[Mesa-users] ZAMS models: a bug or feature?

[Warrick Ball]

Hi Natasha,

I just had a look at the ZAMS models, which you can find in

     $MESA_DIR/data/star_data/zams_models/zams_z2m2_y28.data

and I think you've found a genuine bug.  There are models with masses 1.000, 1.259 and 1.584 and the 1.259 Msun model has a central H abundance of 0.574.

For anyone looking to confirm, each new model has a line with "M/Msun" and the mass, so a quick way to see the issue is to find those lines and look 2 lines before for the last line of the previous model e.g.

     $ grep -B 2 M/Msun $MESA_DIR/data/star_data/zams_models/zams_z2m2_y28.data | less -S

All the other models have X₀ > 0.69 so should be okay.  The inlists to generate the ZAMS models are in

     mesa/mesahub/data/star_data/zams_models/create_z2m2_y28

so we'll check if those work ASAP to create appropriate ZAMS models and, if not, fix them.

We'll investigate this right away.  Note that ZAMS models either side of 1.000 and 1.584 Msun should be okay (but we'll definitely check!).

Cheers,
Warrick


___________

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

On Tue, 18 Jul 2023, Natasha Ivanova 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.
> Hello all,
> 
> By chance, I have found that the use of pre-calculated models (create_pre_main_sequence_model = .false.) in MESA version 23.05.1 leads to a huge discrepancy with the use of the models that are obtained as pre-MS (create_pre_main_sequence_model =
> .true.).
> 
> A simple inlist to reproduce is attached.
> 
> Specifically, the problem seems to be in that the stored low mass ZAMS model(s) have a significantly depleted central hydrogen -- see the table below, where I provided ages at the end of the MS, obtained with and without pre-MS calculations (and so
> MS starts with X=0.7), and also the H content in the center with which the calculations start in the stored zams model (ZAMS X_0).
> 
> The minimum central H content seems to be for 1.25 Msun - it is as small at the zero age as 0.578248. This also leads to a more massive star having a longer MS lifetime.
> 
> For me, it seems that the trend in deleted central H concentration is due to only one stellar model with a very low central H, and the rest of the models are interpolated; I also believe that that specific stored model is wrong. So my question is, is
> that a bug (an erroneous stored model) or a feature, and if that is the physical feature, then why?
> 
> My question is motivated by the fact that I used to use the ZAMS models when teaching my stellar astro class, in the first introductional computational lab (and then later on). But if one can no longer believe in pre-calculated ZAMS models, I need to
> update my lab(s) - unless of course I am missing something and that is, for some reason, how it should be.
> 
> 
> M/Msun Age with preMS Age from "ZAMS"      ZAMS X_0
> 1.0 8.7599E+09     8.7154E+09          0.699498
> 1.1 5.6064E+09     5.0309E+09      0.666448
> 1.2   4.2840E+09     3.4859E+09        0.614212
> 1.23 3.9819E+09 3.1354E+09      0.587010
> 1.25 3.7561E+09 2.9156E+09      0.578248
> 1.27 3.6174E+09 2.7330E+09      0.583083
> 1.28 3.4796E+09 2.6863E+09      0.596698
> 1.29 3.4140E+09 2.6717E+09      0.605975
> 1.3 3.3239E+09 2.6728E+09      0.613373
> 1.4 2.6296E+09 2.2952E+09      0.650336
> 1.5 2.1518E+09 1.9827E+09      0.672308
> 
> Best,
> Natasha
> 
>