[Mesa-users] Caveats to implementing time- and space-varying gravitational constant to MESA RSP Cepheid

[Øyvind Christiansen]

Hi Rob and Jeremy,


Thank you for your input. I will look into using other_momentum for including the effect of non-zero derivatives of G.

I will also explore whether

call gyre_set_constant('G_GRAVITY', standard_cgrav) may be called each time-step and the effect of doing this as a first consideration.
For our first attempt we did as you suggested, basing it on your implementation on zenodo!

Best,
Øyvind



________________________________
From: Jeremy Sakstein <sakstein at hawaii.edu>
Sent: 24 February 2022 19:01:00
To: Rob Farmer
Cc: Øyvind Christiansen; mesa-users at lists.mesastar.org
Subject: Re: [Mesa-users] Caveats to implementing time- and space-varying gravitational constant to MESA RSP Cepheid

Hi Øyvind,

I did something similar to what you are trying to do using RSP a while back (https://ui.adsabs.harvard.edu/abs/2021PhRvD.103b4028D/abstract) but only for a constant scaling of [G]  i.e. [G\rightarrow G(1+\alpha)] .

The way that I did this was to set the value of standard_cgrav in extras_controls at the start of each run. You might be able to include time-variations in G by changing standard_cgrav at the start of each timestep in e.g. extras_start_step but I don't know if this will work or not so you will have to test it.

For spatial variations, you'll need to see if other_cgrav or other_momentum is implemented into RSP.

My Zenodo package for the paper is here in case it's helpful: https://zenodo.org/record/4309065#.Yhe4cPXMK84

Cheers,

Jeremy

On Wed, Feb 23, 2022 at 11:26 PM Rob Farmer via Mesa-users <mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>> wrote:
>
> Hi
>
> Assuming you are using the other_cgrav hook then changes in G should be propagated correctly through mesa/star.  GYRE doesn't take into account a changing G (at least by default), if you look in one of the gyre_* test cases in star/test_suite and look in the run_star_extras.f90 you'll see lines like:
>
> call gyre_set_constant('G_GRAVITY', standard_cgrav)
>
> where we are setting G at the start of the run to the standard value.
>
> You'll get energy errors as other_cgrav assumes G is constant so its change won't be taken into account.
> So you would want to look at the other_momentum_implicit hook where you can provide a change to cgrav as well as the derivatives of your change. Though I'm not sure how this will interact with RSP.
>
> Rob
>
>
> On Mon, 21 Feb 2022 at 13:43, Øyvind Christiansen <oyvind.christiansen at astro.uio.no<mailto:oyvind.christiansen at astro.uio.no>> wrote:
>>
>> Dear Mesa users,
>>
>>
>> I am working on a project where we intend to study the effect of temporally and possibly spatially varying gravitational constant on Cepheids using the radially pulsating stellar library, basing it on the test_suite rsp_Cepheid working folder. We are updating the gravitational constant every time step according to our model, and the results seem well-behaved and reasonable. However, I wanted to ask if anyone with more experience in working with Mesa are aware of possible caveats to doing so?
>>
>>
>> Possible difficulties that I see are
>>
>> 1) Energy conservation constraints and termination criterion, if the time varying G transfers energy into or out of the system.
>>
>> 2) External libraries like gyre, and whether they are using the updated value for G at every time step
>>
>> 3) Equations such as the hydrostatic equilibrium where the gravitational pressure is calculated assuming constant G.
>>
>>
>> The variations that we intend to look at for now are around percentage and below, so possibly we will be able argue that these effects are negligible.
>>
>>
>> Any input is appreciated!
>>
>>
>> Best wishes,
>>
>> Øyvind Christiansen
>>
>> _______________________________________________
>> mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>
>> https://lists.mesastar.org/mailman/listinfo/mesa-users
>>
> _______________________________________________
> mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>
> https://urldefense.com/v3/__https://lists.mesastar.org/mailman/listinfo/mesa-users__;!!IBzWLUs!Fd3Tuv9TSIa86RPISg45lEb5JJuqCj1f6lMQ-T56gWvxEwMy2RbIWFY426_FY8h6W2Q$
>


--
Jeremy Sakstein (He/Him/His)
Assistant Professor of Theoretical Physics
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/20220225/2a56021b/attachment.htm>