[Mesa-users] Astero: delta_nu with a single radial frequency

Tue Oct 8 10:11:25 EDT 2019

Thanks all.

Warrick was able to clarify was I was asking - about how astero treats an observable delta_nu.

Rich and I looked into it a bit more and saw that in astero_support.f if there is only a single radial frequency, delta_nu_sigma is not set to be greater than zero,
In extras_support.f, if delta_nu_sigma is not greater than zero, then chi2_delta_nu=0.
So it looks like in this case delta_nu is bypassed as a check parameter.

Sincerely,

Jacqueline Goldstein
PhD Graduate Student | Astronomy
University of Wisconsin - Madison
http://www.astro.wisc.edu/~jgoldstein

On Oct 8, 2019, at 3:24 AM, Warrick Ball <W.H.Ball at bham.ac.uk<mailto:W.H.Ball at bham.ac.uk>> wrote:

Hi Jacqueline et al.,

Thanks for picking this up!  This looks like a case where the astero module might be doing something fishy and isn't documented well anyway so I'll look into it.  This message is a quick answer of what might be happening (based on the default work folder) but I'll start going through the code and fixing whatever code or docs need fixing.

First, Earl and Kuldeep, I don't think Jacqueline is asking about how MESA computes the model's `delta_nu` without enough radial modes.  Rather, I think the question is where does the astero module get the *observed* `delta_nu` when there's only one radial mode.

The documentation in the default inlist says that it's inferred from the radial mode frequencies if `delta_nu < 0`, which is what the block that Jacqueline highlighted presumably does.  But it doesn't make sense to do a linear fit to frequency as a function of radial order if there's only one point.  So what is the astero module doing?

At the moment, I don't know, but having run the code, my suspicion is that it's simply bypassing the phase where it tries to match `delta_nu`.  As a reminder, the astero module evaluates increasingly expensive components of χ² as the star evolves.  The partial evaluations are

1. the "spectroscopic" (really, non-seismic) components,
2. `delta_nu`,
3. the radial mode frequencies, and finally
4. everything, including the non-radial mode frequencies.

If each evaluation is below some threshold, the next component is evaluated too.  So I think in this case the astero module is just bypassing part 2.  The side effect in the `example_astero` test case is that MESA starts taking much smaller timesteps and stops long before it's even close to matching the one radial mode.  (My run terminated at 0.14 Gyr, whereas the default model should be about 4.64 Gyr old.)

Still, all this is just from running MESA/astero, rather than reading the code, which is what I'll start on.

Cheers,
Warrick

------------
Warrick Ball
Postdoc, School of Physics and Astronomy
University of Birmingham, Edgbaston, Birmingham B15 2TT
W.H.Ball at bham.ac.uk<mailto:W.H.Ball at bham.ac.uk>
+44 (0)121 414 4552

On Tue, 8 Oct 2019, mesa-users at lists.mesastar.org wrote:

Hi Jacqueline,
As Earl replied, delta_nu is calculated using the asymptotic expression (not using the scaling relation though).
Here is what grep has to say about the calculation of delta_nu:
 star/public/star_data.inc:      real(dp) :: delta_nu ! in micro Hz.  = 10^6/(2*photosphere_acoustic_r)
Best wishes,
Kuldeep
On Tue, Oct 8, 2019 at 1:45 AM Earl Bellinger via Mesa-users <mesa-users at lists.mesastar.org> wrote:
     Hi Jacqueline,
I haven't checked, and this is only a guess, but is it possible that it's using the asymptotic value? That is:
    Delta_nu/Delta_nu_solar = sqrt(M/R**3)
with M and R being in solar units and Delta_nu_solar ~= 135.1 muHz.
Best regards,
Earl
On Tue, Oct 8, 2019 at 6:17 AM Jacqueline Goldstein via Mesa-users <mesa-users at lists.mesastar.org> wrote:
     Hi MESA users,
How does the astero module determine/use delta_nu when only provided a single radial frequency?
In my inlist, I am only using first values and only using aster-seismic constraints.
The radial frequency is:

       chi2_seismo_fraction = 1d0          # only seismic constraints
       normalize_chi2_spectro = .false.  #need when chi2_seismo_fraction = 1.

       nl0 = 1 ! number of observed l=0 modes

       l0_obs(1) =  66.704 !                     # microhertz
       l0_obs_sigma(1) = 0.001
          l0_n_obs(1) = 1                            # radial order
I have three additional dipole frequencies.
I am leaving the default value of

     delta_nu = -1
delta_nu_sigma = -1
The module runs, but how does the astero module treat delta_nu with this single radial frequency?
In astero_support.f I see:
if (delta_nu <= 0 .and. nl0 > 1 .and. l0_n_obs(1) > 0) then…
But I can’t determine what happens when delta_nu <=0 I only have a single radial frequency.
Sincerely,
Jacqueline Goldstein
PhD Graduate Student | Astronomy
University of Wisconsin - Madison
http://www.astro.wisc.edu/~jgoldstein
_______________________________________________
mesa-users at lists.mesastar.org
https://lists.mesastar.org/mailman/listinfo/mesa-users
_______________________________________________
mesa-users at lists.mesastar.org
https://lists.mesastar.org/mailman/listinfo/mesa-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20191008/2f2c799c/attachment.html>