issue with 'GAMMA' boundary condition in a polytropic model

[repepo]

I'm trying to compute the lowest order g modes using gyre (8.1) of an n=1 polytropic model (generated with --Gamma_1=3.0 using gyre's own build_poly tool). I need to set a no penetration boundary condition on the outer boundary, and according to the online documentation, the option 'GAMMA' fits the bill. However, such an option doesn't seem to be recognized by gyre. By looking at the source code, the closest matches are options 'GAMMA1' and 'GAMMA2' but can't find much info about them. Running gyre using 'GAMMA1' produces an output but the displacement near the outer boundary does not seem to be correct (i.e. the radial derivative of the displacement is not zero there, see attached fig). Running gyre with option 'GAMMA2' produces a core dump.

The command to build the model is:
$GYRE_DIR/bin/build_poly --n_poly=1.0 --dz=0.001 --Gamma_1=3.0 poly.n1_gamma3.h5

Any advice would be much appreciated. Thanks in advance!

Santiago

[rhdtownsend]

Hi Santiago --

Do you really want to be running with one of the GAMMA boundary conditions? These are designed for use with the gamma-pi mode separation approach pioneered by Joel Ong.

If instead you want a no-penetration boundary condition, that's not something currently offered by GYRE. However, I've already had another request for it, and I'd be happy to implement it over the next few days. Let me know!

cheers,

Rich

[repepo]

Hi Rich,

We are interested in the no-penetration boundary condition indeed, in a context not related to gamma-pi modes. We thought the GAMMA option would be appropriate, but we'll follow your advice. If GYRE could offer a no-penetration bc it would be really helpful for us! Hope it is not too difficult to implement, we will certainly appreciate it!

Thanks and cheers!
Santiago

[rhdtownsend]

Hi Santiago — I’ve added the new outer boundary condition to the development version of GYRE — it’s called ‘ZERO_R’ (same name as the corresponding inner boundary condition). Please try it out and see if it works for you! One word of caution when dealing with polytrope models: these are unlike regular stellar models, in that the surface of the star becomes a singular point of the oscillation equations. The ‘VACUUM’ boundary condition in such cases is the only appropriate choice, since it ensures that eigenfunctions remain regular. If you apply the ‘ZERO_R’ condition at the surface of a polytrope, you will see a steep growth in in xi_r and xi_h as the surface is approached, followed by xi_r dropping to zero (much like the plot you originally posted). This is not physical, it’s a numerical artifact. The solution is to set the outer boundary location to be a little bit inside the formal surface where P and rho reach zero. Best wishes, Rich

…

On Oct 13, 2025, at 2:34 AM, Santiago Andres Triana ***@***.***> wrote: repepo left a comment (rhdtownsend/gyre#37) Hi Rich, We are interested in the no-penetration boundary condition indeed, in a context not related to gamma-pi modes. We thought the GAMMA option would be appropriate, but we'll follow your advice. If GYRE could offer a no-penetration bc it would be really helpful for us! Hope it is not too difficult to implement, we will certainly appreciate it! Thanks and cheers! Santiago — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

-- Rich Townsend • Professor of Astronomy Astronomy Department • University of Wisconsin-Madison Phone: 608-262-1752 • E-mail: ***@***.***

[repepo]

Hi Rich, We appreciate it very much! but I must confess I could not find the development version to try it out, is it on github? Thanks once again! Santiago On Mon, Oct 13, 2025 at 2:46 PM Rich Townsend ***@***.***> wrote:

…

*rhdtownsend* left a comment (rhdtownsend/gyre#37) <#37 (comment)> Hi Santiago — I’ve added the new outer boundary condition to the development version of GYRE — it’s called ‘ZERO_R’ (same name as the corresponding inner boundary condition). Please try it out and see if it works for you! One word of caution when dealing with polytrope models: these are unlike regular stellar models, in that the surface of the star becomes a singular point of the oscillation equations. The ‘VACUUM’ boundary condition in such cases is the only appropriate choice, since it ensures that eigenfunctions remain regular. If you apply the ‘ZERO_R’ condition at the surface of a polytrope, you will see a steep growth in in xi_r and xi_h as the surface is approached, followed by xi_r dropping to zero (much like the plot you originally posted). This is not physical, it’s a numerical artifact. The solution is to set the outer boundary location to be a little bit inside the formal surface where P and rho reach zero. Best wishes, Rich > On Oct 13, 2025, at 2:34 AM, Santiago Andres Triana ***@***.***> wrote: > > repepo left a comment (rhdtownsend/gyre#37) > Hi Rich, > We are interested in the no-penetration boundary condition indeed, in a context not related to gamma-pi modes. We thought the GAMMA option would be appropriate, but we'll follow your advice. If GYRE could offer a no-penetration bc it would be really helpful for us! Hope it is not too difficult to implement, we will certainly appreciate it! > Thanks and cheers! > Santiago > — > Reply to this email directly, view it on GitHub, or unsubscribe. > You are receiving this because you commented.Message ID: ***@***.***> -- Rich Townsend • Professor of Astronomy Astronomy Department • University of Wisconsin-Madison Phone: 608-262-1752 • E-mail: ***@***.*** — Reply to this email directly, view it on GitHub <#37 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AZIHRTDQBVCC6NW6RDMHSPT3XONLPAVCNFSM6AAAAACI52DGR6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGOJXGM4DQNJRG4> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[rhdtownsend]

Hi Santiago — It’s on github — instructions for downloading it are here: https://gyre.readthedocs.io/en/stable/ref-guide/installation.html#git-access cheers, Rich

…

On Oct 13, 2025, at 2:41 PM, Santiago Andres Triana ***@***.***> wrote: repepo left a comment (rhdtownsend/gyre#37) Hi Rich, We appreciate it very much! but I must confess I could not find the development version to try it out, is it on github? Thanks once again! Santiago On Mon, Oct 13, 2025 at 2:46 PM Rich Townsend ***@***.***> wrote: > *rhdtownsend* left a comment (rhdtownsend/gyre#37) > <#37 (comment)> > Hi Santiago — > > I’ve added the new outer boundary condition to the development version of > GYRE — it’s called ‘ZERO_R’ (same name as the corresponding inner boundary > condition). Please try it out and see if it works for you! > > One word of caution when dealing with polytrope models: these are unlike > regular stellar models, in that the surface of the star becomes a singular > point of the oscillation equations. The ‘VACUUM’ boundary condition in such > cases is the only appropriate choice, since it ensures that eigenfunctions > remain regular. > > If you apply the ‘ZERO_R’ condition at the surface of a polytrope, you > will see a steep growth in in xi_r and xi_h as the surface is approached, > followed by xi_r dropping to zero (much like the plot you originally > posted). This is not physical, it’s a numerical artifact. The solution is > to set the outer boundary location to be a little bit inside the formal > surface where P and rho reach zero. > > Best wishes, > > Rich > > > On Oct 13, 2025, at 2:34 AM, Santiago Andres Triana ***@***.***> wrote: > > > > repepo left a comment (rhdtownsend/gyre#37) > > Hi Rich, > > We are interested in the no-penetration boundary condition indeed, in a > context not related to gamma-pi modes. We thought the GAMMA option would be > appropriate, but we'll follow your advice. If GYRE could offer a > no-penetration bc it would be really helpful for us! Hope it is not too > difficult to implement, we will certainly appreciate it! > > Thanks and cheers! > > Santiago > > — > > Reply to this email directly, view it on GitHub, or unsubscribe. > > You are receiving this because you commented.Message ID: ***@***.***> > > -- > Rich Townsend • Professor of Astronomy > Astronomy Department • University of Wisconsin-Madison > Phone: 608-262-1752 • E-mail: ***@***.*** > > > > > > — > Reply to this email directly, view it on GitHub > <#37 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AZIHRTDQBVCC6NW6RDMHSPT3XONLPAVCNFSM6AAAAACI52DGR6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGOJXGM4DQNJRG4> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

-- Rich Townsend • Professor of Astronomy Astronomy Department • University of Wisconsin-Madison Phone: 608-262-1752 • E-mail: ***@***.***

[warrickball]

I'm with @repepo here... I can't see that any branch has been updated since September 11, 2025.

[rhdtownsend]

Ooops, my bad — I didn’t actually make the commits! Should be done now...

…

On Oct 13, 2025, at 3:21 PM, Warrick Ball ***@***.***> wrote: warrickball left a comment (rhdtownsend/gyre#37) I'm with @repepo here... I can't see that any branch has been updated since September 11, 2025. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

-- Rich Townsend • Professor of Astronomy Astronomy Department • University of Wisconsin-Madison Phone: 608-262-1752 • E-mail: ***@***.***