The GYRE Forums

Hello Rich,

I was looking for a code to compute non-adiabatic pulsations and found the GYRE Code.
I'm actually interested in computing solar oscillation eigenfunctions. From the webpage I conclude that GYRE is designed for the computation of stellar oscillations where low degrees are of interest. However, are there any internal restrictions concerning the computation of modes of higher degree, say up to 200?

Anyway, I installed GYRE 3.0 on my Mac and applied it to a FGONG-file of solar model S (see attachment + input list file). First, I wanted to compare the adiabatic outcome from GYRE with the one from ADIPLS. For modes of low harmonic degree it seems to work fine. However, when I increase the harmonic degree, e.g. l>40, I still get a result, but the radial grid is no longer monotonic increasing near the center and shows additional grid points that are negative. The eigenfunctions near the center also look not meaningful; near the surface, they seem to be okay.

I played around with different choices for the shooting and the reconstruction grid, as well as the numerical solver, but this did not helped. Do you have any idea how I can solve this problem?

Thank you very much in advance, pwave.

pwave wrote:Hello Rich,

I was looking for a code to compute non-adiabatic pulsations and found the GYRE Code.
I'm actually interested in computing solar oscillation eigenfunctions. From the webpage I conclude that GYRE is designed for the computation of stellar oscillations where low degrees are of interest. However, are there any internal restrictions concerning the computation of modes of higher degree, say up to 200?

Anyway, I installed GYRE 3.0 on my Mac and applied it to a FGONG-file of solar model S (see attachment + input list file). First, I wanted to compare the adiabatic outcome from GYRE with the one from ADIPLS. For modes of low harmonic degree it seems to work fine. However, when I increase the harmonic degree, e.g. l>40, I still get a result, but the radial grid is no longer monotonic increasing near the center and shows additional grid points that are negative. The eigenfunctions near the center also look not meaningful; near the surface, they seem to be okay.

I played around with different choices for the shooting and the reconstruction grid, as well as the numerical solver, but this did not helped. Do you have any idea how I can solve this problem?

Thank you very much in advance, pwave.

Hi 'pwave' --

GYRE should have no problem calculating high-degree modes; just yesterday, I did a run on model S with l = 2000.

From running with your example input files, I can't see any obvious problems. Could you post one of the eigenfunction output files, plus any plots you might have of the eigenfunctions which demonstrate the problem?

cheers,

Rich

Hello Rich,

I found the problem. I used the 'TXT' format for the output files and found out that there appear floating point numbers like 0.6673297045453201-100 and -0.1283653556399903E-99, i.e. the "E" is missing for numbers with exponents with 3 digits. This lead to erroneous representations of my data after reading the files with matlab. Now, I use the "HDF" format for the output files to omit this problem.

Thank you for your help, pwave.

pwave wrote:Hello Rich,

I found the problem. I used the 'TXT' format for the output files and found out that there appear floating point numbers like 0.6673297045453201-100 and -0.1283653556399903E-99, i.e. the "E" is missing for numbers with exponents with 3 digits. This lead to erroneous representations of my data after reading the files with matlab. Now, I use the "HDF" format for the output files to omit this problem.

Thank you for your help, pwave.

And thank you, for figuring out what the problem was. I'm going to fix this behavior in the next release of GYRE.

Best wishes,

Rich