# Conflicting information about the star Delta Pavonis?

Delta Pavonis has about the same mass as the sun so I would think that the evolutionary path would be about the same (the mass is given as O.991 M(Sun)/ not sure how we get this so accurately). It has been estimated to be about 6.6 to 6.9 Ga with a luminosity of 1.22 L(Sun). This all makes sense if you put it on the main sequence where we will be in a couple of billion years (after T(max) in about a billion years). The problem is that the surface temperature is only 5550K (spectral) and for this reason is thought to be in subgiant phase which isn't supposed to happen to us for about another 5.5 billion years. Delta Pav also has a very high metallicity [Fe/H] of +0.33 which also screams "subgiant". What am I missing?? (data from Wikipedia sources)

To answer a question, we do need the source of information (wikipedia just isn't good enough).

According to the SIMBAD database, delta Pavonis is a G8 subgiant, with a temperature of 5512K and a $\log g$ of 4.23 (in cgs units) (Gray et al. 2006), which looks correct for a subgiant classification. The metallicity is given as $+0.13$ in the same source.

There are a large number of other spectroscopic determinations of the star's parameters listed in the SIMBAD measurements. All seem to agree on the subgiant nature, but for example Bensby et al. (2014) makes the star a little warmer (5635 K) and more metal rich (+0.37).

So what appears to be at odds with this is the age/mass combination that you quote.

The wikipedia entry and information you quote appears to be taking information from Takeda et al. (2007). They estimate the age and mass of the star from the temperature and gravity using a Bayesian fitting method and the YREC stellar evolution models.

If this is really important to you, then you need to read the Takeda et al. paper. They show that the posterior probability distribution of age and mass can have multiple peaks. In the case of delta Pavonis it appears that wikipedia has got things wrong (what a surprise). The age that is quoted is actually for a secondary peak in the posterior probability distribution. The secondary peak in the mass distribution is for $M= 1.101M_{\odot}$. I think that age, mass pair makes much more sense.

They do not provide detailed information/discussion on individual cases. Like you, I cannot see how they arrive at 1 solar mass for a star of any age, unless they have picked up a pre main sequence solution in the HR diagram - for which the age corresponding to 1 solar mass would be very small. However, they claim that their models start at the ZAMS. The fact that they list no primary peak in the age posterior pdf suggest to me that the pdf look like a declining function from zero (the ZAMS), with a secondary peak at the 7Gyr (quoted by wikipedia). This then suggests that this secondary solution is also not a brilliant fit and that the fitting process would really have liked to make this younger than ZAMS - i.e. a PMS star (which it isn't). In turn then, this may suggests that the gravity used by Takeda et al. might be too low.

Takeda et al. used spectroscopically derived parameters of temperature 5590 K, $\log g= 4.31$ and $[M/H]=+0.26$ from Valenti & Fischer (2005). These authors also did their own isochrone fits to these parameters finding $M=1.045 \pm 0.058 M_{\odot}$ and an age of 5.3-6.8 Gyr. They say that the mass at the best-fitting isochronal age is $1.07M_{\odot}$.

I think part of what is puzzling you is that you have to remember that having a metal content that is twice the Sun's does affect the evolutionary tracks considerably. It is not that the evolutionary timescales are massively different, but the effective temperature of the star is changed considerably for a given mass and age.

I haven't looked in too much detail, but the first point is to note that there are two estimates of the age: 6.6–6.9 Gyr and 9.3 Gyr. The latter is given without uncertainties, but I looked it up on Vizier and the range is 5.8–10.7 Gyr, basically from isochrones. The former is computed using gyrochronology or activity-age relations, which are usually only calibrated up to about the age of the Sun: 4.6 Gyr. So this star is probably older than the oldest calibration point. In addition, recent results from asteroseismology (some presented at a conference just a few months ago) suggest that the activity-rotation-age indicators are unreliable in older stars.

All this is basically pointing to the fact that measuring a star's age is difficult and model/calibration dependent. I wouldn't take those results too seriously. The 5.8–10.7 Gyr range is the more realistic range.

Incidentally, higher metallicity stars tend to be redder, so I'm not surprised that it would be a bit redder than an older version of the Sun.