[Mesa-users] Possible bug in low-temperature tau atmosphere tables in MESA r23.05.1

Jeremy Sakstein sakstein at hawaii.edu
Wed Sep 30 22:09:32 UTC 2026


Hi Rich,

I just made a new issue on the GitHub.

Thanks,

Jeremy

On Wed, Sep 30, 2026 at 4:28 AM Richard Townsend <townsend at astro.wisc.edu>
wrote:

> Hi Jeremy —
>
> This does indeed smell like a bug. Could you open an issue on the github
> repository:
>
>
> https://urldefense.com/v3/__https://github.com/MESAHub/mesa/issues__;!!PvDODwlR4mBZyAb0!T1--qwpzzejc51sVz0FyWe5qGLt06H4u2E2L-rpgzKsCl-pS9_pS-__6FhhbzGIs8y4u5Hb1J1zmGKRQjXtHQLqV$
>
> …adding the same analysis as you do here? That way, we can get the problem
> properly tracked.
>
> Thanks for reporting this!
>
> Best wishes,
>
> Rich
>
> > On Sep 29, 2026, at 3:23 PM, Jeremy Sakstein via Mesa-users <
> mesa-users at lists.mesastar.org> wrote:
> >
> > Hi everyone,
> >
> > I think I may have found a bug in the generation of the low-temperature
> tau_10, tau_1, and tau_1m1 atmosphere tables in MESA r23.05.1, and wanted
> to check whether this is known or whether I am misunderstanding the
> intended behavior.
> >
> > Looking at atm/preprocessor/src/create_tau100.f
> >
> > the COND reader appears to contain
> >
> > T(i,j) = vals(num_layers)      (line 403)
> >
> > while the pressures are read using
> >
> > Pg = vals(which_cond_layer)     (line 411)
> > Pe = vals(which_cond_layer)     (line 413)
> >
> > Here num_layers = 50 (line 68), while the relevant layer indices are
> >
> > cond_layer_tau10  = 44  (line 48)
> > cond_layer_tau100 = 50 (line 48)
> >
> > Thus, unless I am missing something, in the COND regime the tau_10 table
> receives the pressure from the tau=10 layer but the temperature from layer
> 50, corresponding to tau=100. The tau_100 table is unaffected because
> cond_layer_tau100 = num_layers = 50.
> >
> > I also checked the atmosphere data shipped with my r23.05.1
> installation. At Teff = 3000 K and log g = 5.0, for example, the
> temperature is identical in tau100_T.data, tau10_T.data, tau1_T.data, and
> tau1m1_T.data:
> >
> > T = 4279.834 K
> >
> > The same equality occurs throughout the high-gravity COND part of the
> tables, whereas the temperatures differ between optical depths at lower
> gravity where a different code path is used.
> >
> > Changing the corresponding source line to
> >
> > T(i,j) = vals(which_cond_layer)
> >
> > looks to me like the intended behavior.
> >
> > I have only verified this in my local r23.05.1 installation and have not
> established whether newer releases are affected.
> >
> > Cheers,
> >
> > Jeremy
> >
> > --
> > Jeremy Sakstein
> > Associate Professor
> > Department of Physics & Astronomy
> > University of Hawaiʻi at Mānoa
> > Watanabe Hall
> > 2505 Correa Road
> > Honolulu, HI, 96822, USA
> >
> https://urldefense.com/v3/__http://www.jeremysakstein.com__;!!PvDODwlR4mBZyAb0!T1--qwpzzejc51sVz0FyWe5qGLt06H4u2E2L-rpgzKsCl-pS9_pS-__6FhhbzGIs8y4u5Hb1J1zmGKRQjdV0bIS3$
> > _______________________________________________
> > mesa-users at lists.mesastar.org
> >
> https://urldefense.com/v3/__https://lists.mesastar.org/mailman/listinfo/mesa-users__;!!Mak6IKo!Otwp8SoChjX0DN7BBkhc62UPBuEW_GS1bynvBlffo0TLVFiva5Qs7RWoU3Q7nwTApVN3Q-x3ysyoYr9kcfFHilXp835Z1HY$
> >
>
> --
> Rich Townsend • Professor of Astronomy
> Astronomy Department • University of Wisconsin-Madison
> Phone: 608-262-1752 <(608)%20262-1752> • E-mail: rhtownsend at wisc.edu
>
>
>
>
>
>

-- 
Jeremy Sakstein
Associate Professor
Department of Physics & Astronomy
University of Hawaiʻi at Mānoa
Watanabe Hall
2505 Correa Road
Honolulu, HI, 96822, USA
http://www.jeremysakstein.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20260930/8c615318/attachment.htm>


More information about the Mesa-users mailing list