Smallest "earth-like" planet found

[granthutchison]

I've added the new object to extrasolar.ssc, designating it "d" to match the usage in the Extrasolar Planets Catalog. You can get the updated file from the CVS tree at:
http://cvs.sourceforge.net/viewcvs.py/celestia/celestia/data/extrasolar.ssc?rev=1.39&view=log

(As with the other extrasolars, I don't use an Epoch statement - instead I calculate the MeanAnomaly from the epoch of periastron given in the Extrasolar Planets Catalog.)

Grant

[Cham]

Thanks a lot, granthutchison.

[Evil Dr Ganymede]

A few things still bug me about this system...

1) the planet appears to have an orbital period (9.5 days) that is shorter than the star's rotation period (31 days). That means that if it's solid it should have spiralled into the star because of the tides by now, shouldn't it? (I'm not convinced it would if it was a jovian, but one that small should still have a significant-sized solid core that would be more affected by them). Either that, or the rotation period of the star (or the orbital period of the planet) is wrong...

2) What's the metallicity of the star in terms of a z=... number (as in, x is the percentage of the star that is hydrogen, y is helium, and z are metals)? They only give it in "dex" (I think that's Fe/H, isn't it?) in the paper, and I don't know how to convert that to z. The solstation page claims that it's is a high metallicity star which means it should be older than the sun, but the paper claims it's only 2 Ga in age.

[granthutchison]

The solstation page claims that it's is a high metallicity star which means it should be older than the sun, but the paper claims it's only 2 Ga in age.

High metallicity implies a younger star, other things being equal ... the interstellar medium had more time to enrich with metals before a younger star formed, so 2 Gyr seems consistent. If it's a young system, maybe the Jovian just hasn't had time to spiral all the way in yet.

Grant

[Evil Dr Ganymede]

The solstation page claims that it's is a high metallicity star which means it should be older than the sun, but the paper claims it's only 2 Ga in age.

High metallicity implies a younger star, other things being equal ... the interstellar medium had more time to enrich with metals before a younger star formed, so 2 Gyr seems consistent. If it's a young system, maybe the Jovian just hasn't had time to spiral all the way in yet.

Grant

As I understand it - this not necessarily true. High metallicity does correspond to shorter lifespan, yes. But while a 1 solar mass star with z=0.1 (as opposed to Sol's z=0.020) has a total lifespan of only about 4.8 Ga, one that has z=0.040 has a total lifespan of 8 or 9 billion years, and a main sequence lifespan of about 6 or 7 Ga. (according to the Geneva Stellar Evolution grids, anyway).

Besides, surely the interstellar medium is rather varied. Places that are near OB associations for example would be more enriched in metals as several generations of massive stars lived and died and blew up. But a region where there are no massive stars wouldn't become enriched by metals. So it could be quite possible for high metallicity stars to form earlier in the history of the galaxy, surely?

That's why I was wanting to know how to convert the "dex" (Fe/H) values given in the paper to z values, because the geneva grid needs the z values to be useful.

[granthutchison]

High metallicity does correspond to shorter lifespan, yes. But while a 1 solar mass star with z=0.1 (as opposed to Sol's z=0.020) has a total lifespan of only about 4.8 Ga, one that has z=0.040 has a total lifespan of 8 or 9 billion years, and a main sequence lifespan of about 6 or 7 Ga.

But we're not interested in lifespan, we're interested in how recently the star formed - two separate issues.

Besides, surely the interstellar medium is rather varied. Places that are near OB associations for example would be more enriched in metals as several generations of massive stars lived and died and blew up. But a region where there are no massive stars wouldn't become enriched by metals. So it could be quite possible for high metallicity stars to form earlier in the history of the galaxy, surely?

Sure. You'll notice I said "... other things being equal ..." I was simply pointing out that high metallicity and young age are not at all inconsistent for a star that formed in the thin disk at the Sun's current orbital radius around the galaxy - quite the reverse, in fact. I was responding to your line:

... it's is a high metallicity star which means it should be older than the sun but the paper claims it's only 2 Ga in age.

which does seem to imply that all high metallicity stars must be old. Sorry if I picked you up wrongly.

BTW: does that stellar rotation velocity come from the paper you quoted (sorry, I haven't had a chance to link to it yet). I was just thinking that a young star should have a higher rotation velocity, which might account for the fact that the planet's orbit hasn't tidally decayed.

Grant

[Evil Dr Ganymede]

But we're not interested in lifespan, we're interested in how recently the star formed - two separate issues.

True... I'm just saying that not all high metallicity stars must have formed recently is all .

Sure. You'll notice I said "... other things being equal ..." I was simply pointing out that high metallicity and young age are not at all inconsistent for a star that formed in the thin disk at the Sun's current orbital radius around the galaxy - quite the reverse, in fact. I was responding to your line:

... it's is a high metallicity star which means it should be older than the sun but the paper claims it's only 2 Ga in age.[/quote]which does seem to imply that all high metallicity stars must be old. Sorry if I picked you up wrongly.
[/quote]

No, my mistake, I think I was being unclear there - I probably made an unstated assumption somewhere. Sorry!

BTW: does that stellar rotation velocity come from the paper you quoted (sorry, I haven't had a chance to link to it yet). I was just thinking that a young star should have a higher rotation velocity, which might account for the fact that the planet's orbit hasn't tidally decayed.

Yeah, they quote it in the paper. They get it from the low chromospheric activity, from which they infer an age "above ~2 Ga and a rotational period of ~31 days". They seem to imply that the age isn't very far above 2 Ga though.

Here's a direct link to the paper, BTW:
http://www.oal.ul.pt/~nuno/hd160691_letter.ps

[granthutchison]

But we're not interested in lifespan, we're interested in how recently the star formed - two separate issues.

True... I'm just saying that not all high metallicity stars must have formed recently is all .

Ah no. All you demonstrated (with the phrase I quoted) was that high metallicity stars, once formed, last a long time. Not that high metallicity stars could form a long time ago.

(There's a cherished joke about an engineer, a statistician and a logician on a train, all of them entering Scotland for the first time. The punchline is "Ah no. All we can say at present is that at least one side of at least one sheep in Scotland is black." The reconstruction of the rest of the joke is left as an exercise for the interested student.)

Grant

[granthutchison]

Bugger. Keep meaning to say that I don't think you can convert between [Fe/H] and z. Reason being that you need to know the status of every other element apart from Fe before you can can estimate z, and the relationship between Fe and other significant contributors to z is quite variable. O, for instance, is about 30 times more cosmically abundant than Fe, so will swamp its effect on z ... but the Fe/O ratio is heavily dependent on the type of star that enriches the local interstellar medium. Type Ia supernovae give you Fe, type II's give you O.

Grant

[Evil Dr Ganymede]

Bugger. Keep meaning to say that I don't think you can convert between [Fe/H] and z. Reason being that you need to know the status of every other element apart from Fe before you can can estimate z, and the relationship between Fe and other significant contributors to z is quite variable. O, for instance, is about 30 times more cosmically abundant than Fe, so will swamp its effect on z ... but the Fe/O ratio is heavily dependent on the type of star that enriches the local interstellar medium. Type Ia supernovae give you Fe, type II's give you O.

Nuts . Well, that's awkward. Is there not even a vague correlation between the two? Something where you can at least get an idea of whether the star has twice the metallicity, or ten times the metallicity of Sol?

[granthutchison]

Nuts :(. Well, that's awkward. Is there not even a vague correlation between the two? Something where you can at least get an idea of whether the star has twice the metallicity, or ten times the metallicity of Sol?

Well, [Fe/H] is just log(Fe/H) for the star minus log(Fe/H) for the Sun. So 10^[Fe/H] should give you the iron abundance relative to the Sun. If you make the assumption that all the other metals are present in Sunlike ratios, then that translates into an estimate of the relative metallicity in terms of z.
So if [Fe/H] = -1 then the star has a tenth the (z) metallicity of the Sun.
Shakey, though ...

Grant

[Evil Dr Ganymede]

...but good enough for my purposes . Thanks!

[granthutchison]

There's an interesting review of metallicity and stellar evolution at http://nedwww.ipac.caltech.edu/level5/March03/McWilliam/McWilliam6.html. (I'm bringing you in in the middle of it with that URL, at the point that has relevance to what we're discussing.)
There is a definite tendency to a higher O/Fe ratio at lower values of [Fe/H], meaning that Z will be higher in a low-metallicity star than the value you'd derive from the naive equation above. The graphs there might help you derive a jigger factor for your calculation.
The reason for this variation is interesting (well, it is to me, anyway ).
The first supernovae in the early galaxy were Type II, so their enrichment of the interstellar gas was heavy on the O; only after there has been time for white dwarfs to form can you see Type Ia's with their preferential iron-enrichment. So the O/Fe ratio decreases with time while the overall metallicity rises.

Grant

[granthutchison]

The graphs there might help you derive a jigger factor for your calculation.

Here's one suggestion to improve your Z estimate, based on the information in the link I quoted.

The (alpha/Fe) ratio is pretty constant at [Fe/H] > -0.2. So above that value of [Fe/H], you can just use the original

10^[Fe/H]

as an estimate of the metallicity as a proportion of solar metallicity.

At [Fe/H] <= -0.2, first estimate [alpha/Fe] from the relationship

[alpha/Fe] = -([Fe/H]+0.2)/2

then calculate the proportional metallicity as

10^[alpha/Fe] * 10^[Fe/H]

It won't be that useful at very low metallicities or in non-disc populations, but it's an improvement.

Grant

[Evil Dr Ganymede]

Thanks grant, that's very useful.

[Evil Dr Ganymede]

I can't quote it here, but on page 2 of the paper, it gives a Fe/H of +0.32+/- 0.05 dex...

but it gives surface gravity as 4.25+/-0.07 dex too. Why is gravity given in dex?! Or does it mean that the surface gravity is 4.25 multiplied by whatever the dex is (which is presumably 0.32)?

So if the Fe/H is 0.32, that means that it's got about 2.09 as much metals as Sol, which means a z value of about 0.04, right?