On the nature of oscillating modes of proto-neutron stars

The paper On the nature of oscillating modes of proto-neutron stars by Dimitra Tseneklidou et al. presents a novel scheme to classify the oscillation modes of a newly born proto-neutron star (PNS) surrounded by a stalled accretion shock in core-collapse supernovae (CCSNe). The classification is physically motivated, based on the energy of the restoring forces of the modes.

The authors investigate the nature of the modes by considering the different regions of the CCSN from which they originate. They apply their scheme to a set of 28 non-rotating 1D, 2D, and 3D CCSN simulations performed with two different numerical codes (Aenus-ALCAR and CoCoNuT) and using different progenitors and equations of state.

The key findings are:

1. Four distinct mode families: The study identifies two families of f- and p-modes associated with the PNS and the shock, respectively, and two families of g-modes originating from the two convectively stable regions—one in the PNS core and one near its surface. Specifically:

• f+p-PNS modes: originating in the PNS interior

• f+p-shock modes: arising from the cavity enclosed by the PNS surface and the shock

• g-core modes: stemming from the stable PNS core region

• g-surface modes: stemming from the stably stratified region at the PNS surface

2. Dominant emission mode: The authors settle the issue of the nature of the dominant high-frequency GW emission mode (the High Frequency Feature, HFF). They show that the HFF is the f-mode of the PNS, associated with the presence of a strong density gradient at the PNS surface. Given continuous accretion, this mode can also be seen as an interface mode between the PNS and the post-shock region.

3. Avoided crossings: The study clarifies the origin of the break in the frequency evolution of the HFF at about 0.4 s post-bounce, observed in many numerical simulations. This break is interpreted as a prominent avoided crossing between the PNS f-mode and the first core g-mode.

4. Methodology: The classification procedure is based on:

• Computing the energy of the modes via the work done by restoring forces (compression, buoyancy, free surface, gravitational potential)

• Using an energy fraction F = Mf+p/M2 to distinguish f+p-modes from g-modes

• Classifying modes by their spatial localization (core vs. surface for g-modes; PNS interior vs. post-shock region for f+p-modes)

• Applying a RANSAC-based outlier elimination filter

5. Validation: The classification is confirmed through:

• Restricted-domain calculations (core, PNS, extended PNS domains)

• Radial energy distribution analysis using the Bhattacharyya coefficient

• Simple analytic estimates for each mode family that reproduce the trends of numerically computed modes

• Propagation diagrams comparing mode frequencies with Brunt-Väisälä and Lamb frequencies

6. Importance of full domain: The authors emphasize that taking the full domain of the system into account, i.e. up to the location of the shock, is crucial to characterize the complete mode spectrum. Using only the PNS surface as the outer boundary leads to misclassification of surface g-modes and makes it impossible to identify shock f+p-modes.

7. Automatic classification: The procedure is automatic, performs consistently for all 28 models considered, and paves the way to systematic studies of the dependence of the mode frequencies on the PNS properties, with direct application to parameter estimation from future observations of gravitational wave signals.

The paper concludes that this work will enable systematic studies of PNS mode dependencies on key physical parameters, focusing on modes prominently observed in simulated GW spectrograms (f-PNS, g-core, and SASI modes), and aims to advance quasi-universal relations between mode frequencies and PNS properties for interpreting future GW observations.

Improvements to AI systems:

1. Physics-informed mode classification for GW data analysis: Train an AI to automatically classify oscillation modes in gravitational wave (GW) spectrograms from core-collapse supernovae using the energy-fraction criterion (F = Mf+p/M2) and spatial localization (core vs. surface, PNS vs. post-shock). The improved system can distinguish f+p-PNS, f+p-shock, g-core, and g-surface modes in real or simulated GW signals, enabling rapid identification of the dominant emission mode (HFF) without manual inspection.

2. Avoided-crossing detection and frequency evolution prediction: Implement an AI that learns the pattern of avoided crossings between PNS f-modes and core g-modes (e.g., the break at 0.4 s post-bounce). The system can predict when and how mode frequencies will cross or repel based on PNS properties (mass, radius, accretion rate), improving parameter estimation from observed GW frequency tracks.

3. Domain-aware mode extraction: Build an AI that incorporates the full computational domain (up to the stalled shock) as a required input, rather than truncating at the PNS surface. This prevents misclassification of surface g-modes and enables identification of shock f+p-modes. The system can flag incomplete-domain data and correct for boundary-induced errors in mode frequency estimates.

4. Automatic multi-model consistency checking: Develop an AI that applies the RANSAC-based outlier filter and restricted-domain validation (core, PNS, extended PNS) to any new CCSN simulation, automatically verifying that mode classifications are robust across different codes (Aenus-ALCAR, CoCoNuT), progenitors, and equations of state. This reduces human bias and speeds up systematic surveys.

5. Quasi-universal relation discovery: Use the classification scheme as a feature extractor to train a regression model that maps PNS properties (e.g., compactness, entropy gradients, Brunt-Väisälä frequency profiles) to mode frequencies for each family. The improved AI can then predict GW spectra for unseen progenitors or EoS, and invert observed frequencies to infer PNS properties—directly aiding future GW observatory data interpretation.

6. Propagation-diagram-guided mode identification: Integrate Brunt-Väisälä and Lamb frequency profiles as auxiliary inputs to an AI classifier, allowing it to distinguish modes based on where their frequencies intersect these diagnostic curves. The system can automatically generate propagation diagrams and use them to validate or correct mode assignments in noisy or low-resolution data.

7. Real-time GW signal interpretation: Deploy a trained model that takes a time-frequency spectrogram of a detected CCSN GW signal (e.g., from LIGO/Virgo/KAGRA) and outputs: (a) which mode families are present, (b) the HFF frequency and its evolution, (c) the time of avoided crossing, and (d) inferred PNS properties (mass, radius, accretion rate) with uncertainties. This enables immediate astrophysical interpretation during alert follow-up.

Newborn proto-neutron stars (PNS) are potential targets of gravitational wave asteroseismology, the study of the (inner) structure of stars via their oscillation modes. To prepare for the eventual detection of such modes by current and future gravitational wave observatories, theoretical studies have obtained the possible spectrum of oscillations. However, there has been disagreement when it comes to identifying and classifying specific modes according to the physical mechanism that excites them. In this paper, we present a novel scheme to classify the oscillation modes of a newly born PNS surrounded by a stalled accretion shock in core-collapse supernovae (CCSNe). Our classification is physically motivated, as it is based on the energy of the restoring forces of the mode. We investigate the nature of the modes by considering the different regions of the CCSN that they stem from. We apply this scheme to a set of 28 non-rotating 1D, 2D and 3D CCSN simulations performed with two different numerical codes and using different progenitors and equations of state. In that way, we find that there are different families of f-, p- and g-modes in the system, living in different areas. More specifically, we identify two families of f- and p-modes associated with the PNS and the shock, respectively, and two families of g-modes originating from the two convectively stable regions, one in the PNS core and one near its surface. We also find that the dominant high-frequency emission mode is the f-mode of the PNS, associated with the strong density gradient at its surface. Our classification procedure is automatic, performs consistently for all the models considered, and paves the way to systematic studies of the dependence of the mode frequencies on the PNS properties, with direct application to parameter estimation from future observations of gravitational wave signals.