Line-of-sight acceleration in compact binaries with higher harmonics and eccentricity

This paper presents a robust framework for incorporating line-of-sight acceleration effects into state-of-the-art gravitational-wave waveform models with higher harmonics and eccentricity, revealing that inconsistent treatment of these effects can bias results while finding no substantial evidence for such acceleration in analyzed LIGO-Virgo events.

The Gist

Imagine two heavy objects, like black holes or neutron stars, dancing around each other in space. As they spiral closer and eventually crash together, they send out ripples in the fabric of space-time called gravitational waves. Scientists on Earth catch these ripples with giant detectors (like LIGO and Virgo) to learn about the objects and how they formed.

Usually, scientists assume these dancing pairs are floating in a quiet, empty void. But what if they aren't? What if they are in a crowded neighborhood, like a busy city center or a dense star cluster? In these crowded places, other massive objects nearby might tug on the pair, causing their whole dance floor to speed up or slow down as they move toward us. This is called Line-of-Sight Acceleration (LoSA).

This paper is about building a better "translator" to hear if that tug-of-war is happening.

The Problem: The Old Translator Was Too Simple

Think of the gravitational wave signal like a song.

• The Old Way: Previous models tried to understand this song by only listening to the main melody (the dominant "quadrupole" note). They also assumed the song was perfectly smooth and circular.
• The Issue: Real cosmic songs are complex. They have harmonics (overtones), and sometimes the dancers are moving in oval paths (eccentricity) rather than perfect circles. If you only listen to the main melody and ignore the harmonics, or if you try to apply a "speed-up" correction meant for the main melody to the overtones incorrectly, you get a distorted understanding of the song. You might think the dancers are speeding up because of a neighbor's tug, when actually, you just didn't listen to the whole band correctly.

The Solution: A New, High-Fidelity Translator

The authors of this paper built a new, sophisticated model that listens to every note in the song, not just the main one.

1. Harmonics: They made sure that if the whole system is being accelerated, the correction is applied correctly to the main note and all the higher-pitched overtones.
2. Eccentricity: They updated the model to handle "oval" dances, not just perfect circles.
3. The Mechanism: They realized that acceleration acts like a time-delay. Imagine the dancers are on a moving walkway. If the walkway speeds up, the time it takes for their "song" to reach you changes in a specific way. The authors figured out exactly how to calculate this time delay for every single note in the song.

What They Found: The "Crowded Neighborhood" Test

The researchers took this new, high-fidelity translator and tested it in two ways:

1. The Simulation Test (The "Fake" Signal)
They created fake gravitational wave signals that did have this acceleration effect.

• Result: When they used the old, simple models (ignoring harmonics), the results were blurry. They couldn't tell exactly how strong the acceleration was. Sometimes, they even got the wrong answer about how far away the dancers were.
• Result: When they used their new model, they could hear the acceleration clearly. However, they also found that if the dancers were moving in very oval paths (high eccentricity), it became harder to hear the acceleration because the "ovalness" of the dance mimicked the "speeding up" effect. It's like trying to hear a car engine revving while it's also driving over a bumpy road; the two effects get mixed up.

2. The Real-World Test (The "Real" Signals)
They took real data from three famous cosmic crashes observed by LIGO and Virgo (GW190814, GW200105, and GW190728) and ran them through their new model.

• The Verdict: They found no strong evidence that these specific events were being tugged on by a nearby neighbor. The data looked like the dancers were in a quiet void, not a crowded city.
• A Correction to Past Claims: There was a previous study that claimed to find acceleration in one of these events (GW190814). The authors of this paper showed that the previous claim likely happened because they used the "simple translator" (ignoring the harmonics). When they re-analyzed that same event with their new, correct method, the evidence for acceleration disappeared.

The Bottom Line

This paper doesn't say that acceleration never happens in the universe. Instead, it says: "If you want to find it, you need to listen to the whole orchestra, not just the lead singer."

They have provided a robust, accurate tool for future searches. As our detectors get better and we can hear these cosmic songs for longer, this new tool will help us determine if compact binaries are forming in quiet isolation or in the chaotic, crowded environments of active galactic nuclei and star clusters. For now, though, the specific events they checked didn't show signs of this cosmic tug-of-war.

Technical Summary

Technical Summary: Line-of-sight acceleration in compact binaries with higher harmonics and eccentricity

Problem Statement
Compact binary mergers observed by the LIGO–Virgo–KAGRA (LVK) network offer a unique probe into astrophysical formation channels. While isolated binary evolution and dynamical formation (e.g., in dense stellar clusters or active galactic nuclei) produce overlapping mass and spin distributions, dynamical environments often induce line-of-sight acceleration (LoSA) of the binary's center of mass. This acceleration introduces a time-dependent Doppler modulation in the gravitational-wave (GW) signal, distinct from constant redshift or velocity.

Previous treatments of LoSA have primarily focused on quasi-circular binaries using the dominant quadrupole mode. However, existing literature highlights that:

1. Many dynamical scenarios may produce eccentric binaries or systems where higher-order harmonics are significant (e.g., high mass-ratio systems).
2. Inconsistent treatment of LoSA corrections across different waveform harmonics can lead to biased parameter inference.
3. Current analyses often rely on dominant-mode approximations, potentially missing signatures in systems where subdominant modes or eccentricity are non-negligible.

Methodology
The authors re-derive the leading-order LoSA effects and implement them consistently across advanced waveform models. The core theoretical framework relies on the integrated time delay induced by the LoS redshift, $z_\ell(t) = z_0 + \Gamma t$, where $\Gamma = a_\parallel/c$.

1. Theoretical Derivation:

   • Phase Correction: Using the stationary phase approximation (SPA), the authors derive the Fourier-domain phase correction $\Delta\Psi(f) = -\pi \Gamma f t(f)^2$. Crucially, they generalize this to higher-order modes $(\ell, m)$ by replacing the quadrupolar stationary time $t(f)$ with the mode-dependent stationary time $t_{\ell m}(f)$.
   • Amplitude Correction: They derive the corresponding fractional amplitude correction, $\delta A/A \propto \Gamma t_{\ell m}(f)$, arising from the mapping between source and detector frame variables.
   • Eccentricity: For eccentric binaries, the formulation utilizes the appropriate stationary times for each contributing harmonic within the quasi-Keplerian parameterization.

2. Implementation:

   • PhenomXPHM: The corrections are implemented in the frequency-domain inspiral–merger–ringdown model for precessing binaries with higher-order modes. The correction is applied mode-by-mode before reconstructing the inertial frame polarizations.
   • pyEFPE: The corrections are implemented in the precessing-eccentric inspiral model (pyEFPE), modifying the stationary-phase quantities ($t^{SPA}_i$, $T^{SPA}_i$, $\psi^{SPA}_i$) for each harmonic.

3. Validation and Analysis:

   • Injection Studies: The authors perform Bayesian inference on zero-noise injections of GW190814-like (high mass-ratio) and GW200105-like (eccentric) signals. They compare recovery using models with and without higher harmonics (PhenomXPHM vs. PhenomXP) and with and without eccentricity (pyEFPE vs. PhenomXP).
   • Real Event Analysis: They analyze three LVK events from the O3 run: GW190814, GW200105, and GW190728, using the updated waveform models.

Key Contributions

• Unified Framework: The paper provides a closed-form, unified framework for applying LoSA corrections to both higher-order harmonics and eccentric waveforms. This avoids the cumbersome, mode-by-mode frequency mapping required by previous propagation methods.
• Mode-Consistent Implementation: The authors demonstrate that applying LoSA corrections consistently to all contributing harmonics is essential. They show that applying a quadrupolar prescription uniformly to a higher-harmonic waveform (an inconsistent approach) introduces systematic biases.
• Quantification of Degeneracies: The study quantifies the degeneracy between LoSA and other parameters:
  • Higher Harmonics: Omitting higher harmonics in the recovery model does not necessarily shift the central value of the inferred acceleration but significantly broadens the posterior (reducing sensitivity) and biases the luminosity distance and source-frame mass estimates.
  • Eccentricity: There is a strong degeneracy between LoSA ($\Gamma$) and eccentricity ($e_{20}$). Both effects enter at similar post-Newtonian orders and can partially compensate for each other. Furthermore, high eccentricity reduces the time-to-coalescence, suppressing the net LoSA phase shift and degrading measurability.

Results

• Injection Studies:
  • For GW190814-like injections, recovery with PhenomXP (no higher modes) yields posteriors for $\Gamma$ that are 3–4 times broader than those from PhenomXPHM. It also introduces biases in luminosity distance and chirp mass.
  • For GW200105-like eccentric injections, the inclusion of eccentricity in the inference leads to a strong negative correlation between $\Gamma$ and $e_{20}$. High eccentricity ($e_{20} \approx 0.4$) significantly degrades the measurability of $\Gamma$, resulting in broader posteriors compared to quasi-circular cases.
• Real Event Analysis:
  • GW190814: The analysis finds no statistically significant evidence for LoSA. The inferred acceleration is consistent with zero within the posterior support (98.7% HPD contains $\Gamma=0$). The study resolves a previous tension in the literature (where some analyses claimed evidence for LoSA) by showing that inconsistent mode treatment and short data segments can produce spurious signals.
  • GW200105: No evidence for LoSA is found. The posterior is dominated by the strong degeneracy between acceleration and eccentricity.
  • GW190728: The data provides no informative constraint on $\Gamma$; the posterior remains prior-dominated.

Significance
The paper establishes a robust framework for searching for environmental signatures in GW signals using current and future detectors. By demonstrating that inconsistent treatment of higher harmonics can lead to biased conclusions, the authors argue that mode-by-mode implementation is critical for reliable population-level studies. While no substantial evidence for LoSA was found in the analyzed O3 events, the developed models enable more accurate probes of dynamical formation channels (such as AGN disks or dense clusters) as detector sensitivity improves and allows for the observation of longer inspirals. The work underscores that eccentricity, while a potential discriminator for dynamical formation, can complicate LoSA measurements due to strong parameter degeneracies.