I was looking for more information about this star, which sparked my interest because of its colour & shape (brilliant blue, 50% wider at its equator, due to its high-speed rotation, and encircled by a Saturn-like ring of gas). When trying to find it's distance in light years from us, I found a wide range of answers:

44 ly, 65 ly, 69 ly, 70 ly, between 60 and 70 ly, 73 ly, 75 ly, 84.8 ly, 85 ly, 118 ly, 120 ly, 125 ly, 127 ly, 140 ly, 143.8 ly, 144 ly, 145 ly, and finally this site (perhaps wisely) just left the "distance in ly" column conspicuously blank for Achernar!!!

So, now I'm obviously bugged...How far away is Achernar? :blink:

Right, but more importantly how can this star be 50%(!) flatter at its poles than at it's equator?
I'm still waiting for the current theories to find a patch to stop this gap.
About the distance, obviously we don't know, can we bet?
80.0 lightyears!
cheers.

Yes, conventional physics can only account for a 20-30% bulge. But this star has a whopping great 50% bulge at its equator! It would look pretty weird in the sky, wouldn't it? - (assuming there was a sky in which to see it!).

But hell, I was under the impression astronomers had all the local stars pretty much pinned down distance-wise? I found this very odd!?

Fascinating & unusual star, tho'...!

The Hipparcos measurement of 144 to 145 light years is probably the most accurate one.

Dave Mitsky

__________________
Chance favors the prepared mind.
De gustibus non est disputandum.
Never attribute to malice that which can be adequately explained by stupidity.

Yes, that should have been the most accurate one, but recently those data were questioned in establishing the distance to the Pleiades. They seemed to be off by a factor 2. There is a thread here somewhere with the details.
Cheers.

This is the article
http://www.universetoday.com/am/publish/di...calculated.html

And somewhere in the story it says that either the Hipparcos measurements of the Pleiades were wrong, or the stellar evolution model is wrong. The decision is that Hipparcos was wrong (based on new data), although I can't find what was wrong with the Hipparcos data exactly.
Cheers.
Btw Arechnar is at least 56% flatter at it's poles!!

According to my Planetarium program on my Palm, it's 142 LY, a little less than the Hipparcos estimate.

__________________
"The stars are my home"
"I've seen things you people wouldn't believe... Attack ships on fire off the shoulder of Orion... I've watched c-beams glitter in the dark, near the Tanhauser Gate... all those moments will be lost, in time... like tears in the rain..."

142 light years? Well I'll slot that estimate in with the others. I heard parallax measurements can take us out to 100 parsecs (or thereabouts)...that's over 300 light years...Again, I'll ask: Why such divergence of estimates?

Another thing...

All the rest of our measurements of cosmic distance are based upon parallax measurements...ie we use parallax to find a Cepheid variable...then we find Cepheid variables further away, in other galaxies, etc...etc etc...on & on into the far distances of outer space...

So if we can get a nearby star wrong by a factor of 3-4, what does that say about our cosmology?

Excellent point... maybe the universe is a lot bigger (and older?) than we thought... I'm sure the HST managed to pin the age of the universe at around the 12 billion year mark, but I'm not sure of the methods used.

__________________
"The stars are my home"
"I've seen things you people wouldn't believe... Attack ships on fire off the shoulder of Orion... I've watched c-beams glitter in the dark, near the Tanhauser Gate... all those moments will be lost, in time... like tears in the rain..."

OK, I'm grappling blindly in the dark here...and I hope an astronomer comes along & puts me right:

I think HST uses spectrographic redshifts...doppler effects...to gauge distances (using Hubble equations etc)...but the theory behind cosmic redshifts are based (loosely?) on a whole series of stepping stones...starting from parallax measurements. The margin of error in Stepping Stone #1 is multiplied exponentially across each further step, amounting to a colossal margin of error. And if Stepping Stone #1 is out by a factor of 3-4, then that throws the whole cosmology out the door.

Damn it, I have to go back & re-read Kitty Ferguson's "Measuring The Universe"!!!

I have no idea why there should be as wide a variety of distances for Achenar (alpha Eridani) as indicated in Faulkner's first post. The Hipparcos parallax of 22.68 mas translates into 44 parsecs or 144 ly (anyone can search the Hipparcos catalog, by HD number, HIP number, or RA & DEC; the HIP number for Achenar is 7588 and the HD number is 10144). I would accept this as the most likely distance.

The explanation for the problem with the Hipparcos data & the Pleiades can be found here: The HR diagram and the Galactic distance scale after Hipparcos, I. Neill Reid, Annual Review of Astronomy and Astrophysics 37: 191-237, 1999, specifically on pp. 197-198. There is a systematic bias in the Pleiades data, there is a clear correlation between the parallax and the correlation coefficient of the parallax and right ascension; where the parallax and right ascension are more strongly correlated, the reported parallax is higher. If one trims the data set to eliminate the stars with a strongly biased correlation between parallax and right ascension, the new Hipparcos Pleiades parallax is 7.49 +/- 0.5 mas, which translates into 133.5 parsecs, which is right in the distance region presented as most likely by Kulkarni's group (A distance of 133-137 parsecs to the Pleiades star cluster, X.P. Pan. M.Shao & S. Kulkarni, Nature 427(6972): 326-328, January 22, 2004). This is one reason why I am distrusting of reports in any news media. It was already known 5 years ago that the Hipparcos data for the Pleiades were biased, so it is hardly a big deal that these new results agree with that. Furthermore, the corrected Hipparcos distance for the Pleiades, published 5 years ago, matches the new "correct" distance determined by the Caltech group. There isn't any news here, despite appearances (and also note that no such problem exists for the Hyades, where the Hipparcos parallax fits in nicely with the main sequence fitting distance based on stellar models).

The first statement is misleading, and the second is just wrong. The reality is that Achernar is a Be class star (a http://www.glyphweb.com/esky/concepts/spectralclassification.html]spectral B class star[/url] with photospheric emission lines), which is one of the more exotic & peculiar types of star (see Classical Be Stars, J.M. Porter & T. Rivinius, Publications of the Astronomical Society of the Pacific 115: 1153-1170, October, 2003 for a detailed examination of the peculiarities). The models for Be stars, such as they are, amount to no more than 1st approximations, primarily because the physical details are too complicated to deal with. As de Souza, et al., point out in their paper from September (The spinning-top Be star Achernar from VLTI-VINCI, de Souza, et al., Astronomy and Astrophysics 407: L47-L50, September 2003), their optical model is based on a uniformly bright disk (which we already know is not really correct), and the internal physics models assume uniform rotation, which is probably also not physically real. They also do not include convection or internal turbulance. The models are incomplete approximations, so it is hardly surprising that they don't properly anticipate the observed shape of the star. This also means that no one can say that "conventional physics" does not account for the full oblateness of the star, since in reality, no one has really tried to use conventional physics on the problem yet. When they try, and fail, then we can make such a claim.

There is a lot of physics to worry about here. For instance, if the internal rotation of the star is not uniform, if it is faster towards the center of the star, then we know that the star will bulge more at the equator than would a uniformly rotating star (which would lead the models in the direction of this observation). There is also the open question of the affect of magnetic fields on the stellar bulge (stronger than "normal" magnetic fields may well be "normal" for Be stars; I think only one Be star magnetic field has been measured, and it is unusually high). And what about the affect on observations of a circumstellar gas shell? De Souza, et al., believe that their observations and implied flattening are not contaminated by a gas shell, since they don't see excess H-alpha emission, but there might still be a systematic bias in the flattening observation. And, there might also be a physical consequence of mass loss, in the equatorial bulge. In short, there is a lot of work to do before one can make definitive statements, one way or the other.

The real interest here is that the observation of the oblateness of Achernar is a first of its kind observation. It's not always possible to derive a complete physical picture of something from "first principles". We can't simply derive a realistic stellar model from scratch, because there are too many free choices to make. Rather, we need to calibrate the models against real observations. That constrains the choices, and allows the modelers to more quickly zero in on what works and what doesn't. Until now, no one has known enough about real Be stars to make detailed physical models. Now we know a lot about at least one Be star, and that's one more than we had. Now we can get down to the business of deciding which details are important, and which aren't, in modeling this kind of star. And, the more observations like this we get, the better off the modelers will be. Give them a few years, and we'll have much more realistic models of a Be star interior, and a much better understanding of the detailed physics. That's the real fun part.

__________________
Don't try this at home - We're what you call "professionals" - MythBusters.

There are some cases where we've been able to directly measure the distance to objects a lot farther than 100 parsecs away. One great example is SN1987A - the supernova that exploded in the Greater Magellanic cloud in 1987.

A little less than a year later (IIRC) a cloud of gas and dust that had been expelled from the star by an earlier eruption "lit up" with the light from the supernova. So, we can measure the angular diameter of this cloud in the sky, and we know it's real size (a little less than a light year in radius) because of the time it took for the light from the supernova to get to it. Simple high school geometry can tell you how far away it is. Turns out to be 169,000 light years (+/- about 3%) which agrees quite well with earlier estimates based on those Cepheid variables.

A nice confirmation of the standard method, no?

For more details just google "SN1987A light echo" and maybe "distance". I've seen several articles describing this.

Or we could say that we don't know very much about stars in general. When you say that there are too many free choices, isn't it almost the same as "we don't know yet". And that's based on the observation of one star, what about all the others where this information is unknown; why do we call those stars peculiar and exotic?

Now that is very true, we want more observations!
Cheers.

But that would be very wrong. We know a great deal about stars, in general, and in many less "general" cases as well. For instance, we know about our own sun in excruciating detail. The internal physics and structure of the sun is not "guessed", it is "known", to the fullest extent that the word "know" can be applied in the empirical sciences. We also know the internal physics of all main sequence stars very well, even though we may not know the detailed structure as well, because most stars are simply too far away to duplicate the detailed observations that helioseismology allows. our knowledge of physics in general, strongly contrains what can and cannot happen inside a star.

However, a stellar interior is nevertheless a complex place. The key to understanding complex systems is that even if you know what all of the individual physical processes are in the system, you may not be able to predict a-priori, how those processes will interact to produce the complex system. So, while we know a lot about stars in general, and a lot about specific classes of stars & individual stars, there are still some classes of stars that are less well understood, such as pulsating stars, rapidly rotating stars, or stars where convection is dominant (so the internal physics of a sun like star is better understood than is the internal physics of an M-dwarf like Van Biesbroeck's Star). And in all cases, internal physics is a lot easier to understand than the stellar atmosphere, which is invariably far more complex. So, not surprisingly, the most difficult stars to get a handle on are the ones with complex interiors and complex atmospheres.

The Be class stars fall in this category. We know most, but not all, of the physical processes at work in the interiors & atmospheres of such stars. The problem is that there are a lot of them, and they all seem important. We could easily create a stellar model from "scratch", and more easily about a million such models. But we would have no way to tell which are the ones we really see, and which not, and in many cases, an arbitrarily created stellar model can be found to be impossible, due to unanticipated interactions between processes (a common problem in complex systems). The solution is to have a real star to compare with. The mode detailed our knowledge of the real star, the more strongly we can constrain the physical processes that are really important in the model. Eventually, the result is a model which predicts the observed properties of the star, based on input physics. This has already been done for the sun with great fidelity (see, for instance, John Bahcall's webpage, and follow the links through "technical articles" and "standard solar models and helioseismology").

I will repeat an earlier point. While it is always important to keep a clear eye on what scientists don't know, it's equally important to keep a clear eye on what we do know. And we do, in fact, know a great deal about stars in general. One result of this knowledge is our ability to use standard physics to replicate stellar evolution with time, which in turn replicates even small details in the Hertzsprung-Russell diagram. That's the kind of behavior one expects from successful theories. Another is our ability to create standard models of the sun that are so correct, they forced particle physicists into the realization that neutrinos really have a non-zero rest mass, a thoroughly unexpected, but very valuable result (as documented in Bahcall's papers reference above).

There is a huge literature on the physics of stars. For reading from current sources, I would recommend Stellar Interiors: Physical Principles, Structure, and Evolution by Hansen & Kawaler, Springer-Verlag, 1995; An Introduction to the Theory of Stellar Structure and Evolution by Dina Prialnik, Cambridge University Press, 2000, and Advanced Stellar Astrophysics by William Rose, Cambridge University Press, 1998. This is a topic you can't learn on the web, only the old fashioned way (books) works.

__________________
Don't try this at home - We're what you call "professionals" - MythBusters.

Ok, would you explain to me then the mechanism of the "11 year cycle" of the Sun? (and I don't mean the things that we can see, but the cause).
My idea is that there is a lot unknown, like the behaviour of stars like V838 Mons or FG Sagittae, that do not follow the stellar evolution model. If you say that we know all the physical processes, but only need to fill in the details, I think you're forgetting that we also need to know about the exotics, erratics and anomalies, and I think there is a lot to learn from all these objects and we need more observational data. A lot has been learned, you're right about that, but in my opnion we're not even close to knowing all there is to know about stars.
Cheers.

Ha ha, OK who's got the guts to come up with an explanation for these 2 crazy stars???

I looked them up, they are indeed baffling...exploding & contracting like bloody bubble-gum...

And the Sun's 11-year cycle? Is that "normal"? Is our Sun unstable?

Well, I never suggested that anybody (including the ubiquitous "we") knew "all there is to know about stars". In fact, I was quite explicit about saying the contrary, so I fail to see the point of constantly trying to imply that I am saying things I don't say. But, to answer Faulkner's questions, I have the "guts".

V838 Mon
On this object, I refer you to the webpage I wrote last year, and updated as a result of your inquiry: V838 Monocerotis. It is not known what this object is, but there are two plausible models in circulation. One is a merger of main sequence stars. The other is a model of an expanding giant absorbing massive planets. Either may be correct, both may be wrong. But they are consistent with observation. So far as I know, V838 Mon remains a unique object, the sole known member of its class of objects, whatever that turns out to be.

FG Sagittae
But FG Sagittae is not so mysterious. It is almost certainly a post helium flash ("born again") AGB star, the same class of object as Sakurai's object and V605 Aquilae. See the webpage The Remarkable Object FG Sagittae, and the paper Sakurai's object, V605 Aquilae, and FG Sagittae: An evolutionary sequence revealed, T.M. Lawlor & J. MacDonald, Astrophysical Journal 583(2): 913-922, February 1, 2003. Lawlor & MacDonald have also suggested that V838 Mon might be another example of this same class of object in a AAS meeting abstract. Also see Lawlor's PhD Thesis.

Sunspot cycle
The cause of the sunspot cycle is not now known, but I think this will not long be the case. We already know that the solar dynamo is generated in the convective zone, the outer 15% or so of the solar radius. The process is chaotic, and this is evident in the fact that the solar cycle is not fixed, only nearly so. Dynamo theory is quite complex, and it will take time to pin down which ingredients dominate the solar cycle. But we do know that, as cycles go, the sun is typical. The spot cylces of other stars have been measure for decades at Mt. Wilson Observatory, in the ongoing HK Project (named after the H & K lines in the singly ionized calcium spectrum). See the activity cycles page for a comparison between the solar cycle and those of other stars (only a representative sample of the 2200 stars studied). Published studies show that the length of the activity cycle is correlated with the stellar mass, which is also correlated with the depth of the convective zone (it is deeper for less massive stars), and that is consistent with the standard theory that the dynamo is seated in the convective zone (An Interpretation of Cycle Periods of Stellar Chromospheric Activity, Soon, Baliunas & Zhang, Astrophysical Journal 414(1): L33-L36, September 1 1993; A dynamo interpretation of stellar activity cycles, S.L. Baliunas, et al., Astrophysical Journal 460(2): 848-854, April 1 1996; Patterns of variation among sun-like stars, R.R. Radick, et al., Astrophysical Journal Supplement Series 118(1): 239-258, September 1998). Since Sallie Baliunas was the PI for the HK project, you can find several more papers along the same lines by simply searching the NASA ADS on the author name "Baliunas, S.L."

Of course there is a lot that is unknown, a point long since conceded. My point is that there is a lot that is known, and one should examine what is known before deciding what is not known.

__________________
Don't try this at home - We're what you call "professionals" - MythBusters.

Point also conceded, so we're basically saying the same thing. We need more observational data and see what the correct models turn out to be, my feeling is that we will encounter surprises, and yours is that there is only room for details and no paradigm shifts are to be expected, ok?

Again thanks for all the links to the relevant data, and I'll need a lot of time to read them all, but that's exactly what I'm asking for.
Cheers.