Extrasolar Planets (updated catalogue)

The home for finished/released addons.
Avatar
Art Blos M
Moderator
Posts: 600
Joined: 31.08.2017
Age: 27
With us: 2 years
Location: Volgodonsk, Rostov Oblast, Russia

Post #21by Art Blos » 18.05.2019, 08:45

LukeCEL wrote: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.
You can’t call your add-on fictional, precisely because the objects are real. Approximate (logically thought out) parameters are often found in other areas of astronomy. Just here we are faced with the limitations of science.
Founder and head of the project "Celestia Origin"

Avatar
Lafuente_Astronomy
Moderator
Posts: 357
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #22by Lafuente_Astronomy » 18.05.2019, 14:40

Art Blos wrote:You can’t call your add-on fictional, precisely because the objects are real. Approximate (logically thought out) parameters are often found in other areas of astronomy. Just here we are faced with the limitations of science.

Agreed. We may have to wait for more advance satellites and technology to find more accurate readings for those stars. But at least millions, if not billions of stars are already being recorded on other galaxies each. We just have to make do with what we have. After all, Celestia can always be modified, updated and upgraded for future astronomical discoveries.
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.

Avatar
Topic author
Sirius_Alpha
Posts: 45
Joined: 21.03.2019
With us: 5 months 28 days

Post #23by Sirius_Alpha » 01.06.2019, 04:21

At the risk of dereailing this thread about Galaxy catalogues :p

I've implemented a way to estimate distances based on the star's V mag, T_eff and R_star. Many missing Kepler candidates were able to be added as a result, though their distances may not be as accurate as hoped. Among all the planets I have in my notes, at the present time there are 13 missing unconfirmed planets, which are not included usually because there is neither a specified mass or radius (long-term RV trends, unconstrained TTV solutions, etc). There is also a TESS CTOI with three planet candidates that have no orbital parameters.

I've finally finished the Kepler-K2 candidate planets. Frustratingly, there are two dissimilar catalogues available, each with some planets the other lacks. Several planets in the NASA Exoplanet Archive and several in the ExoFOP. Because several objects flagged as "possible false positive" or "positive eclipsing binary" went on to become confirmed K2 planets, these tags were not used as a criteria to not add the system. Rather, only Kepler-K2 candidates whose ExoFOP pages have observation notes indicating a significant centroid offset, even-odd transit depth disparity, transit chromaticity, or other conclusive indicators of being a false positive were rejected. Additionally, I have excluded planet candidates that are likely false positives based on their stated parameters -- several have extremely high radii (> 25 R_J) or absurd transit depths (>60% in some cases!). Otherwise, sometimes SIMBAD will actually list an eclipsing binary type (e.g., W UMa for EPIC 212499835). This was also taken to be confirmation of it being a false positive.

There are many new objects listed as Community TESS Objets of Interest (CTOI) since the last update which are clearly eclipsing binary systems. They'll have two planet candidates at basically the same period with rather high transit depths. Either these are incorrectly passing through the TESS pipeline (somehow not getting caught by the odd-even transit depth check), or the CTOI list intentionally includes systems that are clearly eclipsing binaries. I have chosen not to include such systems.

I also discovered that I had two definitions for OGLE-2015-BLG-0051L. One from Hwang, et al. (2015) under the designation KMT-2015-1, and another from Han, et al. (2016). This was fixed, and it inspired me to write a script to cross-reference all IDs I had in my input data for these systems to look for additional duplicates. This allowed me to find three more: K2-68 / EPIC 206146957, Kepler-1651 / KOI-1725, and Kepler-482 / KOI-174. These, too, have been fixed.

As always, I'm open to suggestions for improvements, corrections.

Celestia_Screenshot.png
Planets by detection method. Pink = RV, Green = transit, blue = Microlensing.


2019 Jun 01
- 45 TESS candidate planets added (TOI-691 through TOI-728).
- 6 TESS Community TOIs added.
- 11 TESS candidate planets have been determined to be false positives since the last update and removed.
- Reworked AppMag. Comments to clarify if AppMag is calculated, from Kepler Kp, etc.
- Corrected error converting transit depths to radii in some systems.
- Asteroids now handled explicitly (for SDSS J1228+1040, EPIC 210605073).
- For systems with no known distance, it is now estimated photometrically if possible.
- For TESS Single-Transit candidates, a comment is now in the .ssc file noting the period is a guess.
- Added unconfirmed SWEEPS candidates.
- Finished adding missing Kepler-K2 candidates.
- Fixed HW Vir.
- Removed some duplicate systems.
- Removed KOI-376 (false positive).
- Added scripts to mark *all* confirmed or unconfirmed exoplanet systems.
- Added OGLE-2018-BLG-0532L, OGLE-GD-ECL-11388, DW UMa, numerous (new!) small planets from Kepler-K2 (Heller et al.), LSPM J2116+0234.
- Updated L 98-59, HD 113337.
- Corrected CoRoT-30 position.
- Stopped defining rotation periods for planets in synchronous rotation.
- Calculated missing inclinations from transit impact parameters where possible (458 planets).
- Fixed issue where circumbinary planets with unknown periods/semi-major axis would have only the primary star's mass used to determine the missing parameter. Affected systems: FL Lyr, FW Tau, HD 106906, HD 284149, KOI-2938, KOI-6518, OGLE-2007-BLG-349L, Ross 458, ROXs 42B, SR 12, VHS J125601.92-125723.9.
- Changed planet rotation definition from obselete RotationPeriod, Obliquity and EquatorAscendingNode to a UniformRotation{} code, as per request.
- All planets now have LunarLambert 0.5, as per request.
- An additional alias has been added in each planet definition line for just the planet's letter (b, c, d, etc), as per request.
- Added InfoURL's for stars, as per request.

7291 planets (+2 asteroids).
4317 confirmed.
2974 unconfirmed.

Added after 12 hours:
I sincerely apologise for this, but I just discovered that there was a major error in the .stc files in the addon I posted earlier today. It has been corrected. I'm deeply sorry for the inconvenience.
Attachments
CelestiaExoplanets_2019_06_01.rar
Celestia Exoplanet Catalogue (up to date as of 01 Jun 2019)
(589.2 KiB) Downloaded 92 times

Avatar
Lafuente_Astronomy
Moderator
Posts: 357
Joined: 04.08.2018
Age: 21
With us: 1 year 1 month
Location: Cebu City, Cebu Province, Philippines
Contact:

Post #24by Lafuente_Astronomy » 01.06.2019, 21:37

It's ok. Keep up the good work! As far as I'm concerned, your work is already a huge step towards making Celestia more accurate
Official Administrator of the Celestia Discord Server.
Invite: https://discordapp.com/invite/WEWDcJh
If you don't have a Discord account, register here: https://discordapp.com/register
Have a blessed day.

Avatar
Topic author
Sirius_Alpha
Posts: 45
Joined: 21.03.2019
With us: 5 months 28 days

Post #25by Sirius_Alpha » 13.06.2019, 01:27

TESS Sector 10 planets are out! -- at least the first batch of them, it seems that they come out in smaller groups after an initial large release. Furthermore, a lot of new planets in very nearby systems was reported (Wolf 359, Lacaille 9352, Struve 2398 to name a few). So there's plenty of reason to post an update.

The paper for HAT-P-69 and HAT-P-70 did something a bit rare, and specified the stellar oblateness. They're rapidly rotating stars so they would have been given a calculated oblateness in my programme anyway. This gave me the opportunity to compare calculation to reality and I find that my programme underestimates the stellar oblatenesses by a bit. For HAT-P-69, I calculate an oblateness of 0.975, they report 0.969. For HAT-P-70, I caculate 0.963, they report 0.957. I took their values of course, but it's worth noting that with a whopping set of 2 datapoints, it looks like my programme is consistently underestimating stellar oblateness. This is great, because I would like to have more obvious oblatenesses to look at. I'll wait for more data to come in to better understand this before making a change.

I discovered that the inclination angle of the rotation axes of stars relative to the plane of the sky hasn't been taken into account. Some planets are in-fact known to have a misalignment between their orbital axis and the stellar rotation axis, λ, despite neither a detection of the Rossiter-McLaughlin effect or Doppler tomographic shadow. This is done for transiting planets (with orbital inclinations of i~90°) by determining the inclination of the star's rotation axis, and noting it's significantly inconsistent with i~90°. For example. Kepler-408 b is a transiting Venus-sized planet in a mis-aligned orbit, and is in fact the smallest known misaligned planet (Kamiaka, et al. (2019)). Well it wasn't rendered that way until this update.

Apparently the ExoFOP no longer uses CTOI identifiers for Community TESS Objects of Interest. Now they just use the TIC ID. So I've made the change here as well.

For reasons I don't really understand, a couple transiting circumbinary planets (Kepler-16 and Kepler-35) were rendered as having retrograde orbits. This has been fixed a rather ad-hoc way (giving the stars negative orbital periods) since I don't know why the problem was there in the first place.

Wolf-359b.jpg
The inner planet at Wolf 359


- 60 TESS candidate planets added (TOI-729 through TOI-784).
- Added TOI-# ID's for known transiting planets in TESS Sector 10 (WASP-87, WASP-167, HATS-18, HATS-22, etc).
- 15 TESS Community TOIs added.
- Added new TESS candidate planets in known TOI systems (TOI-171.02, TOI-282.03, TOI-553.03).
- Obsolete KID (Kepler ID) catalogue identifiers replaced with KIC (Kepler Input Catalogue) (see Howell (2010).
- Obsolete CTOI (Community TESS Object of Interest) catalogue identifiers replaced with TIC (TESS Input Catalogue).
- Added PDS 70 c, KELT-24 b, HD 144844 (AB)b, though since the nature of the secondary star is unknown, there is no attempt to render the system as binary.
- Added/updated numerous nearby planetary systems from Doppler spectroscopy, removed GJ 229 Ab (Tuomi, et al).
- Updated HAT-P-69, HAT-P-70, CI Tau b.
- Removed Kepler-429, Kepler-70 (see Blokesz, et al.)
- Stellar rotation axis inclination angle relative to the sky-plane now defined for some systems.
- Fixed issue where some planetary systems had incorrect inclination.
- Corrected HAT-P-12 V mag.

7403 planets (+2 asteroids).
4371 confirmed.
3032 unconfirmed.
Attachments
CelestiaExoplanets_2019_06_12.rar
Celestia exoplanet catalogue, up to date as of 12 June 2019.
(584.93 KiB) Downloaded 70 times

Avatar
LukeCEL
Posts: 189
Joined: 26.09.2017
With us: 1 year 11 months

Post #26by LukeCEL » 14.06.2019, 02:56

LukeCEL wrote: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.

Change of plans. In the interest of not having two separate versions of what is essentially the same project, I'm not going to release my exoplanet catalogue after all. This change came after realizing 1) how busy I was (and still am), and 2) how little I knew about spin-orbit misalignment. I'm not too keen on learning that.

By the way, Sirius_Alpha, can I suggest a change to your "Stars_Binary_Exoplanets.stc" file? For stars of binary systems that are already subsystems, I recommend fixing the designation so the second letter is un-capitalized. So for example, "30 Ari Aa" instead of "30 Ari AA".

Avatar
Topic author
Sirius_Alpha
Posts: 45
Joined: 21.03.2019
With us: 5 months 28 days

Post #27by Sirius_Alpha » 14.06.2019, 04:33

My only concern about that is the eventuality that a planet will be discovered around such a star. Of course 30 Ari A is too tight a binary for this to actually occur, but if for example a planet were to be discovered orbiting 30 Ari Aa, what do you call it? Aab? That being said, I do recognize that lower-letter designations for stellar sub-systems isn't without precedent.

Avatar
gironde M
Posts: 386
Joined: 16.12.2016
Age: 66
With us: 2 years 9 months
Location: Montigny-Les-Metz, Grand Est, Lorraine, Moselle, France

Post #28by gironde » 14.06.2019, 10:37

For a planet b of 30 ARI Aa ===> 30 ARI Aa b
put a space between star and planet

:hi:

Avatar
Topic author
Sirius_Alpha
Posts: 45
Joined: 21.03.2019
With us: 5 months 28 days

Post #29by Sirius_Alpha » 01.07.2019, 04:05

01 July 2019

TESS Sector 11 planets are out! -- at least the first batch of them. See prior update for involved caveats. Also there are some new TESS planets from previous sectors. Teegarden's Star has planets of course.

Thanks to some assistance from Fabers_br, I've implemented correct transit times for transiting planets. The epoch and mean anomaly are now set such that the planet should transit at the epoch listed in its discovery paper. I still need to go through and verify that I have correct transit epochs for all transiting planets though so it's possible not all of the planets transit when they should. This will be fixed in the near-future, but at least the implementation is there.

I have changed the stellar designation scheme as suggested by LukeCEL. Stars like 30 Ari AA are now designed 30 Ari Aa.

I have also fixed a major embarrassing error in that the planets' orbital longitude of periapses were not correctly specified in the .ssc files because, of all things, a typo. This is hugely embarrassing.

Screenshot_1.png
HATS-1b Transiting at the correct transit epoch.

Screenshot_2.png
Transit epoch from the HATS-1b discovery paper.


- 48 TESS candidate planets added (TOI-785 through TOI-843).
- Added TOI-# ID's for known transiting planets in TESS Sector 11 (WASP-16, HATS-17, WASP-15, etc).
- 16 TESS candidate planets have been determined to be false positives since the last update and removed.
- 9 TESS Community TOIs added.
- K2 identifiers added to several EPIC systems (K2-297 through K2-308).
- Added Teegarden's Star b & c (Zechmeister, et al.), and KMT-2018-BLG-0029L.
- Updated LP 791-18, HD 271181, TOI-150, TOI-455, DS Tuc, HD 3167, OGLE2-TR-L9.
- Fixed issue where negative declinations were incorrectly converted from HMS to degrees.
- Fixed longitude of periapses...
- Implementation of correct transit times.
- Corrected Wolf 359's radius (it was 0.046 solar-radii in the base Celestia!).
- Fixed WASP-108 b texture.

7446 planets (+2 asteroids)
4379 confirmed.
3067 unconfirmed.
Attachments
CelestiaExoplanets_2019_07_01.rar
Celestia exoplanet catalogue up to date as of 01 July 2019.
(651.19 KiB) Downloaded 76 times

Avatar
Topic author
Sirius_Alpha
Posts: 45
Joined: 21.03.2019
With us: 5 months 28 days

Post #30by Sirius_Alpha » 01.08.2019, 01:57

TESS Sector 12 planets are out! And a lot from prior sectors. BD+04 701 is a (visual?) binary system with a TESS planet candidate around each star. I don't have data right now on the system to render it as a binary, rather than as two separate single stars. Though with as interesting as this is it'll probably be the target of later research anyway.

A certain likely-well-meaning 'citizen scientist' has gotten into the ExoFOP and started adding numerous CTOIs because "dips in light curve jpg suggests multiple planetary system," rather than offering any parameters for these planets. Based on the description, it looks like they're using a rather unscientific process of looking at the light curve by-eye, noting there are dots that are under the detrended baseline, and saying "Yep! That's a multi-planet system." It's horribly unscientific. I've sent an e-mail to David Latham, an (the?) administrator of the ExoFOP about this, asking for clarification and/or action to resolve this, and a few days later the CTOI submission system was disabled... In the mean time, just note that there are many supposed parameter-less CTOIs that have not been added.

EPIC 210605073 has been the 'other' extrasolar asteroid. But the star's nature has been somewhat ambiguous. Adams, et al. (2016) took a look at the system and found the colours to be consistent with a F-type star, but "no measurements of the metallicity, log g, or photometric extinction are available to better constrain the estimate." Table 2 in Adams, et al's analysis "assumed log g = 4.0." Consistent indeed with an F-type star. But there's also a paper out using SDSS DR7 suggesting EPIC 210605073 has a log g of 8.25. This bit of information was apparently not included in this analyses (or was rejected?). Sluijs & Eylen (2017) did a survey of white dwarfs, noted Adams, et al.'s analysis, stating only that "a recent photometric analysis suggests the star is (contaminated with) an F0 companion," without commenting on the SDSS DR7 data.

Between the choice of the log g measurement in SDSS DR7 and the argument based on photometry but assuming a log g of 4, it would seem that the reasonable choice would be the white dwarf model. But the transit duration as reported in the ExoFOP for EPIC 210605073 is just over 0.9 hours. This is inconsistent with the star being a white dwarf for any plausible orbit for the companion (assuming, of course, it is not an astrophysical false positive). Between a bad transit duration and a bad log g, it seems overwhelmingly more plausible that the log g is bad. Consequently, EPIC 210605073 b has been changed from an asteroid to a planet (or more accurately, the star was changed from a white dwarf to a F0 type star, and my programme scaled up the orbiting body based on it's transit depth accordingly).

There was a paper from Kruse, et al, presenting some 374 new planet candidates from Kepler-K2. Their work was to run the Kepler-K2 data through an algorithm that recovered transit-like signatures, apparently without a lot of regard for prior research done on these systems. As such, many of the new candidates presented in this work are known eclipsing binary stars. Consequently, some discretion was used in which systems were added. At K2-24, Petigura, et al. (2018) reported a M sin i = 54 M_e planet in a 428 day orbit, and this new paper reports a single-transit candidate at this system. I don't know if the two planets are the same but it's possible. I won't make any such assumption though.

Celestia_Screenshot.png
The locations of planets and candidate planets from the Kepler-K2 campaigns, viewed from far away.


01 Aug 2019
- 206 TESS candidate planets added (TOI-844 through TOI-1048).
- 8 TESS candidate planets have been determined to be false positives since the last update and removed.
- Added TOI-# ID's for known transiting planets detected by TESS (WASP-132, WASP-180).
- Added 273 new Kepler-K2 planet candidates from Campaigns 0-8 (Kruse, et al.).
- Added stellar companions to λ2 For, HD 33283, HD 107148, HD 108863, HD 170469, 18 Del, HD 4113, HAT-P-4, HATS-1. (Mugrauer, et al. 2014), HD 17926 (Vanderburg, et al. 2019), KELT-22 (Labadie-Bartz, et al. 2018), KELT-23 (Johns, et al. 2019),
- Added OGLE-2018-BLG-1011, K2-43 c, K2-198 c and d (Hedges, et al.), OGLE-2015-BLG-1649L, WASP-178, WASP-184, WASP-185 & WASP-192 (Hellier, et al.), K2-310 and K2-146 c.
- Added missing planet candidates from Kepler DR5 at Kepler-913, Kepler-1178, Kepler-1185, Kepler-1211, Kepler-1270,
- Added unconfirmed planets 2MASS J06101557+2436535 b, Kepler-210 d (from TTV),
- Updated Kepler-29, Kepler-36, Kepler-177 and KOI-1783 (Vissapragada, et al).
- Removed duplicate GJ 3323 and GJ 682 star definitions.
- Removed HD 10383 b and HD 16559 b. They were in the first version of its discovery paper, but the paper did not include it in its final version.
- KOI-2533.02, KOI-410.01, KOI-531.01, have been found to be false positives and removed.
- Fixed issue determining eccentric anomaly.
- Finished giving transiting planets correct transit times where possible (SWEEPS planets have no published T_transit, for example).
- Implemented correct T_Peri times for radial velocity planets where available.
- Changed disposition of several planets from Rowe, et al. (2014) to 'unconfirmed' in alignment with the NASA Exoplanet Archive.
- Changed structure of programatically generated .ssc and .stc files to conform to Celestia Origin standards. I will do this with the manually-produced Stars_Binary_Exoplanets.stc soon.

7938 planets (+1 asteroid)
4236 confirmed.
3702 unconfirmed.
Attachments
CelestiaExoplanets_2019_08_01.rar
Catalogue of extrasolar planets up to date through 01 August 2019.
(706.65 KiB) Downloaded 108 times

Avatar
Fafers_br
Posts: 9
Joined: 01.12.2018
Age: 44
With us: 9 months 16 days
Location: Belo Horizonte - Brazil

Post #31by Fafers_br » 06.08.2019, 22:12

Very good, Sirius!
:clap: :clap:

Avatar
Topic author
Sirius_Alpha
Posts: 45
Joined: 21.03.2019
With us: 5 months 28 days

Post #32by Sirius_Alpha » 31.08.2019, 23:15

TESS Sector 13 planets are out!

Sirius_Alpha wrote:BD+04 701 is a (visual?) binary system with a TESS planet candidate around each star. I don't have data right now on the system to render it as a binary, rather than as two separate single stars. Though with as interesting as this is it'll probably be the target of later research anyway.

They're both false positives. Oh well. In its place is HD 207496.

Oza, et al. have recently suggested that Io-like exomoons could be generating observed circumplanetary Na I and K I gas seen in transmission spectroscopy at some extrasolar transiting hot Jupiters. Specifically, they find that WASP-49b is a good candidate for such a system. I'm not sure how they get around the problem that a migrating hot Jupiter would cross several resonances between the planetary and satellite orbital period which should remove the moon from the planet (see for example Spalding, et al. (2015)), and they also suggest that such moons could be catastrophically tidally evaporated this close to the planet. Overall, I'm rather skeptical, but since it's not like there aren't obvious false-positives among the TOI candidates, it isn't like physical plausibility is the driving force behind inclusion of a planetary body in this add-on. Since it has been proposed by a serious work of research, it'll be included for now.

The Extreme Solar Systems 4 conference happened recently. Several interesting finds were discussed. I've included them here, but left them as unconfirmed for now, pending the arrival of their papers being posted.

Celestia_Screenshot.png
WASP-49b's possible-ish exomoon candidate.


01 September 2019
- 82 TESS candidate planets added (TOI-1051 through TOI-1129).
- 19 TESS candidate planets have been determined to be false positives since the last update and removed.
- Added TOI-# ID's for known transiting planets detected by TESS (K2-237=TOI-1049, WASP-17=TOI-1050, etc).
- Added nearby compact system of terrestrial planets at GJ 1061, with one in the HZ (Dreizler, et al.).
- Added unconfirmed third planet at HD 200964, and updated parameters based on maximum likelihood 7:5 MMR model (Rosenthal, et al.).
- Added two unpublished planets at AU Mic (source), β Pic c (source), V1298 Tau c, d, e (source), HD 120066 b, KMT-2016-BLG-1836L b.
- Added TESS's first circumbinary planet, TIC 260128333 (AB)b, announced at the ExSS4 meeting (Source slides).
- Added exomoon candidate WASP-49b-I.
- Added several missing KOI candidates from Kepler DR25.

8098 planets (+1 asteroid).
4239 confirmed.
3859 unconfirmed.
Attachments
Celestia_Exoplanets_2019_09_01.rar
Catalogue of extrasolar planets up to date as of 01 Sep 2019.
(728.84 KiB) Downloaded 37 times

Avatar
Diigg M
Posts: 35
Joined: 09.02.2011
With us: 8 years 7 months

Post #33by Diigg » 10.09.2019, 15:33

Hi, how do you install it? I understand that the scripts must be copied to the scripts folder, but I have doubts with the other files.
Should we copy them to the data folder or the extras folder?

Avatar
Anthony_B_Russo10
Posts: 226
Joined: 03.07.2018
Age: 16
With us: 1 year 2 months
Location: Tallahassee, Florida, US

Post #34by Anthony_B_Russo10 » 10.09.2019, 21:14

Diigg wrote:Hi, how do you install it? I understand that the scripts must be copied to the scripts folder, but I have doubts with the other files.
Should we copy them to the data folder or the extras folder?
The ssc and stc files go into Celestia's extras folder, it's the xyz and xyzv trajectories that go into the data folder.
Anthony B. Russo, I like Pluto. Mod of the Celestia subreddit: https://www.reddit.com/r/Celestiasoftware/
My PCs. Dell Latitude D820 D400 D410 D420, Acer Aspire, Dell Optiplex 745 x2 755 GX-620 960, Sony Vaio PCG-F420, Packard Bell Legend 100, and an IBM clone


Return to “Add-on releases”

Who is online

Users browsing this forum: 25 guests