Celestia Forums

Since March, Sirius_Alpha has been doing a great job updating Celestia's exoplanet catalog. Actually I also have an exoplanet catalog generated from the NASA Exoplanet Archive, but theirs is better and more complete. One thing that the exoplanet catalog is missing, however, is new textures. We have great new star textures from FarGetaNik (which I'm hoping will be included in default Celestia), so the standard exoplanet textures should be updated too! Some ideas as to what these textures would look like:

For Jovians, in 2016 Andrew Tribick posted a list of temperature classes on Celestial Matters:

ajtribick wrote:

• Extremely hot Jupiter >2000K (TiO, not sure what this would look like...)
• Hot Jupiter 1700-2000K (relatively dark CaTiO₃/Al₂O₃ clouds)
• Hot Jupiter 1600-1700K (bright MgSiO₃ clouds on western hemisphere)
• Hot Jupiter 1500-1600K (mostly cloud-free except nightside and western limb)
• Hot Jupiter 1200-1400K (bright MnS clouds)
• Hot Jupiter 1000-1200K (mostly cloud-free except nightside and western limb)
• Hot Jupiter 700-1000K (chloride/sulfide clouds, maybe cirrus-like?)
• Warm Jupiter 350-700K (cloud free, Sudarsky class III)
• Temperate Jupiter 250–350K (small amount of water clouds? Or just keep as class III?)
• Temperate Jupiter 200-250K (water clouds, Sudarsky class II)
• Cool Jupiter 150-200K (yellowish from NH₄SH photochemistry? Or just keep as class II?)
• Cool Jupiter <150K (ammonia clouds, Sudarsky class I)

For Super-Earths/Neptunes, I'm not really sure what the textures should look like, but there are a lot of these planets and we need more than just the venuslike texture. Cold planets (<90K according to Andrew) would probably resemble Uranus and Neptune. Other planets might resemble their Jovian equivalents? Or if in the habitable zone, they might be panthalassas?

For terrestrials, the asteroid texture can still be used for the very smallest planets, which are likely dead, airless rocks (although a new asteroid texture would be nice...Askaniy?), but most terrestrial planets will need different textures. Here are my guesses:

• Hot planets (interior to the habitable zone) would probably either lose their atmospheres like Mercury (these would have the asteroid texture) or have runaway greenhouse effects like Venus, with the mass cutoff being...somewhere between these two planets. Some Venus analogs end up classified as super-Earths due to their density (e.g. TRAPPIST-1 b); I'm not sure what to do about this. Of course the hottest planets would be molten.
• Planets in the habitable zone shouldn't all have textures resembling Earth. Maybe more like Mars, or something between Earth and Mars.
• Cold planets might look like the moons of Jupiter, or Pluto, or a more icy Mars. Then again they might look more like Titan. There are several types of worlds here, but what are the cutoffs between them?

Of course, many textures exist for most of these planet types, but if they're going to be included with the Celestia distribution as standard exoplanet textures, they must be free use, which not all Celestia addons are. And there aren't textures for some planet types, such as many of the Hot Jupiters.

Edit: and this is my 100th post!

I mean, for a start, for ocean planets you just need a blue background and some spec+clouds.

Second, well i mean we don't have many artists. Not that many active artists anyway.
Snowfall did Haumea and Kexitt did Celestia Origin's exoclass textures - problem is they're not active (Snowfall SE'd and Kexitt=no account afaik)
Also Askaniy's field of specialty is more of image processing than texture making, most of his textures are based off of JohnVV's or Bjorn Jonsson so i mean.

I totally support what you're saying though, I really wish our exoplanet textures were better.

or.... y'know..... you could ask me.....
The problem with my gas giant textures is that they're not really turbulent and they are monochromatic, so i can't do the special textures with swirly colors and all that. Then again they might be appropriate for the mini neptunes and such idk.

Also another problem is that I don't really know what each class is supposed to look like; like is it green, blue, red, cream colored, or the color of a freshly opened can of pashtet? Who knows?
also join the discord server

But idk how we are going to implement the textures, but it definitely sounds like a good idea to me.

I've also wondered about this issue. IMHO it's one of the hardest things to accurately depict in an exoplanet catalogue, the other being rotations.

I've found two NASA articles about this that are kind of useful. This one shows Hot Jupiters from 1000 to 2100 K. The second image is actually really nice, but the problem is it's only a press release, so it's not as rigorous as a scientific paper would be. Also, the arXiv study doesn't go into too much detail about colors as I hoped I would. (Or maybe it does, but spectra are not my area of expertise, so I wouldn't know.)

There's also this one which points out that many ultra-hot Jupiters would be so hot that they'd glow an orange color from black-body radiation. I think this should actually simplify the colors for the hottest planets, since I suspect the radiation would totally overpower the actual color, at least from looking at the images.

fyr02 wrote:I mean, for a start, for ocean planets you just need a blue background and some spec+clouds.

Yeah, but I don't know which planets should be rendered as ocean planets. If any.

fyr02 wrote:Second, well i mean we don't have many artists. Not that many active artists anyway.
Snowfall did Haumea and Kexitt did Celestia Origin's exoclass textures - problem is they're not active (Snowfall SE'd and Kexitt=no account afaik)
Also Askaniy's field of specialty is more of image processing than texture making, most of his textures are based off of JohnVV's or Bjorn Jonsson so i mean.

Older textures can be used, the problem is that textures included with the Celestia distribution need to be free use. When I mentioned Askaniy I was referring to the fact that he's supposed to be updating the models and textures in the GitHub repository.

fyr02 wrote:or.... y'know..... you could ask me.....
The problem with my gas giant textures is that they're not really turbulent and they are monochromatic, so i can't do the special textures with swirly colors and all that. Then again they might be appropriate for the mini neptunes and such idk.

I was asking everyone, including you. Standard, generic exoplanet textures don't need to be very detailed. Your textures are good, they can be used as long as they're compatible with the license.

fyr02 wrote:Also another problem is that I don't really know what each class is supposed to look like; like is it green, blue, red, cream colored, or the color of a freshly opened can of pashtet? Who knows?

Sudarsky class 1 and 2 gas giants look like the existing textures for them. Cloudless gas giants (including class 3) are blue. For the hot Jovians...probably something like the pictures in the articles Luke linked.

Edit: Asked on the Discord which I have now joined; posting here for visibility. Where can I get non DDS versions of Kexitt's exo-classX textures? I don't see them on DeviantArt. Also does anyone know what the licenses for them are?

Celestia Origin has better quality versions of the existing textures. I am in favor of both Celestia Origin's good quality textures, the existing ones, and the old, long-removed "gasgiant" texture.

Joey P, this was what I was talking about. Kexitt designed the CO exoclasses and we are looking for similar quality versions of other gasgiant types.

Oh, I get it. But why don't we try making some good-quality textures out of poor-quality images, such as this one of 2M1207 b:
https://upload.wikimedia.org/wikipedia/commons/0/ ... irst_image_of_an_exoplanet.jpg

I really appreciate the kind words, StevenSpheres. Another possibility to expand out the range of textures are the possibility of transitional textures. Of course the insolation a planet receives, for the simple case of a non-inclined planet (which I assume we'll be assuming, lacking information to the contrary) scales with latitude, φ as cos φ. Consider a Sudarsky Class III planet just inward of the "habitable zone." You might imagine it would have water cloud condensation at its polar areas. So you could imagine a III-II transition texture, where the polar areas resemble a Sudarsky Class II planet, and the equatorial and tropical region are more Sudarsky Class III. We see something similar going on at Saturn, where the polar areas took on a more Uranus-like appearance. I believe this is from ammonia condensating out of the atmosphere at cooler temperatures but chemistry is really not my thing so I'm very open to being shown otherwise.

SevenSpheres wrote:For Super-Earths/Neptunes, I'm not really sure what the textures should look like, but there are a lot of these planets and we need more than just the venuslike texture. Cold planets (<90K according to Andrew) would probably resemble Uranus and Neptune. Other planets might resemble their Jovian equivalents?

I strongly suspect that super-Earths / mini-Neptunes would look similar to, or identical to gas giant planets, but I don't know if the belt/zone structures do, and I don't know how to make an educated guess based on Juno-Jupiter or Cassini-Saturn data on the depths of the counterrotating atmospheric structure.

With super-Earth and mini-Neptune planets appearing to have H-He dominated upper atmospheres, the upper atmospheres of these planets may represent nothing more than higher-metallicity analogues to Jupiter and Saturn - or maybe not even higher metallicity, see GJ 3470 (apologies for the link, I'm not at my main computer so I don't have the links to actual literature easily accessible).

Ultimately, venuslike.jpg seems to have been chosen for this task at a time when
a) It was not yet fully realized that the planets we now call super-Earths are actually low-density mini-ice giants. It made sense to consider GJ 581 c to be a giant terrestrial planet, pre-Kepler. We now know that these planets are not venus-like.
b) It was more important to distinguish the planet visually from the gas giant planets that were known, to indicate that this is not a gas giant planet.

SevenSpheres wrote:Or if in the habitable zone, they might be panthalassas?

We need to understand that these mini-Neptunes likely have a range of pressure-temperature-water abundance profiles that would make such predictions about the phase state of water a bit outside our ability to confidently predict. The reason Neptune is not an ocean planet is because it is too hot and too dry (i.e., where the P-T parameter space allows it, the water abundance is too low), but you wouldn't know that just looking at its insolation, which is really all we have for these extrasolar planets. As you point out in your response to fyr02, I do not think attempting to render ocean planets is advisable at this time, especially given how poorly such planets are understood, and the fact that not a single one has been confidently identified.

SevenSpheres wrote:Hot planets (interior to the habitable zone) would probably either lose their atmospheres like Mercury (these would have the asteroid texture) or have runaway greenhouse effects like Venus, with the mass cutoff being...somewhere between these two planets.

Okay, but where is that 'somewhere'? We know so little about this right now. M dwarfs, for example, require a lot of time to contract to their equilibrium Main Sequence stellar luminosity. Even planets in the "habitable zone" of such stars may have spent 100 Myr, or 1 Gyr for the coolest M dwarfs, being interior to the HZ as the HZ slowly crept inward. There is a legitimate debate about whether or not terrestrial-mass planets in and inward of the habitable zone retain their atmospheres, or if outgassing of secondary atmospheres is a common thing, etc. Clearly there's a mass/insolation regime where planets hold onto their atmospheres, but it's not clear where that is. This is the kind of information that we will be answered by TESS, ground-based radial velocity surveys and JWST.

I would be very interested in revisiting this conversation in 2030 where we have a statistically interesting sample of extrasolar terrestrial-mass planets with atmospheric constraints.

SevenSpheres wrote:Edit: and this is my 100th post!

Congratulations!

Something I'd like to do for extrasolar planet night sides is to have a range of night side textures of increasing brightness, representing different temperatures. Some planets are efficient at heat redistribution to their night side, while others aren't. There's an interesting paper here (Fourtney, et al. 2007) that has really guided a lot of my view on that sort of thing. More recently it was discovered that for a lot of hotter hot Jupiters, the night side temperatures tend to be roughly about the same up to a certain point where molecular hydrogen breaks down and the nightside temperature increases again (Keating, et al.).

It will be interesting to see what ultimately comes out of the HST Panchromatic Survey (lecture on the topic).

Joey P. wrote:Oh, I get it. But why don't we try making some good-quality textures out of poor-quality images, such as this one of 2M1207 b:

Because in that image, 2M1207b is an unresolved dot (in the infrared!), and the most you can get from it is a 2n x n image that is a uniform colour.

Sirius_Alpha wrote:I really appreciate the kind words, StevenSpheres.

It's "SevenSpheres", which is supposed to be a reference to the TRAPPIST-1 system. Not "Steven".

Sirius_Alpha wrote:I strongly suspect that super-Earths / mini-Neptunes would look similar to, or identical to gas giant planets, but I don't know if the belt/zone structures do, and I don't know how to make an educated guess based on Juno-Jupiter or Cassini-Saturn data on the depths of the counterrotating atmospheric structure.

So the same as the gas giant textures, but less "banded".

Sirius_Alpha wrote:As you point out in your response to fyr02, I do not think attempting to render ocean planets is advisable at this time, especially given how poorly such planets are understood, and the fact that not a single one has been confidently identified.

K2-18 b maybe? But yeah, there's no point in making textures for panthalassas since they would be cloud-covered anyway.

Sirius_Alpha wrote:Okay, but where is that 'somewhere'? We know so little about this right now. M dwarfs, for example, require a lot of time to contract to their equilibrium Main Sequence stellar luminosity. Even planets in the "habitable zone" of such stars may have spent 100 Myr, or 1 Gyr for the coolest M dwarfs, being interior to the HZ as the HZ slowly crept inward. There is a legitimate debate about whether or not terrestrial-mass planets in and inward of the habitable zone retain their atmospheres, or if outgassing of secondary atmospheres is a common thing, etc. Clearly there's a mass/insolation regime where planets hold onto their atmospheres, but it's not clear where that is. This is the kind of information that we will be answered by TESS, ground-based radial velocity surveys and JWST.

I would be very interested in revisiting this conversation in 2030 where we have a statistically interesting sample of extrasolar terrestrial-mass planets with atmospheric constraints.

I'm just guessing. I may be completely wrong. But the textures should be based on whatever the current "best guess" is. I don't like the asteroid texture for terrestrial exoplanets; the worlds in the Solar System are more...dynamic.

Sirius_Alpha wrote:Something I'd like to do for extrasolar planet night sides is to have a range of night side textures of increasing brightness, representing different temperatures. Some planets are efficient at heat redistribution to their night side, while others aren't. There's an interesting paper here (Fourtney, et al. 2007) that has really guided a lot of my view on that sort of thing. More recently it was discovered that for a lot of hotter hot Jupiters, the night side temperatures tend to be roughly about the same up to a certain point where molecular hydrogen breaks down and the nightside temperature increases again (Keating, et al.).

It will be interesting to see what ultimately comes out of the HST Panchromatic Survey (lecture on the topic).

Yeah, hot Jupiters are the best known exoplanets, and they have the fewest textures for them. I would really like to see a set of hot Jupiter textures. I probably couldn't make them though; I have no idea how to make textures from scratch, I usually just depth merge textures that already exist and change the colors. Can anyone here make some hot Jupiter textures based on the articles linked by ajtribick, LukeCEL, and Sirius_Alpha?

SevenSpheres wrote:It's "SevenSpheres", which is supposed to be a reference to the TRAPPIST-1 system. Not "Steven".

I'm terribly sorry about that.

Sirius_Alpha wrote:I'm terribly sorry about that.

It's fine!

Here's a quick and dirty grayscale gas planet texture that i made... perhaps it can work?

Possibly for a mainly Helium gas giant.

Added after 36 seconds:
Possibly for a mainly Helium gas giant.

In another community, specifically the Orion's Arm community, a member has made a nice list of cloud types that could form at different temperatures for use in that setting, and I think it will be useful here as well. I'm currently asking for their permission to put it here.

EarthSolar wrote:In another community, specifically the Orion's Arm community, a member has made a nice list of cloud types that could form at different temperatures for use in that setting, and I think it will be useful here as well. I'm currently asking for their permission to put it here.

Like Andrew Tribick's list?

I do have some possible textures that I've been posting on the Discord server. Also, it was suggested by pedro_jg on Discord - and Fridger on Celestial Matters - to automatically give appropriate textures to exoplanets based on temperature, mass, and density. I would very much prefer that the code for doing this be somewhere user modifiable, such as in celestia.cfg. The more that's user modifiable, the better!

Added after 1 hour:
Would also be nice to have new brown dwarf textures, since I'm not sure about the licenses for the T and Y dwarf textures I'm currently using (one unclear, the other missing).

Helium gas giants are actually a suspected, eventual outcome of white dwarf stars:
https://en.wikipedia.org/wiki/Helium_planet#White_dwarf_remnants

Joey P. wrote:Helium gas giants are actually a suspected, eventual outcome of white dwarf stars:

I do wonder what the planetary mass objects in those "black widow" systems would look like. In my pulsar planets addon I just used Edasich's textures. Probably as Luke said above they would just be a solid color from the extremely high temperature.

Alright, I've received the owner's permission to post it here. The list, as part of their attempt to overhaul the planetary classification system in OA, is here. Check the second sheet.

EarthSolar wrote:Alright, I've received the owner's permission to post it here. The list, as part of their attempt to overhaul the planetary classification system in OA, is here. Check the second sheet.

Wow that is very detailed! I assume the colors of the table cells are (at least close to) the actual expected cloud colors?

I assume so. Sad they didn't include albedo, but that's expected...