Gravitational waves from three-dimensional core-collapse supernova models: The impact of moderate progenitor rotation

TL;DR

This study models gravitational wave emissions from 3D supernova simulations with varying progenitor rotation, revealing how rotation influences GW signal characteristics and correlates with SASI activity and shock dynamics.

Contribution

First detailed 3D supernova GW predictions including moderate progenitor rotation effects, highlighting the impact of SASI and shock behavior on GW signals.

Findings

GW signals are stochastic with deterministic low- and high-frequency components.

Rotation affects GW amplitude and SASI activity, with faster rotation producing stronger signals.

GW frequency peaks shift during explosion, indicating shock expansion or contraction.

Abstract

We present predictions for the gravitational-wave (GW) emission of three-dimensional supernova (SN) simulations performed for a 15 solar-mass progenitor with the Prometheus-Vertex code using energy-dependent, three-flavor neutrino transport. The progenitor adopted from stellar evolution calculations including magnetic fields had a fairly low specific angular momentum (j_Fe <~ 10^{15} cm^2/s) in the iron core (central angular velocity ~0.2 rad/s), which we compared to simulations without rotation and with artificially enhanced rotation (j_Fe <~ 2*10^{16} cm^2/s; central angular velocity ~0.5 rad/s). Our results confirm that the time-domain GW signals of SNe are stochastic, but possess deterministic components with characteristic patterns at low frequencies (<~200 Hz), caused by mass motions due to the standing accretion shock instability (SASI), and at high frequencies, associated with gravity-mode oscillations in the surface layer of the proto-neutron star (PNS). Non-radial mass motions in the post-shock layer as well as PNS convection are important triggers of GW emission, whose amplitude scales with the power of the hydrodynamic flows. There is no monotonic increase of the GW amplitude with rotation, but a clear correlation with the strength of SASI activity. Our slowly rotating model is a fainter GW emitter than the non-rotating model because of weaker SASI activity and damped convection in the post-shock layer and PNS. In contrast, the faster rotating model exhibits a powerful SASI spiral mode during its transition to explosion, producing the highest GW amplitudes with a distinctive drift of the low-frequency emission peak from ~80-100 Hz to ~40-50 Hz. This migration signifies shock expansion, whereas non-exploding models are discriminated by the opposite trend.