[Mesa-users] astero grid

[Warrick Ball]

Dear Bruno,

Thanks for sending all of your input files.  A couple of things stand out to me, though I haven't tried to re-run your inlists yet.

First, your `inlist_astero_search_controls` seems to contain a number of controls that don't exist.  Specifically, I couldn't find any of `search_for_best_match`, `search_*_param`, `grid_size_*` or `freq_data_file` defined anywhere in the `astero` module, nor among any of the defaults for any inlist.  I'm not entirely sure how this even runs: the `astero` module should produce errors!  If you could point to the part of the documentation that suggested using those, let me know so we can correct it.  I apologise that there's still no high-level module documentation for `astero` but the `astero_search_controls` reference [1] should help.  You might also find this old guide I wrote [2] useful, though note that the specific controls it describes are *very* out of date.

So my hunch is that you aren't actually using the `astero` module.  Did you create a new work folder by copying `astero/work` instead of `star/work`?  Or perhaps the `astero/test_suite/example_astero` test case?  The latter is the most complete example, as opposed to the `astero/work` folder, which is minimal.

That would probably also explain why the PGSTAR plots don't start.  I've not used them much but also note that I'm not sure they're always produced anyway.  The star might have to get "close enough" (defined by the user) to the observed parameters for the various `astero` related quantities to show.

Regarding the stopping condition, the condition in `inlist_project` will basically override anything in `astero` and stop the evolution.  If `astero` is in any sort of search mode and no χ² has been computed yet, it'll report χ²=NaN or ∞ and move onto the next iteration with that information.  It's usually best to start by experimenting with `search_type = 'use_first_values'` and find some parameters that are good enough to trigger the `astero` routines before starting a search.

Regards,
Warrick

[1] https://docs.mesastar.org/en/latest/reference/astero_search.html
[2] https://mesastar.org/marketplace/guides/astero-module/

___________

Warrick Ball
Senior Research Software Engineer
University of Birmingham
W.H.Ball at bham.ac.uk

On Tue, 15 Jul 2025, Bruno Lustosa via Mesa-users wrote:

> CAUTION: This email originated from outside the organisation. Do not click links or open attachments unless you recognise the sender and know the content is safe.
> Dear MESA developers and users,
> 
> I am trying to perform an asteroseismology grid search using MESA version r24.08.1. My goal is to evolve a star to a specific giant branch phase and then trigger a grid search to find the best-fitting model for a set of observed frequencies.
> 
> The problem I am facing is that while the stellar evolution runs perfectly and stops at my specified condition (using a sequence of 1 to 4 stops), the subsequent asteroseismology search phase is not initiated. The run simply terminates after saving
> the final model, and the PGSTAR plots for 'astero' (Echelle, χ2, etc.) never appear.
> 
> I have already checked the following configurations, based on the documentation:
>
>  * 
>
>     My inlist_project contains the &astero namelist with do_astero_search_op = 'grid'.
>
>  * 
>
>     The astero_search_controls_inlist_name parameter correctly points to my inlist_astero_search_controls file.
>
>  * 
>
>     My inlist_astero_search_controls uses the modern grid definition syntax (grid_type = 'uniform', along with the uniform_grid_... parameters for age, Z, and alpha_mlt). The older, deprecated search_*_param flags are not being used.
>
>  * 
>
>     My frequency data file (freqs.dat) has been checked for duplicates and is correctly formatted (l, nu, nu_err).
> 
> My questions are: Is there any other configuration, flag, or known issue that might prevent the transition from the evolution phase to the search phase after a clean stop? Could the type of stopping condition be interfering with the 'astero' trigger?
> And why do no plots appear, even when astero_pgstar_controls is specified?
> 
> I am attaching my main inlist, inlist_project, and inlist_astero_search_controls for your review.
> 
> Thank you very much for any help or suggestion.
> 
> Sincerely,
> 
> Bruno
> 
> 
> 
> 
>