature may play a roll in the appearance of the atmosphere, but for the purposes of the graph let's assume standard earth-temps. $\endgroup$ – n_bandit Dec 18 '17 at 18:40
  • $\begingroup$ I'm not sure about a couple of these claims. Why is distance from the star important? Why do you say atmospheres have different types and concentrations of gasses? In our atmosphere that isn't really true. It is changes in liquid water concentration that makes the biggest impact of color/visibility. $\endgroup$ – kingledion Dec 19 '17 at 0:38
  • 1
    $\begingroup$ I have added an updated and improved chart. $\endgroup$ – n_bandit Dec 21 '17 at 1:58
  • 1
    $\begingroup$ On the topic of sunsets, color is a funny thing! Check out pumpkin seed oil as an example of what trace gasses might be able to do to your sky. $\endgroup$ – Draco18s Dec 21 '17 at 19:36
  • $\begingroup$ @kingledion As distance from light source increases, the amount of light the planet receives will decrease. This will slowly and gradually darken the skies. That likely has no significant impact on habital zone sky color, and that factor is not taken into account in my charts. $\endgroup$ – n_bandit Dec 23 '17 at 19:02
0
$\begingroup$

You could improve it by means of adding all variations of elements and compounds in the atmosphere to add the atmospheric color association to those elements/compounds. For example, a sky with sulfur would look like..., ozone with hight traces of carbon would look like...

$\endgroup$
  • 1
    $\begingroup$ With 120 elements in our periodic table and an infinite range of possible concentration and combinations, this suggestion sounds at least unpractical $\endgroup$ – L.Dutch Sep 8 '18 at 6:48
  • $\begingroup$ Not all 120 elements are natural, take that into account. Roughly there are about 10200 chemicals, now make concentrations a factor. The number is truly large however not infinite. He could potentially run computer algorithms to change the sky based upon chemical and elemental concentrations present. $\endgroup$ – user55117 Sep 18 '18 at 21:48
  • $\begingroup$ I am curious where did you found that 10200 number and if any chemist would agree with that. $\endgroup$ – L.Dutch Sep 19 '18 at 3:19
  • $\begingroup$ I agree with that number, what tells you I am not a chemist, would a Ph.D. in chemistry qualify? It's a well-known fact, about how many chemicals have been discovered up to date, then add all of them together. $\endgroup$ – user55117 Sep 23 '18 at 0:40
  • $\begingroup$ The CAS registry number "currently identifies more than 143 million unique organic and inorganic substances and 67 million protein and DNA sequences", way more than 10200, then $\endgroup$ – L.Dutch Sep 23 '18 at 3:56

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.