Hello,
I was analyzing the normalized growth rates on a suite of 13 WD models. I'm particularly interested in the g-mode with radial order 2 (g2). For all 13 models, gyre reports a positive normalized growth rate. I verified this by numerically integrating the work integrals for the g2 mode for all 13 models (following the Stellingworth definition for normalized growth rate). However, I'm confused because gyre is reporting both positive and negative values for Im(omega) for the g2 mode. From the documentation and prior posts, I expected the Im(omega) values to all be negative for g2 if all of my models are showing a positive normalized growth rate for g2. The magnitude of Im(omega) is relatively small, so I'm wondering if this is a numerical issue?
I've attached two figures that may help explain my confusion. One shows the nonadiabatic normalized growth rates with respect to radial order. The other is the imaginary angular frequency values reported, zoomed in for g2. (The y values have been multiplied by 10^{11} for formatting. So a 0.01 value is 10^{-13}).
I saw this behavior with both the 'Adiabatic' and 'Minmod' methods. I also ran some tests by increasing the n_freq value and decreasing the dx_max value. In general, the normalized growth rates were consistently positive for g2 regardless of method, n_freq, and dx_max resolution values. But the Im(omega) values continued to have show both positive and negative values for g2.
Have others ran into this before? Thank you in advance for any help or guidance!
Morgan