Talk:List of most luminous stars

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
WikiProject Astronomy / Astronomical objects  (Rated List-class, Mid-importance)
WikiProject iconThis article is within the scope of WikiProject Astronomy, which collaborates on articles related to Astronomy on Wikipedia.
 List  This article has been rated as List-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
Taskforce icon
This article is supported by WikiProject Astronomical objects, which collaborates on articles related to astronomical objects.

Need references !![edit]

It seems to me that this table really must require a reference of some sort for bolometric luminosity, before a star is accepted for display. I'm not going to wage war on the data we have (yet), but a reasonable step in that direction would be to try to get the most luminous candidates documented within a short time, say a week or a month, or else deleted. Thanks -- Wwheaton (talk) 20:53, 21 October 2008 (UTC)

Sources and Eta Carinae[edit]

I see that we have had a couple of pairs of edit reverts recently over the status of Eta Carinae. Neither its (currently listed) low luminosity (5,500,000 solar L; M = -12.1) nor its reverted higher L (55,000,000×solar) have any supporting references at all. I hope every one of the brighter entries in that table will soon be supported by a reliable source, in order for anyone to be able to take the table as a whole very seriously. It is clear that the 1843 magnitude could only be (roughly) in the V band, so we really do not know the bolometric absolute magnitude at that time very accurately. Also, it is now believed to be a binary, which confuses things a little, though probably does not the 1843 luminosity very much. Anyhow, please supply some supporting reference when adding material to this table, otherwise it just becomes silly. I also exhort editors to discuss complex situations here on the talk page rather than just reverting without explanation in the edit summary or here. Thanks, Wwheaton (talk) 03:39, 28 December 2008 (UTC)

Use Bolometric![edit]

In demoting Zeta Puppis (cited in link), I noticed that this article is using visible and bolometric magnitudes randomly. Deneb down under 50,000 solar? Please ONLY use bolometric magnitude estimates! Wayne Hardman (talk) 23:42, 7 February 2009 (UTC)

I am all for this in principle, but there is the severe practical problem that one must either assume a spectral shape (eg, a black body with a specified temperature) or else do a search of the literature for each object, compiling observations at various wavelengths, and combining them, still with some assumptions about the spectrum in unmeasured regions. There seems no perfect way to deal with this, but it would be nice to have a concept that will move us onwards in a consistent and logical way as observations improve. Right now it is chaos bordering on meaningless. Of course a distance must also be measured or estimated, as well as the extinction as a function of wavelength. Most of these very luminous objects deserve an article in their own right, and that would be the place to present a careful (referenced, of course) estimate of the bolometric luminosity. Once that is done, the resulting information can be entered here. Wwheaton (talk) 22:15, 8 February 2009 (UTC)


I think all stars with positive bolometric luminosity magnitude values should be removed, and the list should be limited to a maximum of 100 stars unless a good cut-off point exists to slightly extend or slightly contract the number listed. 100 is an arbitrary limit, but a commonly used one, so should be acceptable. As for allowing only negative values... if it's positive, it isn't very luminous, now is it? (talk) 08:13, 9 February 2009 (UTC)

I tend to agree that the list should be limited (and trimmed), both because of the significant labor involved in doing a good job, and simply because it gets gigantic at some point. My only alternate suggestion would be perhaps to limit it, not to a particular absolute magnitude threshold, but to the 100 most luminous stars known in the Milky Way galaxy, or something like that. In any case small errors in photometry and distance estimates will make the threshold somewhat ambiguous and arbitrary. I favor limiting it to our galaxy mostly because of the difficulty in excluding close double stars, which has already been a problem for Eta Carina. Another question is what to do about variability. I would tend to favor excluding transient outbursts (novas & supernovas, surely) and maybe try to get the average bolometric luminosity for others. The need to get data at different wavelengths all at the same time complicates to problem for variables (possibly fatally for non-periodic variables?), but the bolometric luminosity may be more stable for periodic variables, I'm not sure. I suppose we would not want to exclude the 19th century outburst of Eta Carina, but I have no clue about a fair estimate of its bolometric luminosity, except to extrapolate from V, assuming a black body. Wwheaton (talk) 17:32, 9 February 2009 (UTC)
Open-ended lists like this tend to end up on the AfD chopping block, so a cut-off criteria is a good idea.—RJH (talk) 22:07, 11 February 2009 (UTC)

LBV 1806-20[edit]

The information about very first star listed on this table looks dubious to me, based upon:

Figer, Donald F.; Najarro, Francisco; Kudritzki, Rolf P. (2004). "The Double-lined Spectrum of LBV 1806-20". The Astrophysical Journal. 610 (2): L109–L112. doi:10.1086/423306. Unknown parameter |month= ignored (help)CS1 maint: Multiple names: authors list (link)

I'm not sure I completely trust the data in the linked article. It says that LBV 1806-20 "shines up to 40 million times brighter than the Sun", but 40 million is 107.6; the paper lists luminosities of 106 or 106.3. This entry's luminosity appears in excess by an order of magnitude at least.—RJH (talk) 22:03, 11 February 2009 (UTC)

A proposal for selection criteria[edit]

I'd like to propose a criteria for list membership. In order for a star to be included, it must satisfy at least one of the following minima:

What do you think? (Granted the first two are semi-redundant, but it may be that only one or the other is available.) This will, of course, significantly truncate the current content.—RJH (talk) 22:09, 18 March 2009 (UTC)

I think better documentation and referencing at the top of the list is more important than a sharp cut at the bottom. Measurement uncertainties are going to make cuts on mass especially difficult, but also on luminosity. Bolometric luminosity is hard to measure accurately—people will publish whole articles on the bolometric luminosity of one or a few stars. I think just limiting the table at the bottom to a manageable size (100?) may be better than an arbitrary cut on mass or luminosity. I'd like to see all the stars at the top of the list have a solid article about the object with at least a couple of good references from the literature (not just popular news notes or press releases) documenting the luminosity. Magnitudes ought to specify color band or filter, and distance is essential as well. Because of the difficulty in resolving doubles and tight clusters, I think we should consider limiting it to the Milky Way Galaxy, or maybe imposing a distance cut. I'd rather see a list with 30 solid objects at the top than what we have now, frankly. Wwheaton (talk) 18:22, 24 March 2009 (UTC)
Thanks for the reply. While a top 100 list would work, the part I don't like is the culling of people's research at the bottom as new members are added. Plus if you find out some members don't belong, then you may have already truncated the prior bottom members of the list. You might need a spill-over list to keep the bottom members around.
Still, I'd be happy with any sort of limit at this point. Defining an inclusion limit is needed make the documentation both more essential (to prove the star belongs) and easier to accomplish (since the list will be shorter). My preference would be to basically start over with a fresh list at the top of the page of just those stars that can be cited. Maybe start with a top ten list and then I can try to dig up citations for those. Afterward we can work on the other stars.
A related idea I had was to have multiple lists: the most luminous stars within a particular distance (10 parsecs?); the most luminous stars within the Milky Way, and maybe a third table for more distant objects. (The list has a bunch of objects in the Local Group, so I wanted to capture those.)—RJH (talk) 18:46, 24 March 2009 (UTC)
I don't disagree. I just think a hard criterion for cutting at the bottom is not so important, more a matter of practicality, dealing with numbers. But I'm tempted to remove all the ones at the top that do not have a credible reference, and the very top ones should all have articles thoroughly documenting their claimed luminosity, since it is so complex to determine -- distance, spectral energy distribution, and extinction all enter in, in an essential way. The thing is, there are inevitably going to be a lot of entries crowded together in luminosity near the bottom, the more so the lower you place the cut. So it is kind of arbitrary, and a 2% change in the best estimate of the distance or extinction could move stars across the cut. We just don't know things that accurately, except for a tiny fraction of nearby, well-studied objects. Cheers Wwheaton (talk) 20:44, 9 April 2009 (UTC)
I agree on all of your points. Perhaps it would be a good idea to copy the existing list to a linked file off the talk page so we have it for reference?—RJH (talk) 16:02, 10 April 2009 (UTC)
OK, certainly. I think this page is not archived at the moment, but if it ever should be, do you know any way to keep such a file from being lost in the great Wiki anthill? I suppose we could put it in either of our user pages. Wwheaton (talk) 19:01, 11 April 2009 (UTC)
I think we just add a {{Archive}} template at the top of the page, then add it to our watch lists. Hopefully after a while it won't matter, because the new page will be more reliable.—RJH (talk) 23:51, 12 April 2009 (UTC)

After reading this discussion, I decided to take the bull by the horns and impose a cutoff of a bolometric luminosity of at least 105 solar luminosities (which, using Mbol(Sun)=4.75, is equivalent to an absolute bolometric magnitude of −7.75 or lower.) Removed entries are still accessible in the history. The idea of a mass cutoff is not a good one as the list is of most luminous and not most massive stars. Of course, I agree with Will Wheaton's point that we must have references certifying the luminosity of any list member. I left a few familiar stars in with Lbol below the threshold for comparison. Spacepotato (talk) 03:09, 21 November 2009 (UTC)

Please note Wwheaton is not Will Wheaton. Sorry for the confusion, but I had that name first, afik. Thanks -- Wwheaton (talk) 23:12, 13 September 2012 (UTC)
Might it be of interest to have a separate article giving a list of massive eclipsing binaries? That way we could list more concise data about the larger stars.—RJH (talk) 17:07, 31 March 2010 (UTC)
Looks like we need to address the issue of selection criteria again. I think the table is now moderately complete down to about a million times solar luminosity, but there are literally dozens of stars that could be added down to maybe half a million. Most of them are meaningless names and numbers in clusters far far away, but the table says "most luminous" so it seems a bit naughty to deliberately list things like Betelgeuse at about #50 when it probably doesn't even make the top thousand. One possibility is to move out objects not in our own galaxies to separate tables, although we're still left with an ever-increasing list of meaningless numbered objects swamping things people have heard of. I just suggest it because the lists of most luminous stars for the Magellanic clouds is effectively complete down to perhaps half a million times solar luminosity and certainly down to around a million. Lithopsian (talk) 22:15, 23 May 2012 (UTC)

Cygnus OB2-12[edit]

The first star in this list is given an Mbol of −12.2. The only source I could find for this value was the Tim Thompson web site:

(I suggest perhaps trying to check with stellar expert and author James Kaler's on his site & via the following books & articles : Kaler, James B., ‘Extreme Stars’, Cambridge University Press, 2001. Kaler, James B., ’The Hundred Greatest Stars’, Copernicus books, 2002. Kaler, James, "Hypergiants", Astronomy March 1994, Kalmbach publishing co. Kaler, James B., Astronomy, May 1991, "The Brightest Stars in the Galaxy.", Kalmbach publishing Co. - ed = StevoR. PS. I agree 40 million x L solar sounds massively excessive for LBV 1806-20. Around 4-6 mil. would seem a more reasonable guesstimate to me. Maybe an "extra zero" error & really 4 million?)

Thompson, Tim. "Cygnus OB2 #12 and the Cygnus OB2 Association". Tim Thompson's Home Page. Retrieved 2009-10-25.

His web site lists him as a physicist at the JPL. However, all the journal sources I checked based their values on:

Massey, Philip; Thompson, A. B. "Massive Stars in Cyg OB2". Astronomical Journal. 101 (4): 1408−1428. doi:10.1086/115774.CS1 maint: Multiple names: authors list (link)

who gave an approximate value for Mbol of −11 after correction for reddening. It might help if we had a corroborative journal article confirming the −12.2 value, but I haven't been able to turn one up. I checked all of the references Tim Thompson listed, but they showed only −11 (where the data existed). Any ideas? This star doesn't even have a properly defined spectral class.—RJH (talk) 19:41, 25 October 2009 (UTC)

I agree the lower luminosity limit is more credible, and have re-located the star downwards. The luminosities for -11 & -12.2 come out to be 2,000,000 and 6,000,000 solar luminosities, respectively. The geometric mean of these is 3,500,000. I have moved the star's position down accordingly, but there is obviously a wide range where it could end up. I also moved a fact tag from the luminosity numbers, since the conversion from absolute magnitude (which is referenced) should be routine. But this also may want further discussion though. Wwheaton (talk) 20:04, 24 July 2010 (UTC)

Inconsistency in brightest star?[edit]

The brightest star in the table is listed as 6,000,000 solar units, but the first sentence following the table reads: "Note that even the most luminous star (40 million times the luminosity of the Sun) is much less luminous than extragalactic objects like quasars" I'm wondering if the values in the table have been updated, but not that piece of text. I don't know enough on the subject to know which is correct, just that someone should probably reconcile the two values. —Preceding unsigned comment added by (talk) 00:06, 10 December 2009 (UTC)

I have changed the text sentence (to six million) to maintain internal consistency. The changes to the table were made Oct 25, 2009 by an IP editor without edit summary or other comment, and I make this change without prejudice re. the correctness of that edit, but based on the discussion above it seems defensible. If anyone restores LBV 1806-20's previous exalted 4e7 luminosity, please change that sentence to maintain consistency. (Actually, it would probably be better just to say "the most luminous quasars are much more luminous than the brightest known stars", or words to that effect, so we do not fall into this pit over and over every time the top value changes. I think I will do that myself, in fact.) Wwheaton (talk) 01:07, 10 December 2009 (UTC)

Zeta Puppis[edit]

What is the bolometric luminosity of Zeta Puppis ? 360 000 solar units as cited in this article or 790 000 solor units as cited in the article Zeta Puppis? Nedim Ardoğa (talk) 21:16, 9 February 2010 (UTC)

R136a1 ?[edit]

This object was recently added, but I think it may not be quite well enough established to merit inclusion here yet. I have two problems, the main one being that as far as I could see from the article, it is based on a conference talk, with as yet no refereed paper in the literature. My second objection is just that it is in another (albeit nearby) galaxy, and allowing distant candidates without limit obviously opens the door to "stars" that appear to be single observationally, but which turn out to be multiple later. This actually happened previously, in the 1970s I think, for the whole R136 cluster, which was first claimed to be a single super-luminous star. Then it was resolved into a cluster (by HST, I presume) with several tight clumps. So it is a sore subject.

The VLT folks have been doing a fantastic job with adaptive optics and interferometry lately, and I think it is quite likely that this report is valid and will be confirmed soon. Anyway, I would like to wait at least until we have a reference to a peer-reviewed publication to look at. (Less good would be a "press release" web source from ESO or the VLT, which should give some more details that might be convincing, like spectroscopic evidence that it is not a tight binary.) Even better would be independent confirming observations or third-party commentary, which I would expect soon.

This list article is inevitably plagued with poorly characterized and referenced candidates. If we do not exercise tight and conservative control over it, it will very soon become simply meaningless. There are other objects listed needing better (or any...) sourcing, and we need to fix or remove those too, but I believe that is a poor reason to fail to deal with this one, especially as it would be the celebrity, at the top of the list.

I would also like to propose a maximum distance rule for this article, to prevent our listing extragalactic false candidates again and again. Because observations are rapidly improving (with no end in sight) and because even very rare and distant examples are of great fundamental interest, I think any distance limit would need to be revised upwards now and then as the instruments improve; the day will come when we will have optical interferometric arrays in space that will change all the rules we now live by. For now I think the LMC & SMC (& nearer dwarf satellites of the Milky Way) are probably OK, and soon the entire Local Group likely will be too. Wwheaton (talk) 19:55, 21 July 2010 (UTC)

I have just noticed that R136a1 was a duplicate, already listed further down the list, also without an explicit reference. I am leaving it in for now, but yet another warning on the ease with which problems arise here. Someone should systematically check SIMBAD for all these brighter candidates and see if we can get good solid references for each. Wwheaton (talk) 20:20, 21 July 2010 (UTC)

There is now an ESO press release giving a reference to a paper by Crowther et al in Monthly Notices of the RAS; preprint available here. I do not have time to read it at the moment, but assuming it looks convincing, I have no objection to restoring R136a1 to the head of the table. If anyone else does this, please also add the reference, and remove the duplicate entry for it half a dozen lines down in the table. Thanks. Wwheaton (talk) 23:57, 21 July 2010 (UTC)

Table format[edit]

Since this page is titled "Most Luminous Stars", and since the stars are arranged in order of bolometric luminosity, I would think that "Bolometric Luminosity" should be the first column (not the last). I would think "Apparent Magnitude" should be the last column, since that figure is really incidental to a star's absolute magnitude. I would be glad to make this edit, but I wanted to see if there's agreement among those of you who are more invested in this page. Niobrara (talk) 20:10, 5 December 2011 (UTC)

I went ahead and did it.Niobrara (talk) 21:30, 27 December 2011 (UTC)
Sure would be great if we could make this table sortable by column... anyone know how to do that?Niobrara (talk) 21:46, 27 December 2011 (UTC)

Distance to DL Crucis[edit]

Currently the distance given is 20,000 l.y. However, Simbad seems to estimate a distance of 1125 parsecs, which is considerably less.Niobrara (talk) 21:26, 10 April 2012 (UTC)

The Simbad data is meaningless. The margin of error is much larger than the value itself. The parameters for this star are poorly referenced though, both here and on the detail page. Plus they don't match. — Preceding unsigned comment added by (talk) 16:52, 8 September 2012 (UTC)
Which suggests to me that we should not touch it. We have to have a reasonably reliable distance to be able to say anything useful about luminosity, of course. Parallax is the gold standard, but we would lose all our luminous blue variables if we demanded that. Having a pretty solid lower limit of 1,000,000 solar luminosities must be good enough, I guess. (This means that our table cannot be complete beyond about 1 kpc, right?) Wwheaton (talk) 16:34, 14 September 2012 (UTC)

Time to cull again[edit]

I intend to move the cutoff point for the table to 1,000,000 times solar luminosity. The list is moderately complete above that level, woefully patchy below it. Any star below that level which isn't a household name will be deleted and the rest highlighted in pink. Niobara will then delete half the ones I thought were household names ;) Speak now or forever hold your peace. Lithopsian (talk) 11:14, 13 September 2012 (UTC)

Given the ? 100,000,000,000 ? or so stars in The Galaxy, we have to cut pretty severely, because otherwise the human labor involved in reducing apparent brightness to bolometric luminosity in a reliable way involves an impossible amount of scholarship for an encyclopedia.
SIMBAD helps a lot, so things should get better. Yet the observations are improving even faster, so it seems hopeless to me to try to go any deeper than you suggest. I think a table with 100,000 entries is not really appropriate here. Besides, professional astronomers construct huge tables of all kinds, and publish them nowadays on the web. They are surely better equipped than we to manage this sort of thing.
I must say I am delighted to see how much this table has improved since I was so appalled by its condition a few years back. Thanks to all who have labored over it since. Wwheaton (talk) 16:09, 14 September 2012 (UTC)

Bold truncation of obscure stars[edit]

A block at the bottom of the table is headed with the statement "The following well-known stars are listed for the purpose of comparison", yet all the stars listed between "Pismis 24-17" and "θ1 Ori C" seem relatively (or utterly) obscure, while all the stars listed after Betelgeuse are indeed well known.

I cut the first range (preserved and displayed below) and I invite anyone so inclined to restore individual deleted entries that can honestly be called "well-known". Please don't just revert the whole edit! Harold f (talk) 21:19, 6 June 2013 (UTC)

Pismis 24-17 891,000 −10.1 5,600 11.84
Cygnus OB2-#8C 870,000 −10.1 5000 13.25
Cygnus OB2-#11 790,000 −10.0 5000
Pismis 24-1SW 646,000 −9.8 5,600 10.43 (Pismis 24-1 total for three components)
HD 33579 (in LMC) 650,000 −9.7 160,000 9.13
P Cygni 610,000 −9.7 5,900 4.8
ρ Cas 550,000 −9.6 12,000 4.1 to 6.2
VY CMa 450,000[1] −9.4 4900 6.5 to 9.6
ε Ori 380,000 −9.2 1300 1.70
KW Sgr 370,000 −9.17 10,000 8.9
ζ Pup 360,000 −9.0 1090 2.21
V354 Cep 360,000 −9.15 9000 10.82 to 11.35
RW Cep 350,000 −9.11 11,500 6.52
μ Cep (the Garnet Star) 340,000 −9.08 1900 4.04
VV Cep A 315,000 −9.0 2400 4.91
WOH G64 (in LMC) 280,000   163,000
KY Cyg 270,000 −8.84 5000  
Plaskett's Star A 224,000 −8.6 6600 6.06 (A + B)
θ1 Ori C 220,000 −8.6 1500 5.13
I would suggest keeping:
  • P Cygni (well known variable star and class prototype);
  • ρ Cas (naked eye yellow hypergiant);
  • VY CMa (contender for largest known star);
  • ε Ori (naked eye);
  • ζ Pup ? (hottest naked eye star?);
  • μ Cep (the Garnet Star);
  • Plaskett's Star A? (once the most massive known star);
  • θ1 Ori C? (Trapezium member). Lithopsian (talk) 22:33, 6 June 2013 (UTC)
Like Lithopsian, I agree with some, but not all, of the truncations. Part of the issue is the phrase "well-known", which is subjective depending upon how "well" one "knows" the field of astronomy. So I changed it to the slightly-more-objective phrase "naked-eye", and added back in the eight stars mentioned above, plus two others. I have no objection if someone wants to re-delete those two (RW Cep and VV Cep A); I kept them simply because deleting is easier than adding back in. Niobrara (talk) 15:39, 14 June 2013 (UTC)
  1. ^ Humphreys, Roberta M. (2006). "VY Canis Majoris: The Astrophysical Basis of Its Luminosity". EPrint: 10433. arXiv:astro-ph/0610433.

Individual articles for some of these stars[edit]

I'd like to make individual articles for some of these stars, but I end on this list. How can I avoid that?. --U-95 (talk) 15:17, 27 August 2013 (UTC)

I argued against this sort of pointless self-referential linking, but no matter. Click onto the page of the star you want. You'll come back to this page, but at the top you'll see a message that you were redirected. Click on the name of you star in that message and you'll go to the actual redirect page. It doesn't look much like a Wikipedia article but it is and you can replace the redirect directive with an actual article. Lithopsian (talk) 18:32, 27 August 2013 (UTC)

Apparent/absolute magnitude[edit]

The apparent magnitude of a star is how bright it appears from Earth. The absolute magnitude is how bright the star actually is. Many bright star look quite dim from Earth since they are much further away — Preceding unsigned comment added by $spunkynova1 (talkcontribs) 20:50, 13 January 2014 (UTC)

And? Lithopsian (talk) 22:00, 13 January 2014 (UTC)
Luminosity, by definition, is absolute, not apparent.   ~ Tom.Reding (talk|contribs|dgaf)


The first star in 'naked eye' section is P Cygni, with absolute magnitude of -9,7, distance of 5900 light years and apparent magnitude of 4,8. A few lines above, there are the Cygnus OB2 stars, with absolute magnitude of -10,4, distance of 5000 light years and apparent magnitude of 12,7.
So, I wonder, how can Cygnus OB2 be 1.8 times more luminous than P Cygni and have apparent magnitude of only 12.7, even when it is closer to us? (talk) 00:23, 17 August 2014 (UTC)

Dust ;) Lithopsian (talk) 10:15, 17 August 2014 (UTC)

Error with WR 25[edit]

The list says WR 25 was more than 100000 lightyears away, with apparent magnitude of about 18. However, the article says it's only 7500 lys away, at mag 8,8. The contradiction (and thus, error in the list) is obvious. — Preceding unsigned comment added by (talk) 23:51, 7 November 2014 (UTC)

Expanding table with effective temperature[edit]

As of today, the "Effective temperature" column only goes down to Var A-1. Should I expand it all the way down to S Doradus (or even to the comparison chart)? If anybody wants it, here's a list I made.

Sources are from the articles for individual stars; if they don't have an article, the source is listed below.

Loooke (talk) 02:26, 28 January 2017 (UTC)

Also, here are the temperature values for the comparison stars. Again, these all come from the individual star's articles:

Loooke (talk) 20:41, 25 March 2017 (UTC)