[Mesa-users] Schwarzschild and Ledoux Criteria

Sat Jul 20 15:35:59 EDT 2019

Dear Tomek,

Schwarzchild and Ledoux criteria should indeed give the same position for the convective boundaries, on fact even whether or not semiconvective mixing is used, as explained in the Gabriel et al  paper you cite. 

We also explain in the last two MESA instrument papers (https://ui.adsabs.harvard.edu/abs/2018ApJS..234...34P <https://ui.adsabs.harvard.edu/abs/2018ApJS..234...34P> and https://ui.adsabs.harvard.edu/abs/2019ApJS..243...10P <https://ui.adsabs.harvard.edu/abs/2019ApJS..243...10P>) that, in MESA (and in several other stellar evolution codes) this is clearly not always the case due to the simple condition used to position the convective boundaries (grad rad - grad ad =0 at the boundary).Using this simple equation to position the convective boundaries fails in the presence of composition discontinuities, or composition gradients.
It is therefore necessary, to correctly position the convective boundary, to be very careful with the implementation of this condition. 
This is the reason why we implemented two new algorithms in MESA, called « predictive mixing » and « convective pre-mixing » to remedy this problem. 

I suggest you carefully read the sections on Convective Boundaries in the last two MESA instrument papers.
Then, try one of those two new algorithms by adding the appropriate controls in your inlist. You should then get the same evolutionary tracks whether you use Schwarzschild or Ledoux criteria.

Below, I attach the info contained in the controls_default file concerning these new controls. They are not set to true by default, so you have to explicitly put them in your inlist and set them to true (note: you can only use one of these two prescriptions).

Do not hesitate to get back to us after you do that if things are still unclear or if you have additional questions about this topic.

Cheers

Anne

!## Predictive mixing

         ! Predictive mixing is an approach for expanding convective boundaries until
         ! gradr = grada on the convective side of the boundary (as required by the criterion
         ! that the convective velocity and luminosity vanish at the boundary). It is discussed
         ! in detail in Paxton et al. 2018, ApJ, in press: "Modules for Experiments in Stellar 
         ! Astrophysics (MESA): Convective boundaries, element diffusion, and massive star explosions"

         ! Predictive mixing is controlled by specifying a set of parameters, which combines matching
         ! criteria (determining which boundaries to apply the predictive mixing to) together with
         ! values (determining how the predictive mixing should operate at those boundaries). Up to
         ! `NUM_PREDICTIVE_PARAM_SETS` of these parameter sets can be defined (see `star_def.inc` for value).

            !### predictive_mix

            ! Set to .true. to enable this set of parameters

         predictive_mix(1) = .false.

            !### predictive_zone_type

            ! Matching criterion for the type of the convection zone. Possible values are `burn_H`
            ! (hydrogen burning), `burn_He` (helium burning), `burn_Z` (metal burning), `nonburn`
            ! (no burning) or `any` (which matches any type of zone).

         predictive_zone_type(1) = ''

            !### predictive_zone_loc

            ! Matching criterion for the location of the convection zone. Possible values are `core`
            ! (the core convection zone), `shell` (a convective shell), `surf` (the surface convection
            ! zone) or `any` (which matches any location).

         predictive_zone_loc(1) = ''

            !### predictive_bdy_loc

            ! Matching criterion for the location of the convective boundary. Possible values are
            ! `top` (the top of the convection zone), `bottom` (the bottom of the convection zone)
            ! or `any` (which matches any location).

         predictive_bdy_loc(1) = ''

!## Convective premixing

         ! Convective premixing is a approach to handling mixing in convection zones that improves
	 ! upon the predictive mixing scheme described above. Like predictive mixing, it expands
	 ! convective boundaries until gradr = grada on the convective side of the boundary. Unlike
	 ! predictive mixing, it directly modifies abundances in the stellar model, via a iterative
	 ! series of mixings-to-homogeneity over a shifting window of cells. This iterative approach
	 ! allows convective premixing to build 'classical' semiconvection regions, where the abundance
	 ! gradient is tuned to yield convective neutrality.

            !### do_conv_premix

            ! Set to .true. to perform convective premixing. Note that this cannot be enabled at the
	    ! same time as the `predictive_mix` control

         do_conv_premix = .false.

> Le 20 juil. 2019 à 20:41, Tomasz Plewa via Mesa-users <mesa-users at lists.mesastar.org> a écrit :
> 
> In theory, see for example Eq. 10 in Gabriel et al. (2014, AA, 569, A63), one should be able to obtain a series of Ledoux-based models by gradually increasing \alpha_sc from zero. All other parameters being equal, we expected that the first model in that Ledoux series to match the Schwarzschild model. But that did not seem the case as the two trajectories diverged forcing us to produce a Schwarzschild series separately.
> 
> Perhaps there exists a simple explanation for the observed lack of equivalence between the two models in the alpha_sc limit without getting buried in implementation details. That may not be possible, though, in which case a reference would be helpful.
> 
> Tomek
> --
> On 7/20/19 12:59 PM, anne.thoul at uliege.be <mailto:anne.thoul at uliege.be> wrote:
>> Hi Antigoni
>> 
>> Could you give us more information about what you are doing?
>> 
>> Which version of MESA are you using? 
>> 
>> There has been a lot of recent improvement done in MESA concerning the convective boundaries.
>> In the 4th instrument paper we introduced « predictive mixing » and in the fifth instrument paper we introduced « CPM » or « convective pre-mixing ». 
>> 
>> Which of these prescriptions are you using?
>> 
>> I suggest you read those instrument papers (at least the sections about convective boundaries), and try to use one of these prescriptions for the convective boundaries. They improved considerably what the older versions of MESA did. Those prescriptions are not, however, the default at this point, so you have to specifically put the necessary
>> controls in your inlists.
>> 
>> Those controls are explained in the instrument papers, and in the controls_default file.
>> 
>> If, however, your problems persist after having used the newer convective boundaries settings, please let us know and send us the inlist you use so that we can try to reproduce your problem.
>> 
>> Cheers
>> 
>> Anne
>> 
>> 
>> 
>> 
>>> Le 20 juil. 2019 à 18:26, Antigoni Georgiadou via Mesa-users <mesa-users at lists.mesastar.org <mailto:mesa-users at lists.mesastar.org>> a écrit :
>>> 
>>> Hello,
>>> 
>>> I am working on stellar evolution applications with the MESA code and I wanted to ask about the way MESA operates for the Schwarzschild and Ledoux criteria. 
>>> 
>>> For example, for a massive star of 30 Msun using the Schwarzschild criterion does not seem to be equivalent to using the Ledoux criterion with semiconvective parameter equal to 0.
>>> 
>>> I would really appreciate it if you can help me understanding this difference. I think it shouldn't appear any different.
>>> 
>>> Many thanks,
>>> 
>>> Antigoni Georgiadou
>>> PhD Candidate
>>> Department of Mathematics
>>> Florida State University
>>> 
>>> _______________________________________________
>>> mesa-users at lists.mesastar.org <mailto:mesa-users at lists.mesastar.org>
>>> https://lists.mesastar.org/mailman/listinfo/mesa-users <https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.mesastar.org_mailman_listinfo_mesa-2Dusers&d=DwMFaQ&c=HPMtquzZjKY31rtkyGRFnQ&r=BAkmbcCLNe7YO36WZjdRtg&m=KeZX2uMm225woNN9WtCp_JVbG4KW80spkdH8-rpHKcg&s=itTxP3kXHuKQ7z6vFIVEGPWXjzdSizwOHnbg1fZSX4E&e=>
>> 
>> 
>> _______________________________________________
>> mesa-users at lists.mesastar.org <mailto:mesa-users at lists.mesastar.org>
>> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.mesastar.org_mailman_listinfo_mesa-2Dusers&d=DwICAg&c=HPMtquzZjKY31rtkyGRFnQ&r=BAkmbcCLNe7YO36WZjdRtg&m=KeZX2uMm225woNN9WtCp_JVbG4KW80spkdH8-rpHKcg&s=itTxP3kXHuKQ7z6vFIVEGPWXjzdSizwOHnbg1fZSX4E&e= <https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.mesastar.org_mailman_listinfo_mesa-2Dusers&d=DwICAg&c=HPMtquzZjKY31rtkyGRFnQ&r=BAkmbcCLNe7YO36WZjdRtg&m=KeZX2uMm225woNN9WtCp_JVbG4KW80spkdH8-rpHKcg&s=itTxP3kXHuKQ7z6vFIVEGPWXjzdSizwOHnbg1fZSX4E&e=> 
>> 
> 
> _______________________________________________
> 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/20190720/1cdb6c76/attachment-0001.html>