Visibility of unstable oscillation modes in a rapidly rotating B star
Space missions like CoRoT and Kepler have provided numerous new observations of stellar oscillations in a multitude of stars by high precision photometry. This work compares the observed rich oscillation spectrum of the rapidly rotating B3 IV star HD 43317 with the first results obtained by a new method to calculate unstable oscillation modes in rapidly rotating stars in order to see whether some of the observed modes can be identified. The new numerical method consists of two parts. We first search for modes resonant with a prescribed forcing symmetry by moving through relevant regions of complex frequency space and monitoring any increase of the stellar response to the applied forcing and zooming in onto the resonance. These resonant non-adiabatic 2D-solutions are then fed into a 2D relaxation code with the same equations but without forcing terms. The complex oscillation frequency used in the forcing is now no longer prescribed, but added as an extra unknown. The corresponding free oscillation mode is usually obtained after a few ($<10$) iterations with only minor adjustment of the complex oscillation frequency. To compare with the observed light variations we calculate the `visibility' of the found unstable oscillation modes, taking into account the cancellation of the various parts of the radiating oscillating stellar surface as seen by the observer. The frequencies of unstable axisymmetric g-modes, which have the highest visibility, appear to nearly coincide with the observed largest amplitude photometric variations of HD 43317, making an identification of the latter oscillations as $m$=0 modes plausible. The identification of $m$=1 g-modes is less straightforward, while many of the unstable even $m$=2 g-modes may correspond to observed weaker photometric variations.