Re: [mesa-users] secular modes with GYRE

[RICHARD H D T. <tow...@as...>]

Hi Alfred —

There already is some support in GYRE for this kind of calulation. In the &osc namelist group, there is an (undocumented) parameter ‘time_factor’ which allows one to control the assumed time dependence of modes. The two options are ‘OSC’, which has time dependence proportional to exp(-i omega t), and ‘EXP’, which has time dependence proportional to exp(-omega t). The ‘EXP’ option in effect allows one to scan along the imaginary frequency axis, which I use when searching for dynamically unstable modes.

The more general case of scanning along an arbitrary line in the complex plane isn’t supported by GYRE, but there are some (experimental) executables that ship with GYRE, which provide this sort of functionality. In particular, the file gyre/src/frontend/gyre_map.fpp allows a 2-D sampling of the discriminant D(omega) over the complex plane, which is a robust way to find very non-adiabatic modes. Attached below for your amusement is a map showing the real(D)=0 (blue) and imag(D)=0 (red) contours for radial modes in a massive star, as calculated using gyre_map. The crossings of these contours correspond to the star’s modes. Note the almost-complex-conjugate pairs of modes — we’re in the NAR limit here!

To build gyre_map, all you need do is download the most recent release candidate of GYRE (5.0rc3), edit gyre/src/build/Makefile to add gyre_map to the definitions of the TARGETS variable, and then build as usual; gyre_map will then be placed in gyre/bin. An example of gyre_map in action is given in the gyre/test/map directory; just do ‘./gyre_map gyre_map.in’ to run the example. The resulting map.h5 file can be plotted as a contour map using the plot_map.py python script in the same directory.

Since this is a very GYRE/pulsation specific topic, perhaps we should take further discussion to the GYRE forums (unless people on the MESA list want to hear about very non-adiabatic modes and the numerical challenges they present?).

cheers,

Rich



> On Apr 16, 2017, at 5:13 AM, al...@ga... wrote:
>
>
> For selected phases of a star's evolution I am frequently interested
> to learn about the prevailing secular (in-)stability situation. GYRE
> is a tool that has all components to answer such questions. In
> contrast to pulsation modes it is, however, necessary to be able to
> perform frequency scans along the imaginary axis [assuming the
> temporal behavior to be proportional to exp(i omega t)] on the
> frequency plane. As far as I have figured out, only scans along the
> real frequency axis are currently implemented.
>
> Being able to perform frequency scans along straight lines that are
> arbitrarily oriented in the *complex* frequency plane comes in handy
> also when dealing with strongly nonadiabatic pulsations; i.e. when
> eigenfrequencies lie far off the real frequency axis.
>
> So, here is my plea: Would it be possible, in a future release of
> GYRE, to be able to specify a
> freq_min and a freq_max, both being complex numbers, as lower and
> upper bounds for frequency scans.
>
> Thank you for considering the above as an item on the GYRE wishlist,
> Alfred
>
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users