The influence of model parameters on the prediction of gravitational wave signals from stellar core collapse

We present the gravitational wave (GW) analysis of an extensive series of 3D MHD core-collapse simulations. Our 25 models are launched from a 15 solar mass progenitor, a spherically symmetric effective general relativistic potential, the Lattimer-Swesty or the Shen equation of state (EoS), and a neutrino parametrisation scheme which is accurate until about 5ms postbounce. For 3 representative models, we also include long-term neutrino physics by means of a leakage scheme. Non- or only slowly rotating models show GW emission due to prompt and proto-neutron star convection, allowing the distinction between the two different nuclear EoS. For moderately or fast rotation rates models, we find, in agreement with recent results, only a type I GW signature at core bounce. Models which are set up with an initial central angular velocity of >~ 2pi rad/s emit GWs due to the low T/|W| dynamical instability during the postbounce phase. Weak B-fields do not notably influence the dynamical evolution of the core and thus the GW emission. However, for strong initial poloidal B-fields (~1e12 G),flux-freezing and field winding leads to conditions where P_{mag}/P_{mat} ~ 1, causing the onset of a jet-like supernova explosion and hence the emission of a type IV GW signal. In contradiction to axisymmetric simulations, we find evidence that nonaxisymmetric fluid modes can counteract or even suppress jet formation for models with strong initial toroidal B-fields. We point out that the inclusion of the deleptonisation during the postbounce phase is an indispensable issue for the quantitative prediction of GWs from core-collapse supernovae, as it can alter the GW amplitude up to a factor of 10 compared to a pure hydrodynamical treatment.