Friday, August 19, 2011
Mamajek & Hillenbrand 2008 Fortran Code
I've translated the relevant part of the fortran code to python and am now running it on the list of stars I have. I believe the ages taken from the Mamajek & Hillenbrand 2008 table 13 were derived from R'HK, not B-V. The code uses the actual gyrochronology equation from Barnes 2007 with the coefficient values from Mamajek&Hillenbrand2008. It actually calculates 3 different ages - all using the Barnes 2007 equation, but with coefficients from Barnes, Mamajek, and Meibom et al. 2009. There doesn't seem to be convincing evidence to use one set of coefficients over another, so I'll take the "trusted age" to be the average over these three sets. The errors in the ages are also calculated by the code.
Subscribe to:
Post Comments (Atom)
Greetings Laura--
ReplyDeleteRe: "There doesn't seem to be convincing evidence to use one set of coefficients over another".
I provided the Barnes and Miebom versions of the gyro curves as a courtesy to the user, and to fairly acknowledge the work of others and allow comparison, but I do not advocate using them. My reasons for making a new fit in preference to the Barnes (2007) law are outlined in Mamajek & Hillenbrand (2008). I don't advocate the Miebom law because little work has been done on age-dating M35, and given the large number of points that they provide for that cluster, the M35 members would swamp the curve-fitting (and hing the fit on a cluster whose age is not as well-studied as the Pleiades and Hyades). The Miebom paper quotes "gyro ages" for M35, which defeats the purpose of using the cluster to define the gyro calibration. I'm also concerned about the published ages and reddening E(B-V) for M35 -- I'm not sure it is sufficiently well-characterized to provide a useful calibrator cluster at this time. Long story short, I would still adopt the MH2008 calibration in preference to the B2007 and M2009 calibrations. cheers, Eric M.
Eric,
ReplyDeleteThanks - that makes sense. Also, I was thinking about A stars the other day and thought you might have some input:
I'm working to date A stars using Teff vs logg isochrones from Song et al. 2001. But is there some reason that gyrochronology can't be used? The equation from Barnes 2007 describes the evolution resulting from magnetic braking, but shouldn't there be some critical rotational isochrone above which stars are fully radiative and evolve Skumanich-ly?
Thanks for your help!
-Laura
Interesting question. The B/A/early-F stars do spin down and they evolve on the main sequence and off of it. Unfortunately rotation periods are not easily measured for the high mass stars as they tend not to show variability due to starspots at the same level of detectability as the late-F/G/K/M stars. The rotational evolution of the host stars has been studied via their projected rotational velocities (vsini), so one has the sine(inclination) ambiguity. Hot stars also spend much shorter times on the main sequence, so they have less time to brake than do the longer lived lower-mass stars. So unfortunately, among high mass stars in a coeval cluster, one will see fast rotators rotating near brake-up speeds (hundreds of km/s) and "slow" rotators (~tens km/s; which are almost certainly fast rotators seen pole-on) -- but one doesn't see nice sequences of vsinis among coeval samples (let alone among field stars of varying ages). So I have not attempted using rotation for stars earlier than about ~F5/F6 or so. - Eric M.
ReplyDelete