Celestia Forums

I have created some files for an updated extrasolar planets catalogue. I will be updating it as discoveries are made. It contains the following files:
> One .ssc file for the confirmed planets.
> One .ssc file for the unconfirmed planets.
> One .stc file for the star systems with confirmed planets.
> One .stc file for the star systems without confirmed planets.
> Several scripts to mark planets by discovery method, or by TESS/Kepler/Kepler-K2.

Download link to most recent version (02 December 2024):
viewtopic.php?p=163303#p163303

The information contained below is outdated, but it is left for historical reasons. This project has evolved and improved considerably in the years since it started.

The orbits of the planets are converted from using the reference plane of the sky to Celestia's internal reference plane through the formulae given in Grant Hutchison's starorbs.xls. spread sheet. For systems with a combination of transiting and non-transiting planets, the non-transiting planet was given an inclination equal to the transiting planet, plus five degrees. I may change this logic later to only do so if having the same inclination produces an orbit with an impact parameter of b < 1. Some circumbinary planets do not have known inclinations, yet their hosts are eclipsing binary stars. For most of them, I have simply not defined the stellar hosts as an eclipsing binary system (the system is thus coplanar with the planet), though for non-transiting Kepler circumbinary planets, they are rendered as eclipsing binaries but non-coplanar with the planet orbit (whose inclination is not declared). I may normalize this in the future. (Fixed!)

Otherwise, for planets with orbit inclinations determined astrometrically, only the planet with a measure inclination has been modified to reflect that. This makes some systems look atrocious (EPS Eri, GJ 676 A). I may rework the logic on this as well to assume system coplanarity. This will become very important as Gaia reveals intermediate-period planets around stars with already-known close-in RV planets. (Fixed!)

There are many extrasolar planets in multi-stellar systems. For systems where the components have known masses and separations, a simple binary stellar orbit was constructed for it, though I acknowledge that for some systems, it may place stellar components at incorrect position angles and separations. For multi-stellar systems where I could not find the mass ratios or separations, yet RA and Dec coordinates are available for the secondary component(s), I simply defined the two stars separately as nearby single stars. For multi-stellar systems where I could not find enough information to do either, I simply declared the host star as a single star. In all cases, the stars are given 'A', 'B', etc in their identifier.

The microlensing systems still need work. (Fixed!)

Often, especially for Kepler-# and KOI-# stars, spectral types are not known. If this is the case, I estimated the spectral type from the star's effective temperature according to the relation given by Eric Mamajek here.

Planets with masses above 3000 Earth-masses are assigned a brown dwarf texture. This is true even the planet's mass is known only as an upper limit. There's a few cases where this is particularly bad, such as at Kepler-453,Kepler-27, when clearly sub-Jovian planets are represented as brown dwarfs. This will be fixed in the next update. (Fixed!)

Star distances where possible have been determined from Gaia DR2 parallaxes. AppMag values (V magnitudes) are not known for quite a few stars. In such cases, they have been estimated from the 2MASS JHK fluxes. For stars where parallaxes or distances are unknown, neither the star nor its planetary system are included in this add-on. This is limited to just a few KOI systems whose planets are unconfirmed anyway, so it shouldn't be particularly obvious.

Finally, this add-on is intended to represent as complete an exoplanet catalogue as possible. You must remove both the extrasolar.ssc and extrasolar.stc files from your data folder, otherwise you will have duplicate planet definitions.

I'm open to any ideas for improvement that fit the spirit of representing the extrasolar planets in a scientifically accurate way, within the constraints of what is known.

Download Link here (01 Aug 2019)

You need to switch to the advanced editor ("Full Editor & Preview" button under a textarea) to have access to attachment upload.

Thanks. It's there now, but wasn't as my post was awaiting moderation.

I am pleased to announce that I have completed this updated extrasolar planets catalogue. I will provide regular updates as new planets are discovered, especially as new TESS planet candidates are added or old ones discarded as false-positives.

2019 May 06
- Numerous Kepler KOI False Positives have been removed.
- Systems fixed: KOI-1725, FL Lyr,
- Added planets and updates to known planetary systems in arXiv:1904.01573
- Updated reference for MWC 758b and added two uncnonfirmed planets (arXiv:1903.06537).
- Added TESS planet candidates.
- Added Kepler-K2 planet candidates.
- Microlensing planetary systems fixed.
- Corrected a bug in my code where stars with negative declinations were given incorrect ascending nodes in Celestia's reference frame.
- Spin-orbit (mis-)alignments from literature are taken into account. Retrograde planets are now rendered as retrograde. For systems without measured spin-orbit angles, stellar rotation axes are aligned with planets' orbital axes.
- Stellar oblateness has been estimated from v sin i measurements through the methods given in this document (pdf).
- Fixed issue where upper-mass limits in the super-Jovian regime were used to justify rendering sub-Jupiter-sized planets as brown dwarfs.
- Proxima c updated.
- TESS single-transit candidate planets are assigned a period of 50 days. Arbitrary, but I'm not sure what else to do for now. (update 07 May 2019)

6897 planets.
4281 confirmed.
2616 unconfirmed.

Download link: here.

NICE! This should be considered for the next Celestia Origin update

Lafuente_Astronomy wrote:NICE! This should be considered for the next Celestia Origin update

Not a fact at all. The quality of the base still requires a lot of checks.

Nice work!

Wow, this is really cool. Yours has a lot of advanced stuff like spin-orbit alignment, as well as well as stellar oblateness from v sin i. I am a bit jealous... besides the fact that I was also working on an exoplanet catalog like yours, but you beat me to it

P.S. What source did you use for your exoplanet catalogue? I used the NASA Exoplanet Archive.

Thank-you all for your kind words.

LukeCEL wrote:What source did you use for your exoplanet catalogue? I used the NASA Exoplanet Archive.

I made extensive use of the NASA Exoplanet Archive, as well as the Exoplanet Follow-up Programme (ExoFOP) for Kepler-K2 and TESS candidates. I've also made considerable use of literature on arXiv. For distances, Gaia DR2 data is accessible here.

It's already out of date, I'm afraid. There's 52 new TESS candidates that have been added to the ExoFOP since yesterday. They'll be in the next update.

Good job, Sirius.
I line myself up with LukeCEL in the team of the "bit jealous people", because I was also working in my exoplanet catalog.

We did somethings in common like, for example, using Eric Mamajek's table to estimate the spectral types from effective temperatures (at least for the main sequence stars).

I'm curious: how did you calculate the Spin-orbit (mis-)alignments, given the orbital parameters and the lambda, so that they apeared correct in Celestia? This is something that I haven't figured out yet.

I noticed that some of the distances to stars in the stars.stc file are in parsecs, not in light years as required in Celestia. I haven't checked all of them, but you can see it, for example, in Kepler-1638.

Best regards.

Fafers_br wrote:I line myself up with LukeCEL in the team of the "bit jealous people", because I was also working in my exoplanet catalog.

You too? Funny how so many people were working independently of each other.

That being said, I still plan on finishing up and releasing mine soon. Textures are a problem though, we should have a discussion on this.

Fafers_br wrote:I'm curious: how did you calculate the Spin-orbit (mis-)alignments, given the orbital parameters and the lambda, so that they apeared correct in Celestia? This is something that I haven't figured out yet.

Great question! Basically, I determined the planetary system's ecliptic plane (if there were transiting planets, it was set as coplanar to the innermost transiting planet, otherwise as might be the case for multi-planet systems from direct imaging or astrometry, it was set as the average orbit plane). The ecliptic plane was transformed into Celestia's system, then used to write the relevant information in the .stc file.

For the planet orbit, I went back to the original planetary orbit plane and simply added the spin-orbit (mis-)alignment angle λ to the planet's ascending node Ω. Then I transformed this new plane into Celestia's system and used that to write the planet's orbit.

Determination of the ecliptic plane:

Spoiler

Code: Select all

                // First, determine the ecliptic plane for the system, ignoring spin-orbit (mis-)alignments.
                for (int i = 0; i < StarList.Count; i++)
                {
                    bool WriteStarRotation = true;
                    OrbitPlane Ecliptic = new OrbitPlane(); // Needed for transiting planet systems.
                    if (has3DOrbit)
                    {
                        // If there is a transiting planet, we will use it's orbit to define the ecliptic.
                        bool HasEcliptic = false;
                        for (int j = 0; j < PlanetList.Count; j++)
                        {
                            string Host = PlanetList[j].Host.Substring(PlanetList[j].Host.Length - 1, 1);
                            if (PlanetList[j].hasTransits && Host == StarList[i].Host)
                            {
                                Ecliptic = new OrbitPlane(CurrentSystem.RA, CurrentSystem.Dec, PlanetList[j].inc, PlanetList[j].Ω, PlanetList[j].ω);
                                HasEcliptic = true;
                                break;
                            }
                        }
                        if (!HasEcliptic)
                        {
                            // Otherwise, just take from the system average.
                            double µinc = 0; // Average inclination.
                            double µΩ = 0; // Average ascending node.
                            double µω = 0; // Average longperi.
                            double npl = 0; // Number of planets.
                            foreach (Planet P in PlanetList)
                            {
                                string Host = P.Host.Substring(P.Host.Length - 1, 1);
                                if (Host != StarList[i].Host)
                                    continue;
                                µinc += P.inc;
                                µΩ += P.Ω;
                                µω += P.ω;
                                npl++;
                            }
                            µinc /= npl;
                            µΩ /= npl;
                            µω /= npl;
                            Ecliptic = new OrbitPlane(CurrentSystem.RA, CurrentSystem.Dec, µinc, µΩ, µω);
                            if (npl == 0) // No data.
                                WriteStarRotation = false;
                        }
                    }

                    // Terrific. Now write the star.
                    WriteStar.WriteLine(CurrentSystem.getPrefixNum() + CurrentSystem.getNames());
                    WriteStar.WriteLine("{");
                    WriteStar.WriteLine(CurrentSystem.getRA());
                    WriteStar.WriteLine(CurrentSystem.getDec());
                    WriteStar.WriteLine(CurrentSystem.getDist());
                    WriteStar.WriteLine(StarList[i].getAppMag());
                    WriteStar.WriteLine(StarList[i].getSpectral());
                    if (StarList[i].Radius != 0)
                        WriteStar.WriteLine(StarList[i].getRadius());
                    if (has3DOrbit && WriteStarRotation) // Let's define the orientation of the star.
                    {
                        double Prot = StarList[i].Prot; // Do we have a rotation period?
                        if (Prot == -1) Prot = 25; // Guess if not.
                        Ecliptic.Transform(); // Convert to Celestia reference frame.
                        WriteStar.WriteLine("   UniformRotation{");
                        WriteStar.WriteLine("      Period " + Math.Round(Prot, 3));
                        WriteStar.WriteLine("      Inclination " + Math.Round(Ecliptic.inc, 2));
                        WriteStar.WriteLine("      AscendingNode " + Math.Round(Ecliptic.Ω, 2));
                        WriteStar.WriteLine("   }");
                    }
                    if (StarList[i].f > 0.015) // Below which it isn't worth rendering.
                    {
                        double Oblateness = Math.Round(1 - StarList[i].f, 3);
                        WriteStar.WriteLine("SemiAxes [ 1 " + Oblateness + " 1 ]");
                    }
                    WriteStar.WriteLine("}");
                    WriteStar.WriteLine("");
                }


Incorporate spin-orbit angle, λ.

Spoiler

Code: Select all

                    // Construct orbit.
                    OrbitPlane Orbit = new OrbitPlane(CurrentSystem.RA, CurrentSystem.Dec, P.inc, P.Ω, P.ω, P.M, P.P, P.Tperi);
                    if (P.λ != 0)
                        Orbit.Ω += P.λ; // Adjust for spin-orbit (mis-)alignment.
                    Orbit.Transform(); // Convert to Celestia coordinates.


The actual plane transformation.

Spoiler

Code: Select all

        public void Transform()
        {
            // Do reference frame transformations.
            double b = Tools.ConvertDegtoRad(90 - inc);
            double α0 = Tools.ConvertDegtoRad(RA) - Math.PI;
            double δ0 = 0 - Tools.ConvertDegtoRad(Dec);
            double i270 = Tools.ConvertDegtoRad(Ω); // Orbit plane adjusted for spin-orbit alignment.
            double o270 = (Ω - 270) / 180 * Math.PI;
            double ε = Tools.ConvertDegtoRad(23.4392911);
            double α = Math.Atan(Math.Cos(b) * Math.Cos(i270) / (Math.Sin(b) * Math.Cos(δ0) - Math.Cos(b) * Math.Sin(δ0) * Math.Sin(i270))) + α0;
            if (Math.Sin(b) * Math.Cos(δ0) - Math.Cos(b) * Math.Sin(δ0) * Math.Sin(i270) < 0)
                α += Math.PI;
            double δ = Math.Asin(Math.Cos(b) * Math.Cos(δ0) * Math.Sin(i270) + Math.Sin(b) * Math.Sin(δ0));
            double λ = Math.Atan((Math.Sin(α) * Math.Cos(ε) + Math.Tan(δ) * Math.Sin(ε)) / Math.Cos(α));
            if (Math.Cos(α) < 0)
                λ += Math.PI;
            double β = Math.Asin(Math.Sin(δ) * Math.Cos(ε) - Math.Cos(δ) * Math.Sin(ε) * Math.Sin(α));
            double δo = Math.Asin(Math.Cos(δ0) * Math.Sin(o270));
            double αo = Math.Atan(Math.Cos(o270) / -Math.Sin(δ0) / Math.Sin(o270)) + α0;
            if (-Math.Sin(δ0) * Math.Sin(o270) < 0)
                αo += Math.PI;
            double λo = Math.Atan((Math.Sin(αo) * Math.Cos(ε) + Math.Tan(δo) * Math.Sin(ε)) / Math.Cos(αo));
            if (Math.Cos(αo) < 0)
                λo += Math.PI;
            double βo = Math.Asin(Math.Sin(δo) * Math.Cos(ε) - Math.Cos(δo) * Math.Sin(ε) * Math.Sin(αo));
            double dExc = Math.Acos(Math.Cos(βo) * Math.Cos(λo - λ - Math.PI / 2));
            if (βo < 0)
                dExc *= -1;
            // Update orbit parameters.
            inc = 90 - β / Math.PI * 180;
            Ω = (λ / Math.PI * 180 + 90) % 360; // Base orbit plane.
            ω = (ω + 86400 / Math.PI * 180) % 360;
            M = (((2000 - (Tperi / 365.25)) / P) % 1) * 360;
        }
       

Fafers_br wrote:I noticed that some of the distances to stars in the stars.stc file are in parsecs, not in light years as required in Celestia. I haven't checked all of them, but you can see it, for example, in Kepler-1638.

Great catch! the overwhelming majority of systems' input data had parallaxes, but this wasn't one of them, so it used a separate method for determining distance that didn't make the correct unit conversion. I'm really appreciative that you've found this and I will post an update with the fix.

LukeCEL wrote:Textures are a problem though, we should have a discussion on this.

It's an interesting problem. I've stuck with the default Celestia texture scheme because it doesn't require anyone to download more than is necessary to view the new catalogue, and seems reasonably objective. Of course we know now that super-Earths on the larger side of the Fulton Gap are more Neptune-like in composition than Earth-like, so the "Venuslike" texture ends up being a texture for fluid planets now -- it was originally used for Neptune-like planets originally anyway (GJ 581 c, GJ 436 b, etc).

There might be a use for a lava-type texture to be used for, for example, Kepler-10b, but I worry about the empiricism aspect. Do we just cut it off at a certain temperature? (i.e., planets > 1500 K are assigned such a texture). We do that for the current Sudarsky classification anyway. So I guess I don't see a huge problem. Ultimately I want to avoid falling into the realm of speculation as much as possible, preferring instead to texture planets in rather broad ways to signify what is known about them. To that end I will certainly not be assigning habitable zone terrestrial planets textures that make them appear to have lush fields with kittens running through the grass swatting at butterflies.

That being said, I've been considering making a separate stand-alone addon for just this. Spitzer 4.6µm phase curve map texture for HD 189733 b, a cloud map for Kepler-7 b, etc. Rather planet-specific improvements that do not really fit into a uniform catalogue of exoplanets. Another task I'd like to look into is using transmission spectroscopy data (quite a bit of it is piling up on the NASA Exoplanet Archive) to create atmosphere definitions that approximate true colour.

Added after 25 minutes 53 seconds:

2019 May 15
- 52 TESS candidate planets added: TOI-640 through TOI-690.
- 13 TESS candidate planets have been determined to be false positives since the last update and removed.
- Added TOI-# ID's for 10 known transiting planets in TESS Sector 9 (WASP-66, WASP-106, HATS-19, etc).
- Added KMT-2018-BLG-1990L b, K2-133 e, HR 858 bcd, KMT-2018-BLG-1292L b.
- Removed LkCa 15 planetary system (link).
- With an improved programme able to translate transit-depths into radii, added some missing Kepler-K2 candidates.
- Tidal equilibrium rotation periods: Pseudosynchronous for fluid planets. 1:1 or 2:3 spin-orbit resonance for solid planets.
- Fixed distance error in some systems where declared distance should have been in light years rather than parsecs.
- Added a comment on the texture line clarifying the type of planet.

6931 planets.
4286 confirmed.
2645 unconfirmed.

Sirius_Alpha wrote:Great question! Basically, I determined the planetary system's ecliptic plane (if there were transiting planets, it was set as coplanar to the innermost transiting planet, otherwise as might be the case for multi-planet systems from direct imaging or astrometry, it was set as the average orbit plane). The ecliptic plane was transformed into Celestia's system, then used to write the relevant information in the .stc file.

Thank you for sharing this, you were very kind! I'll study your math this weekend (I hope I'll understand it).

Sirius_Alpha wrote:I'm really appreciative that you've found this and I will post an update with the fix.

Glad to be of service!
By the way, I have another thing to appoint: while comparing some of your OGLE systems to mine (just out of curiosity about your choices of radius, spectral classes, etc) I noticed that the periods of the planets may not be quite right. Take, for example, OGLE-2015-BLG-1670L. I assume you used the Kepler's laws and simply calculated the period, given the star mass and the planet's semimajor axis, correct? Well, for the said system, M=0.55Ms and a=3.2AU. That should result in a period ~7.7yr, not 2.440641. I suspect that your code is, somehow, showing the results not in years, but in seconds (multiplied by 1E-8), because I saw the same "pattern" in two other systems I tested. Please, check it out.

LukeCEL wrote:You too? Funny how so many people were working independently of each other.

Well, I think we all feel compeled to keep the exoplanets in our Celestias up to date.
In the matter of textures, I agree with Sirius that by using the default ones you make it easier to share the Catalogue.

Best regards to all!

Fafers_br wrote:I assume you used the Kepler's laws and simply calculated the period, given the star mass and the planet's semimajor axis, correct?

Certainly. After checking into it, it turns out that I used a value for G of "6.67408 * 10e-11" rather than the correct 6.67408e-11. This would have applied to all planets without a specified orbital period (mostly microlensing and direct imaging planets). Idiot error on my part. Thank-you for catching it!

--

2019 May 16
- Fixed incorrect value for the gravitational constant (d'oh!).
- Added OGLE-2018-BLG-0596L b.

6932 planets.
4287 confirmed.
2645 unconfirmed.

LukeCEL wrote:I still plan on finishing up and releasing mine soon. Textures are a problem though, we should have a discussion on this.

Of course you decide. But personally, I expect from you first of all a new catalog of galaxies.
Sirius_Alpha's exoplanetary base already exceeds its analogues in many parameters. It must be recognized.

Art Blos wrote:Of course you decide. But personally, I expect from you first of all a new catalog of galaxies.

Yup. Exoplanets and Stars have already been dealt with. A galaxy catalog is needed, as well as stars within those galaxies, which LukeCEL has already started some time ago.

Lafuente_Astronomy wrote:A galaxy catalog is needed, as well as stars within those galaxies, which LukeCEL has already started some time ago.

My "stars within galaxies" add-ons are semi-fictional; they shouldn't really be in the main Celestia package. Sure, they represent real stars as accurately as possible. However, their actual positions in 3D space are completely wrong because we don't know the distance to them. Other parameters, like spectral type and apparent magnitude, are also highly approximate guesses.

Also, before I do a galaxy catalogue, I need to know which data sources to use. I think it is possible to add the UGC galaxies in Celestia, a bit less sure about MCG, and probably not PGC.

LukeCEL wrote:Also, before I do a galaxy catalogue, I need to know which data sources to use. I think it is possible to add the UGC galaxies in Celestia, a bit less sure about MCG, and probably not PGC.

While it's not wholly accurate but you can use the SDSS Catalog as well, since so far, it's the most massive and comprehensive galaxy catalog. And you can take some of the NGC and IC galaxies, provided they are not members of other catalogs. And provided a barycenter for galactic clusters can be made, you can also have the Abell Catalog.

Lafuente_Astronomy wrote:While it's not wholly accurate but you can use the SDSS Catalog as well, since so far, it's the most massive and comprehensive galaxy catalog.

Mmm, can you clarify? The SDSS is a huge survey, with many data releases and resulting catalogues. And I couldn't find any of them have galaxy morphological classifications, which are necessary to write galaxies in Celestia.

Lafuente_Astronomy wrote:And you can take some of the NGC and IC galaxies, provided they are not members of other catalogs.

I already plan to use Wolfgang Steinicke's Revised NGC/IC Catalogue, as it's what the original file is based on.

Lafuente_Astronomy wrote:you can also have the Abell Catalog.

I don't think so. The original Abell catalogue of galaxy clusters doesn't have individual galaxies. Although, speaking of galaxy barycenters, are we really planning to add galaxies orbiting other galaxies? This would be a daunting task.

LukeCEL wrote:Mmm, can you clarify? The SDSS is a huge survey, with many data releases and resulting catalogues. And I couldn't find any of them have galaxy morphological classifications, which are necessary to write galaxies in Celestia.

They are merely spots which indicate galaxy locations, hence the "Not wholly accurate" part. If I remember, Selden has an addon for that, and apparently Gaia Sky has galaxies from the SDSS included as well. However, because accuracy is important, if you feel that you shouldn't add them in Celestia, then feel free to not add them.

LukeCEL wrote:I don't think so. The original Abell catalogue of galaxy clusters doesn't have individual galaxies. Although, speaking of galaxy barycenters, are we really planning to add galaxies orbiting other galaxies? This would be a daunting task.

Well, galaxies do have movements right? There's already suggestions for star orbits, which would make stars orbit around a galactic barycenter, thus adding some more accuracy in the positions of stars and planets over time. If the star orbits can be done, perhaps galaxy movements and therefore, galaxy clusters can be done too. I think IllustrisTNG might help out on that one.