[Mesa-users] Issue with rotation in MESA r22.11.1

[Torres, Guillermo]

Hi Matteo,

Thanks for all your comments. See responses below.

On Thu, Jan 23, 2025 at 4:01 PM Matteo Cantiello <matteo.cantiello at gmail.com>
wrote:

> Hi Willie,
>
> On Thu, Jan 23, 2025 at 3:48 PM Torres, Guillermo <gtorres at cfa.harvard.edu>
> wrote:
>
>> Hi Matteo,
>>
>> Thanks for the quick reply.
>>
>> On Thu, Jan 23, 2025 at 3:09 PM Matteo Cantiello <
>> matteo.cantiello at gmail.com> wrote:
>>
>>> Hi Willie,
>>>
>>> It seems like you’re suggesting that the shape of the evolutionary track
>>> is ‘odd,’ but I think it’s more accurate to say that it differs from the
>>> PARSEC model. This difference isn’t entirely surprising, as the MESA
>>> calculation reaches critical rotation, while the PARSEC model does not.
>>>
>>
>> Yes, I certainly agree with that.
>>
>> There could be a variety of factors contributing to the discrepancy --
>>> such as differences in initial conditions, the treatment of mixing, or
>>> angular momentum transport -- and it’s difficult to pinpoint the exact
>>> cause based solely on the inlist and plot you’ve provided.
>>>
>>> If the goal is to reproduce the PARSEC results, you’ll want to ensure
>>> the physics and initial conditions in your MESA setup match those in the
>>> PARSEC calculation as closely as possible. Once that alignment is achieved,
>>> if differences still persist, it would be ideal to run both PARSEC and MESA
>>> side by side and compare results throughout the evolutionary stages. This
>>> approach might help identify where the divergence arises.
>>>
>>
>> My goal is not really to reproduce the PARSEC results exactly or
>> quantitatively (e.g., in terms of luminosity or temperature), as there are
>> probably so many differences in the physics of the two codes that it would
>> be unlikely one could do that. Instead, I was simply struck by the
>> morphology of the blue hook in MESA, which is so different from the
>> 'familiar' shape seen with PARSEC, as well as the few other models for
>> rotating stars that I have seen in the literature.
>>
>
> Have you compared to models in the literature that reach critical rotation
> like yours?
>

No, I haven't. The publicly available tracks I've been able to find so far
in the literature typically only go up to initial w/w_crit = 0.4, and a
track from MESA r22.11.1 at that initial rotation does not reach breakup at
the blue hook (it only gets up to 0.56), so the morphology I get in that
region is the 'familiar' one. But I'll keep looking.


>> Ideally one would want to run PARSEC and MESA in parallel, as you
>> suggest, though this is more easily said than done, and would obviously
>> require access to the PARSEC code. In any case, from feedback from others,
>> it appears I'm not the first to encounter this problem,
>>
>
> I am not sure it is a problem. It would be a problem if we had
> measurements of stars with well defined parameters that are critically
> rotating at H-exhaustion and show a very different HRD position as compared
> to your calculations. But I don't think we have that constraint.
>

This is true. But I also wonder why the PARSEC models for a similar mass
and initial rotation *do not* reach critical rotation at the blue hook,
whereas MESA does. I've been assuming that PARSEC manages to get rid of
angular momentum more efficiently at this mass and initial w/w_crit,
somehow. Still, I agree that having the right observations would be very
useful.


>
>> and it would seem prudent to ignore the part of the tracks that come so
>> close to breakup.
>>
>
> I think it is prudent to assume that we do not have a great grasp of the
> physics of breakup and how to treat the stellar surface layers and mass
> loss at critical rotation.
>

I certainly agree!

But I don't think ignoring the track just because it doesn't look like a
> track of a non-critically rotating star is warranted.
> I understand that there is a preference to select for tracks that look
> smooth and more similar to the ones published in stellar physics books (I
> certainly do not like to publish HRDs with messy tracks, although I have),
> but again I would stress that unless we have clear observational evidence
> that favor a numerical model over another, selecting for aesthetic is not a
> great way to proceed. It is conceivable that stars in short evolutionary
> phases and undergoing some extreme process (like critical rotation) might
> have very complicated tracks.
>

Here, too, I agree. One of the motivations for my initial message was to
find out if anyone else has seen the same behavior, and if nobody has, that
might be an indication that I need to work on my inlist.

Willie




> On Thu, Jan 23, 2025 at 2:53 PM Torres, Guillermo via Mesa-users <
> mesa-users at lists.mesastar.org> wrote:
>
>> All,
>>
>> I am using MESA r22.11.1 to model the evolution of a rapidly rotating
>> star of 5.04 solar masses (initial w/w_crit = 0.55), and I'm finding an odd
>> shape for the evolutionary track in the blue hook region, as the star is
>> contracting. During this phase, I see the model reaching very high values
>> of w/w_crit up to the limit of 0.99, and as it approaches that value, the
>> morphology goes haywire, at least compared to a similar track from the
>> PARSEC v2.0 series (see figure attached, with the odd part highlighted in
>> red). I'm wondering if this really is an issue with the physics in MESA, or
>> just that I don't have the right controls in my inlist. It appears the star
>> would need to get rid of more of its angular momentum to avoid this, but
>> evidently it is not doing that with the controls I am using. I attach my
>> inlist for reference. The pre-main-sequence part is computed separately. I
>> would be thankful for any suggestions.
>>
>> Willie
>> _______________________________________________
>> 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/20250123/279e64b9/attachment.htm>