Low-Frequency Gravitational Waves in Three-Dimensional Core-Collapse Supernova Models

This paper presents corrected low-frequency gravitational wave signals from three-dimensional core-collapse supernova models, addressing a computational error that underestimated neutrino-induced waveforms by a factor of $4\pi$ while confirming that the original conclusions remain valid and are further reinforced.

The Gist

Imagine a massive star, much heavier than our Sun, reaching the end of its life. It doesn't just fade away; it collapses in on itself and then explodes in a spectacular event called a core-collapse supernova. For a long time, scientists have been listening to these explosions using "ears" called gravitational wave detectors. However, they have mostly been tuning their ears to high-pitched sounds (high frequencies), like the squeaks and squeals of the star's core vibrating.

This paper, however, is about listening to the low, rumbling bass of the explosion. The authors are using supercomputer simulations to predict what these low-frequency "bass notes" sound like and how we might catch them in the future.

Here is a breakdown of their findings using simple analogies:

1. The Two "Musicians" in the Explosion

The paper explains that the low-frequency rumble comes from two different sources, acting like two musicians playing in an orchestra:

• The Fluid Musician (The Star's Matter): As the star explodes, chunks of hot gas and matter are thrown out in all directions. If this explosion is perfectly round, it's silent. But if it's lopsided (like throwing a ball that wobbles), it creates a gravitational wave.
• The Neutrino Musician (The Ghost Particles): Stars also shoot out trillions of tiny, ghost-like particles called neutrinos. Usually, we think of these as invisible. But the paper shows that if these particles are shot out unevenly (more to the left than the right), they also create a gravitational wave.

The Big Surprise: The authors found that for the low-frequency rumble, the Neutrino Musician is actually louder than the Fluid Musician. Even though the neutrinos are only slightly uneven in their direction, they create a bigger "bass note" than the churning matter does.

2. The "Ramp-Up" Analogy

The paper focuses on a specific type of signal called "memory." Imagine a car accelerating from a stop.

• High-frequency waves are like the engine revving up and down quickly (squeaks and rattles).
• Low-frequency memory is like the car slowly speeding up and then staying at a steady speed. The "memory" is the permanent change in the road's surface after the car passes.

The authors discovered that this "slow speed-up" (the ramp-up to the memory) follows a very predictable pattern, like a smooth hill. They found they could describe this hill with a simple mathematical curve (a logistic function). This is important because it means we can build a "template" or a "mold" of what this signal should look like, making it easier to find in the noise later.

3. The "Shape" of the Signal

The team ran three different simulations using stars of different sizes (9.6, 15, and 25 times the mass of our Sun).

• The Small Star (9.6 solar masses): This explosion was very round and quiet. The "bass note" was very faint, almost like a whisper.
• The Big Stars (15 and 25 solar masses): These explosions were more chaotic and lopsided. They produced much louder, stronger bass notes.

They also looked at the signal from every possible angle (like listening to a speaker from the front, side, or back). They found that while the loudness changes depending on where you stand, the shape of the low-frequency signal remains consistent.

4. Can We Hear It? (The Detection Challenge)

The authors tested if current detectors (like LIGO) could hear this low rumble.

• The Problem: Current detectors are like ears that are very good at hearing high-pitched squeaks but are "deaf" to very low rumbles. They have a "noise floor" that drowns out these low frequencies.
• The Solution: The paper suggests that while we might not hear the full "memory" (the final steady state) with current ground-based detectors, we might be able to hear the ramp-up (the part where the signal is building up) if the event happens close by (like in our own galaxy).
• Future Ears: The paper highlights that future space-based detectors (like LISA) and next-generation ground detectors (like the Einstein Telescope) will have much better "ears" for these low frequencies. They might be able to hear the entire signal clearly.

5. The "Ghost" in the Machine

In a specific test, the researchers tried to reconstruct the signal using real data from a detector. They found that the current tools used to find these explosions (which look for high-pitched, chaotic sounds) completely missed the low-frequency "neutrino" part of the signal. It was as if the detector was looking for a violin solo but the neutrino signal was a cello playing in a different room.

Summary

This paper tells us that when a massive star explodes, it creates a deep, low-frequency gravitational wave "bass note" that is mostly caused by the uneven shooting of ghost-like neutrino particles. While our current listening equipment is a bit deaf to these low notes, the signal has a predictable shape that we can use to build better "molds" for future searches. As our listening technology improves, we will finally be able to hear this deep rumble, giving us a new way to understand the heart of a supernova.

Technical Summary

Technical Summary: Low-Frequency Gravitational Waves in Three-Dimensional Core-Collapse Supernova Models

Problem Statement
While gravitational wave (GW) astronomy from core-collapse supernovae (CCSNe) has traditionally focused on high-frequency emission ($\gtrsim 500$ Hz) associated with proto-neutron star (PNS) oscillations, there is a growing recognition of the importance of the low-frequency band ($\lesssim 250$ Hz). This paper addresses the detection prospects and characteristics of low-frequency GW signals, specifically focusing on the "linear gravitational wave memory" (the 0-Hz limit) and the ramp-up phase leading to it. The authors investigate signals sourced from two distinct mechanisms: the anisotropic motion of stellar fluid and the anisotropic emission of neutrinos. A critical gap addressed is the lack of detailed modeling for the neutrino-sourced component in three-dimensional (3D) simulations and its contribution to the total low-frequency signal.

Methodology
The study utilizes three state-of-the-art 3D CCSN models generated with the Chimera supernova code, corresponding to progenitor masses of 9.6 $M_\odot$ (D9.6-3D), 15 $M_\odot$ (D15-3D), and 25 $M_\odot$ (D25-3D). The models are non-rotating, with varying metallicities.

1. Fluid-Sourced GWs: The authors extract gravitational waveforms from fluid motions for 2,664 different observer orientations (5-degree resolution in azimuthal and polar angles). These are derived from the transverse-traceless (TT) projection of the second time derivative of the mass quadrupole moment.
2. Neutrino-Sourced GWs: The authors provide a detailed derivation of GWs sourced from anisotropic neutrino emission, following the methodology of Epstein and extending the work of Müller and Janka. They treat neutrinos as massless fermions transported via a ray-by-ray approximation with flux-limited diffusion. The stress-energy tensor is constructed based on the directional energy loss rate ($dL/d\Omega$) and angular distribution. The resulting strain is calculated by integrating the neutrino luminosity anisotropy over time and solid angle, projected into the TT gauge.
3. Signal Analysis:
   • Templatability: The low-frequency ramp-up (below 100 Hz) is fitted using a logistic function, $f(t) = L / (1 + e^{-k(t-t_0)})$, to characterize the signal for matched filtering. Parameters $k$ (ramp-up frequency) and $L$ (amplitude/memory) are analyzed across all observer orientations.
   • Reconstruction: The authors inject a D15-3D waveform (fluid, neutrino, and total) into real interferometric data from the LIGO O3 run and attempt reconstruction using the coherent WaveBurst (cWB) pipeline.
   • Detection Prospects: Signal-to-noise ratios (SNR) are calculated for current detectors (aLIGO, aVirgo, KAGRA) and future detectors (LISA, Cosmic Explorer, Einstein Telescope), assuming a source distance of 1 kpc. To assess LISA detectability, signals are extended with a half-cosine tail to simulate long-duration memory.

Key Contributions and Results

• Neutrino Anisotropy Derivation: The paper presents a rigorous derivation of the GW strain sourced specifically from neutrino anisotropy in 3D models, calculating contributions from electron, anti-electron, heavy, and anti-heavy neutrino species.
• Dominance of Neutrino Signal: The analysis reveals that the low-frequency portion of the total GW signal is dominated by the neutrino-sourced component. While the fluid-sourced signal amplitude is comparable in some cases, the neutrino anisotropy drives the low-frequency ramp-up and the final memory amplitude.
• Model Characteristics:
  • D9.6-3D: Represents a "complete" signal where the explosion has finished and accretion has diminished. The GW strain reaches a non-zero memory value. The anisotropy is low ($<5\%$) and evolves slowly, resulting in a low-amplitude, low-frequency signal.
  • D15-3D and D25-3D: These models are still evolving at the end of the simulation. The authors assume the signals continue to develop according to their best-fit logistic functions for detection studies. These models exhibit higher amplitudes and more complex anisotropy evolution.
• Templating: The low-frequency signals can be effectively fitted with a logistic function. The ramp-up frequency $k$ is found to be in the range of roughly 20–60 Hz (e.g., $42.15 \pm 2.03$ Hz for the D9.6-3D fluid), which falls within the detectable band of current ground-based detectors.
• Reconstruction in Real Data: Using cWB on real interferometric noise, the authors successfully reconstructed the fluid-sourced signal (identifying g/f-mode features). However, the neutrino-sourced component was not distinguishable in the reconstruction, nor was it visible in the total signal reconstruction. This suggests that for current burst searches, the neutrino contribution may not need to be explicitly modeled to detect the event, though it dominates the low-frequency physics.
• Detection Prospects:
  • Ground-Based: Matched filtering combined with linear predictive filtering predicts memory detection for Galactic events (e.g., D15-3D and D25-3D) in current detectors.
  • Space-Based (LISA): The study highlights LISA as a promising candidate for memory detection, provided the memory signal lasts longer than a few seconds. With a 10,000-second tail extension, the models show detectability out to the center of the Galaxy for D15-3D and D25-3D.
  • Future Ground-Based: Cosmic Explorer and the Einstein Telescope show significantly higher SNRs due to improved low-frequency sensitivity.

Significance and Claims
The authors claim this work provides the first presentation of GWs sourced from neutrino luminosity anisotropy in these specific Chimera 3D models. The primary significance lies in:

1. Establishing the Neutrino Contribution: Demonstrating that neutrino anisotropy is the dominant source of low-frequency GWs and the linear memory in CCSNe.
2. Template Development: Providing a parameterized template (logistic function) for the low-frequency ramp-up, which facilitates matched filtering searches in the 20–60 Hz band.
3. Detection Strategy: Arguing that while current burst algorithms (cWB) may not isolate the neutrino component, future strategies involving matched filtering and multi-messenger approaches (using neutrino detections to trigger low-frequency GW searches) are viable.
4. Future Outlook: The paper concludes that the detection of low-frequency GWs from CCSNe is a promising component of future multimessenger astronomy, particularly with next-generation detectors (LISA, CE, ET) that can access the memory signal and the ramp-up phase.

The authors remain modest regarding immediate detection, noting that the D15-3D and D25-3D signals are not yet saturated and that the neutrino component is currently undetectable by burst pipelines in real noise. They position this work as a foundational step for future parameter estimation and detection studies.