Extracting SASI signatures from Gravitational Waves of Core-Collapse Supernovae using the Hilbert-Huang Transform

TL;DR

Core-collapse supernovae (CCSNe) emit gravitational waves (GWs) and neutrinos, with SASI imprinting a characteristic low-frequency GW signature around $f \sim 100 \ \mathrm{Hz}$ that correlates with neutrino flux. The authors apply Hilbert-Huang Transform (HHT), combining ensemble empirical mode decomposition (EEMD) and Hilbert spectral analysis (HSA), to a set of 3D CCSN simulations to isolate the SASI contribution and track its instantaneous frequency. They assess detectability with the Einstein Telescope (ET) network, finding high efficiency (>$90\%$) out to $\sim 100$ kpc for favorable models and lower efficiency for weaker signals, demonstrating a practical path to probing the inner engine of CCSNe with multi-messenger observations. The study provides a robust, data-driven time-frequency analysis that can be extended to real data and future neutrino analyses.

Abstract

Core collapse supernovae are among the most energetic astrophysical events in the Universe. Despite huge efforts on understanding the main ingredients triggering such explosions, we still lack of compelling evidences for the precise mechanism driving those phenomena. They are expected to produce gravitational waves due to asymmetric mass motions in the collapsing core, and emit in the meanwhile neutrinos as a result of the interactions in their high-density environment. The combination of these two cosmic messengers can provide a unique probe to study the inner engine of these processes and unveil the explosion mechanism. Among the possible detectable signature, standing accretion shock instabilities (SASI) are particularly relevant in this context as they establish a direct connection between gravitational wave emission and the outcoming neutrino flux. In this work, Hilbert-Huang transform is applied to a selected sample of 3D numerical simulations, with the aim of identifying SASI contribution and extract its instantaneous frequency. The performance of the method is evaluated in the context of Einstein Telescope.

Figures (8)

• Figure 1: Taken from doi:10.1142/S1793536909000047. IMF computation via EMD process. Panel (a) is the input; panel (b) identifies local maxima (gray dots); panel (c) plots the upper envelope (upper gray dashed line) and low envelope (lower gray dashed line) and their mean (bold gray line); and panel (d) is the difference between the input and the mean of the envelopes. The input data, which is composed of high-frequency intermittent oscillations riding on the fundamental low-frequency part, results in a first IMF that is a mixture of both components (mode mixing).
• Figure 2: Kuroda et al. 2016 Kuroda_2016 plus polarization in time-frequency plane computed via short-time Fourier transform.
• Figure 3: List of IMFs obtained by applying EEMD to the Kuroda et al. 2016 Kuroda_2016 signal. The blue curve shows the original input signal, while the orange curve represents the corresponding mode strain. The final panel displays the residual component from the decomposition.
• Figure 4: HHT-based time-frequency representations of the first six IMFs extracted from the Kuroda et al. 2016 simulation Kuroda_2016. IF and IA are calculated using HSA, as described in Section \ref{['hht']}. To improve plot readability, the data have been downsampled to 200 Hz, following the approach in PhysRevD.104.084063.
• Figure 5: