[mesa-users] A new parametrization of the surface effect

Bill Paxton paxton at kitp.ucsb.edu
Wed Aug 6 12:27:57 EDT 2014


Hi Warrick,

Thanks for the news.  Let's get those new options for surface corrections into mesa so future users don't have to change the sources to try them.  At the same time, let's add hooks so that in the future new methods can be tried more easily -- i.e., without modifications to the mesa sources.  Either Rich Townsend or I will make the changes in time for the next release; we'll be in touch with you if we get confused!

Thanks again,
Bill




On Aug 6, 2014, at 2:37 AM, Warrick Ball wrote:

> Hi all,
> 
> Our recently accepted paper on surface effects appeared on arXiv this morning:
> 
> "A new correction of stellar oscillation frequencies
> for near-surface effects"
> http://arxiv.org/abs/1408.0986
> 
> In it, we argue for the formula frequency^3/(mode inertia) as a good-fitting functional form for the surface effect. (i.e. the systematic difference between modelled and observed oscillation frequencies.)  An even better fit is found when there's an extra term proportional to frequency^(-1)/(mode inertia), but, though it fits the difference with the Sun quite well, the extra term doesn't really improve the fit for the distant star we tried, HD 52265.  (The same goes for other stars that I've been fitting between submission and acceptance of the paper.)
> 
> We made use of MESA models, and the fit was included by modifying the surface correction subroutines in astero_support.f.  In the spirit of the open sourceness of MESA, I've attached the modified versions that I wrote, as well as the tweaked makefile.  The .cube file corresponds to the "cubic" surface term (eqn 3 in the paper), and the .both file to the "combined" term (eqn 4).  In short, I simply replaced the get_freq_corr subroutine with my own routine, which calculates the best-fitting coefficients for the surface term (in a least squares sense).  I put the modified astero_support.f files in the src/ directory, hence the modification to the makefile.
> 
> Watch out though: I also modified the chi2 function because I didn't want a weighted mean of reduced chi-squareds from different sources, which is the default.  If you use these files, you may wish to remove the changes to the chi2 subroutines.
> 
> I'm no master coder, and my snippet is ugly, brutal, minimally-functional code.  When I have a bit of time, I intend to add an option for which correction to use. e.g. to make an option in &astero_search_controls something like
> 
>    which_surface_correction = 'none' ! default
>      ! options are:
>        !  'none'        No frequency correction
>        !  'power_law'   Correction of Kjeldsen et al. (2008)
> 	!  'cubic'       Cubic correction of Ball & Gizon (2014, eqn 3)
>        !  'combined'    Combined correction of Ball & Gizon (2014, eq 4)
> 
> so I don't have to choose the correction by using differently compiled binaries.  I'll update the list if/when I get that far.  I'm also interested in implementing other new options that might be in the works.
> 
> All comments welcome!
> 
> Cheers,
> Warrick
> 
> 
> ------------
> Warrick Ball
> Postdoc, Institut für Astrophysik Göttingen
> wball at astro.physik.uni-goettingen.de
> +49 (0) 551 39 5069<astero_support.f.cube><astero_support.f.both><makefile.txt>------------------------------------------------------------------------------
> Infragistics Professional
> Build stunning WinForms apps today!
> Reboot your WinForms applications with our WinForms controls. 
> Build a bridge from your legacy apps to the future.
> http://pubads.g.doubleclick.net/gampad/clk?id=153845071&iu=/4140/ostg.clktrk_______________________________________________
> mesa-users mailing list
> mesa-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mesa-users





More information about the Mesa-users mailing list