pith. machine review for the scientific record. sign in

arxiv: 2604.20971 · v1 · submitted 2026-04-22 · 🌌 astro-ph.GA · astro-ph.HE· gr-qc

Recognition: unknown

Chaotic migration of LISA Extreme Mass Ratio Inspirals in a turbulent accretion disk: effect on waveform de-phasing

Mudit Garg , Lucio Mayer , Yinhao Wu , Yacine Ali-Ha\"imoud , Douglas N.C. Lin
Authors on Pith no claims yet

Pith reviewed 2026-05-09 23:14 UTC · model grok-4.3

classification 🌌 astro-ph.GA astro-ph.HEgr-qc
keywords extreme mass ratio inspiralsLISAgravitational wave dephasingaccretion disk turbulencechaotic migrationwaveform modelinggas torques
0
0 comments X

The pith

Turbulent gas torques on extreme mass ratio inspirals can produce observable dephasing in LISA gravitational wave signals when mean linear torques alone do not.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper examines how turbulence in galactic accretion disks affects the orbital evolution of extreme mass ratio inspirals that LISA will detect. It models the turbulent torque as random fluctuations around the expected linear torque from a smooth disk. For certain ranges of disk properties like high Eddington ratios and sufficient turbulence strength, these fluctuations lead to chaotic migration that accumulates enough phase shift to be seen by LISA. A reader would care because accurate waveform modeling is needed to extract source parameters and learn about the environments where these black hole binaries form. The work shows that ignoring turbulence underestimates the gas effects on the signals.

Core claim

The central claim is that when the turbulent torque is modeled as a Gaussian distribution around the linear torque T_lin with normalization C, the resulting chaotic migration of the EMRI produces a gas-induced dephasing Delta psi_gas that exceeds the detectability threshold of 8/SNR for f_Edd ≳ 0.3, C ≳ 300, h0 ≳ 0.03, and α ≳ 0.1 in the final four years of a golden circular EMRI with M=10^6 M_sun, q=5e-5 at z=0.276 and SNR=50, whereas considering only T_lin yields unobservable dephasings.

What carries the argument

The turbulent torque T_turb, represented as a Gaussian random variable centered on the linear torque T_lin with a normalization factor C drawn from hydrodynamical simulations, which introduces stochasticity that drives chaotic orbital migration and accumulates waveform phase shifts.

Load-bearing premise

The assumption that turbulent torques act as Gaussian fluctuations around the mean linear torque, with strength fixed from prior simulations, without self-consistently evolving the coupled disk-EMRI system.

What would settle it

Detection or non-detection of dephasing signals larger than 8/SNR in actual LISA EMRI events from high-accretion turbulent disks would confirm or refute the model predictions.

Figures

Figures reproduced from arXiv: 2604.20971 by Douglas N.C. Lin, Lucio Mayer, Mudit Garg, Yacine Ali-Ha\"imoud, Yinhao Wu.

Figure 1
Figure 1. Figure 1: We show ∆ψlin, the gas-induced dephasing due to the linear torque, as a function of the Eddington ratio (fEdd), the turbo-viscous coefficient (α), and the disk aspect ratio (h0). For each panel, we increase α vertically from 10−1.25 to 10−0.25 and h0 horizontally from 0.01 to 0.1, while keeping the Eddington ratio constant. Then we increase fEdd from left to right from 0.1 to 1. The colorbar shows the rang… view at source ↗
Figure 2
Figure 2. Figure 2: For the “golden” EMRI in Eq. (15), we show Λ that is defined in Eq. (17) to signify where the tail of the turbulent dephasing distribution (set to |∆ψlin| + 2ˆσturb) becomes observable in comparison to the linear dephasing ∆ψlin for a given {fEdd, C, α, h0}. We horizontally increase fEdd from 0.1 to 1 and vertically increase C from 102 to 103 panel-by-panel. And for each panel, we vertically increase α fro… view at source ↗
read the original abstract

Gravitational wave (GW) detector LISA will observe near-coalescence extreme mass ratio inspirals (EMRIs), which typically form in galactic central accretion disks. Gas torques on EMRI will alter its GW-driven inspiral trajectory from the vacuum expectation, leading to potentially LISA-observable GW dephasing ($\Delta\psi_{\rm gas}$). Most studies compute $\Delta\psi_{\rm gas}$ for a thin, laminar disk, with negligible flow turbulence, where the disk exerts a fairly well-understood linear torque ($T_{\rm lin}$). However, these disks must be turbulent due to magneto-rotational instability in the inner regions. Hence, we present a proof-of-concept general, agnostic prescription for the turbulent torque ($T_{\rm turb}$) acting on an EMRI by modeling it as a Gaussian distribution around $T_{\rm lin}$, based on recent advances from a global hydrodynamical (HD) study. We compute $\Delta\psi_{\rm gas}$ for the ``golden'' circular EMRI with total source mass $M=10^6~{\rm M}_\odot$ and mass ratio $q=5\times10^{-5}$ in its final four-year evolution at redshift $z=0.276$ and signal-to-noise ratio (SNR) $=50$ by varying Eddington ratio ${\rm f}_{\rm Edd}$, turbulence normalization $C$ ($=~360$ in the aforementioned HD study), disk aspect ratio $h_0$, and turbo-viscous coefficient $\alpha$ in a reasonable parameters space. We find that for ${\rm f}_{\rm Edd}\gtrsim0.3$, $C\gtrsim300$, $h_0\gtrsim0.03$, and $\alpha\gtrsim0.1$, gas-induced dephasings are unobservable if only considering $T_{\rm lin}$ but could become detectable ($\Delta\psi_{\rm gas}>8/$SNR) if EMRIs exhibit chaotic migration due to turbulent gas flow. Hence, this work motivates running MHD simulations of accretion disks with embedded LISA EMRIs in the early in-spiral phase over long enough timescales to understand the evolution of their orbital elements and the imprint of the turbulent environment on their gravitational waveforms.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 3 minor

Summary. The manuscript presents a proof-of-concept model in which the turbulent torque T_turb on a LISA EMRI is drawn from a Gaussian distribution centered on the linear torque T_lin, with width set by a normalization C drawn from a prior hydrodynamical study. For a fiducial circular EMRI (M=10^6 M_⊙, q=5×10^{-5}, z=0.276, SNR=50), the authors integrate the orbital evolution over the final four years and report that gas-induced dephasing Δψ_gas exceeds the detectability threshold 8/SNR for f_Edd ≳ 0.3, C ≳ 300, h0 ≳ 0.03 and α ≳ 0.1, whereas the same parameters yield undetectable dephasing when only T_lin is used. The work concludes by motivating self-consistent MHD simulations of disks with embedded EMRIs.

Significance. If the Gaussian turbulence prescription can be shown to produce a net random walk that survives orbital and viscous correlations, the result would indicate that disk turbulence can push otherwise undetectable gas effects into the LISA band, directly affecting waveform modeling and parameter estimation. The numerical integration itself is straightforward and the parameter survey is systematic; the paper also correctly flags the need for future coupled simulations. However, the claimed detectability window is defined by the same free parameters that set the torque variance, limiting the strength of the prediction.

major comments (3)
  1. [§2] §2 (Turbulent torque prescription): The Gaussian model T_turb ~ N(T_lin, C) is introduced without derivation from the MHD equations or direct validation against simulation data that include an embedded point mass; C is taken from a single prior global HD run that does not contain the EMRI, leaving open whether the quoted variance remains appropriate once the secondary perturbs the local flow.
  2. [§3] §3 (Numerical integration): The torque time series is sampled independently at each timestep, implicitly assuming white, uncorrelated fluctuations; the manuscript does not evolve the disk and EMRI self-consistently, so possible correlations on orbital or viscous timescales are not tested and could cause the random walk in semi-major axis (and thus Δψ_gas) to saturate rather than accumulate to detectable levels.
  3. [§4] §4 (Results and detectability): The statement that Δψ_gas > 8/SNR holds for f_Edd ≳ 0.3, C ≳ 300, etc., is obtained after scanning the four-dimensional parameter space; the threshold is therefore applied post hoc rather than as a blind, a-priori prediction, making the central claim dependent on the chosen values of the free parameters.
minor comments (3)
  1. [§3] The definition of Δψ_gas and the precise numerical implementation of the phase integral should be stated explicitly in the text (currently referenced only to prior work).
  2. [Figures] Figure captions and axis labels could more clearly indicate which curves correspond to the laminar (T_lin only) versus turbulent cases and whether error bands reflect Monte-Carlo realizations.
  3. [§3] A short convergence test with respect to timestep or number of turbulence realizations would strengthen the numerical results section.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the thoughtful and constructive report. We address each major comment below and have revised the manuscript to incorporate clarifications and additional caveats where appropriate. Our responses aim to strengthen the presentation of this proof-of-concept study while honestly acknowledging its limitations.

read point-by-point responses

  1. Referee: §2 (Turbulent torque prescription): The Gaussian model T_turb ~ N(T_lin, C) is introduced without derivation from the MHD equations or direct validation against simulation data that include an embedded point mass; C is taken from a single prior global HD run that does not contain the EMRI, leaving open whether the quoted variance remains appropriate once the secondary perturbs the local flow.

    Authors: We agree that the Gaussian prescription is a phenomenological model rather than a first-principles derivation from the MHD equations with an embedded secondary. The value of C is drawn from a prior global hydrodynamical simulation that did not include an EMRI, so the local flow perturbation by the secondary is not accounted for. This is an inherent limitation of the current approach. In the revised manuscript we have expanded the discussion in §2 to explicitly state these caveats, to clarify that the model is intended only as a proof-of-concept, and to reiterate the motivation for future self-consistent MHD simulations that include the embedded EMRI. revision: partial

  2. Referee: §3 (Numerical integration): The torque time series is sampled independently at each timestep, implicitly assuming white, uncorrelated fluctuations; the manuscript does not evolve the disk and EMRI self-consistently, so possible correlations on orbital or viscous timescales are not tested and could cause the random walk in semi-major axis (and thus Δψ_gas) to saturate rather than accumulate to detectable levels.

    Authors: The independent sampling of the torque at each timestep is indeed an assumption that corresponds to white-noise fluctuations. We recognize that real disk turbulence is expected to exhibit correlations on orbital and viscous timescales, which could in principle cause the random walk to saturate. Because the present work does not perform self-consistent disk+EMRI evolution, we cannot quantify the impact of such correlations. We have added a dedicated paragraph in the revised §3 and in the conclusions that highlights this assumption and its possible consequences, while emphasizing that the study is designed to motivate precisely the long-term, self-consistent simulations needed to test these effects. revision: partial

  3. Referee: §4 (Results and detectability): The statement that Δψ_gas > 8/SNR holds for f_Edd ≳ 0.3, C ≳ 300, etc., is obtained after scanning the four-dimensional parameter space; the threshold is therefore applied post hoc rather than as a blind, a-priori prediction, making the central claim dependent on the chosen values of the free parameters.

    Authors: We accept that the reported detectability thresholds emerge from a systematic scan of the (f_Edd, C, h0, α) parameter space rather than from a blind, a-priori forecast. The purpose of the survey is exploratory: to identify the regions of parameter space in which turbulent torques could push gas-induced dephasing above the LISA detection threshold. In the revised manuscript we have rephrased the abstract, §4, and the conclusions to present the results explicitly as an exploration of the viable parameter regime, to stress the dependence on the free parameters, and to avoid any implication of a definitive prediction. revision: yes

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper introduces an agnostic Gaussian model for T_turb centered on T_lin with free normalization C drawn from an external HD study, then performs a forward parameter sweep over f_Edd, C, h0, and α to compute Δψ_gas for a fixed EMRI. This is a model-based exploration of detectability thresholds rather than a closed derivation; the output dephasing scales with the input variance by design of the random-walk integration, but no step equates a claimed prediction to its own fitted inputs or reduces via self-citation to an unverified premise. The central result is conditional on the assumed model parameters and does not invoke uniqueness theorems or rename prior results. The derivation remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

4 free parameters · 2 axioms · 1 invented entities

The model rests on three main external inputs: the linear torque T_lin from prior analytic work, the turbulence normalization C taken from an earlier HD simulation, and the assumption that the disk remains thin and Keplerian while the EMRI migrates.

free parameters (4)
  • C (turbulence normalization)
    Width of the Gaussian turbulent torque; set to 360 from the cited HD study and then varied.
  • f_Edd (Eddington ratio)
    Controls gas density and therefore torque strength; scanned above 0.3.
  • h0 (disk aspect ratio)
    Affects torque magnitude; scanned above 0.03.
  • α (turbo-viscous coefficient)
    Additional viscosity parameter; scanned above 0.1.
axioms (2)
  • domain assumption The disk is turbulent due to magneto-rotational instability and the torque fluctuations can be modeled as a stationary Gaussian process around the mean linear torque.
    Invoked to justify the T_turb prescription without deriving the distribution from MHD.
  • domain assumption The EMRI remains on a circular orbit while the torque is applied over the final four years.
    Simplifies the waveform calculation to a single phase integral.
invented entities (1)
  • Gaussian turbulent torque T_turb no independent evidence
    purpose: To represent chaotic migration without running full MHD.
    New statistical entity introduced to stand in for unresolved turbulent fluctuations.

pith-pipeline@v0.9.0 · 5750 in / 1753 out tokens · 28956 ms · 2026-05-09T23:14:08.799606+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

47 extracted references · 46 canonical work pages · 2 internal anchors

  1. [1]

    2023, Living Reviews in Relativity, 26, 2, doi: 10.1007/s41114-022-00041-y

    Amaro-Seoane, P., Andrews, J., Arca Sedda, M., et al. 2023, Living Reviews in Relativity, 26, 2, doi: 10.1007/s41114-022-00041-y Arca Sedda, M., Amaro Seoane, P., & Chen, X. 2021, A&A, 652, A54, doi: 10.1051/0004-6361/202037785

    work page doi:10.1007/s41114-022-00041-y 2023

  2. [2]

    doi:10.1103/PhysRevD.89.104059, https://arxiv.org/abs/1404.7149 arXiv:1404.7149 [gr-qc]

    Barausse, E., Cardoso, V., & Pani, P. 2014, PhRvD, 89, 104059, doi: 10.1103/PhysRevD.89.104059

    work page doi:10.1103/physrevd.89.104059 2014

  3. [3]

    M., Petterson J

    Bardeen, J. M., & Petterson, J. A. 1975, ApJL, 195, L65, doi: 10.1086/181711

    work page doi:10.1086/181711 1975

  4. [4]

    Baruteau, C., & Lin, D. N. C. 2010, ApJ, 709, 759, doi: 10.1088/0004-637X/709/2/759

    work page doi:10.1088/0004-637x/709/2/759 2010

  5. [5]

    M., Mac Low, M.-M., McKernan, B., & Ford, K

    Bellovary, J. M., Mac Low, M.-M., McKernan, B., & Ford, K. E. S. 2016, ApJL, 819, L17, doi: 10.3847/2041-8205/819/2/L17

    work page doi:10.3847/2041-8205/819/2/l17 2016

  6. [6]

    Sathyaprakash, B. S. 2009, PhRvD, 80, 084043, doi: 10.1103/PhysRevD.80.084043 Cevallos Soto, A., & Zhu, Z. 2026, ApJ, 998, 209, doi: 10.3847/1538-4357/ae355c

    work page doi:10.1103/physrevd.80.084043 2009

  7. [7]

    2025, MNRAS, 540, 1998, doi: 10.1093/mnras/staf867

    Chen, Y.-X., Wu, Y., Li, Y.-P., et al. 2025, MNRAS, 540, 1998, doi: 10.1093/mnras/staf867

    work page doi:10.1093/mnras/staf867 2025

  8. [8]

    LISA Definition Study Report

    Colpi, M., Danzmann, K., Hewitson, M., et al. 2024, arXiv e-prints, arXiv:2402.07571, doi: 10.48550/arXiv.2402.07571 9

    work page internal anchor Pith review doi:10.48550/arxiv.2402.07571 2024

  9. [9]

    The Astrophysical Journal891(2), 154 (2020)

    Deng, H., Mayer, L., & Latter, H. 2020, ApJ, 891, 154, doi: 10.3847/1538-4357/ab77b2

    work page doi:10.3847/1538-4357/ab77b2 2020

  10. [10]

    2021, MNRAS, 501, 3540, doi: 10.1093/mnras/staa3976

    MacFadyen, A. 2021, MNRAS, 501, 3540, doi: 10.1093/mnras/staa3976

    work page doi:10.1093/mnras/staa3976 2021

  11. [11]

    J., Franchini, A., et al

    Derdzinski, A., Dittmann, A. J., Franchini, A., et al. 2025, arXiv e-prints, arXiv:2512.10893, doi: 10.48550/arXiv.2512.10893

    work page doi:10.48550/arxiv.2512.10893 2025

  12. [12]

    2025a, PhRvD, 111, 084006, doi: 10.1103/PhysRevD.111.084006

    Duque, F., Kejriwal, S., Sberna, L., Speri, L., & Gair, J. 2025a, PhRvD, 111, 084006, doi: 10.1103/PhysRevD.111.084006

    work page doi:10.1103/physrevd.111.084006

  13. [13]

    Duque, L

    Duque, F., Sberna, L., Spiers, A., & Vicente, R. 2025b, arXiv e-prints, arXiv:2510.02433, doi: 10.48550/arXiv.2510.02433

    work page doi:10.48550/arxiv.2510.02433

  14. [14]

    2024a, MNRAS, 532, 4060, doi: 10.1093/mnras/stae1764

    Garg, M., Derdzinski, A., Tiwari, S., Gair, J., & Mayer, L. 2024a, MNRAS, 532, 4060, doi: 10.1093/mnras/stae1764

    work page doi:10.1093/mnras/stae1764

  15. [15]

    2022, MNRAS, 517, 1339, doi: 10.1093/mnras/stac2711

    Mayer, L. 2022, MNRAS, 517, 1339, doi: 10.1093/mnras/stac2711

    work page doi:10.1093/mnras/stac2711 2022

  16. [16]

    2025, ApJ, 993, 145, doi: 10.3847/1538-4357/ae10b4

    Garg, M., Franchini, A., Lupi, A., Bonetti, M., & Mayer, L. 2025, ApJ, 993, 145, doi: 10.3847/1538-4357/ae10b4

    work page doi:10.3847/1538-4357/ae10b4 2025

  17. [17]

    2024b, MNRAS, 535, 3283, doi: 10.1093/mnras/stae2605

    Garg, M., Sberna, L., Speri, L., Duque, F., & Gair, J. 2024b, MNRAS, 535, 3283, doi: 10.1093/mnras/stae2605

    work page doi:10.1093/mnras/stae2605

  18. [18]

    Garg, M., Tiede, C., & D’Orazio, D. J. 2024c, MNRAS, 534, 3705, doi: 10.1093/mnras/stae2357

    work page doi:10.1093/mnras/stae2357

  19. [19]

    2024d, MNRAS, 528, 4176, doi: 10.1093/mnras/stad3477

    Garg, M., Tiwari, S., Derdzinski, A., et al. 2024d, MNRAS, 528, 4176, doi: 10.1093/mnras/stad3477

    work page doi:10.1093/mnras/stad3477

  20. [20]

    Isoyama, R

    Gong, X., Xu, S., Gui, S., Huang, S., & Lau, Y.-K. 2021, in Handbook of Gravitational Wave Astronomy (Springer Singapore), 24, doi: 10.1007/978-981-15-4702-7 24-1

    work page doi:10.1007/978-981-15-4702-7 2021

  21. [21]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357, doi: 10.1038/s41586-020-2649-2

    work page doi:10.1038/s41586-020-2649-2 2020

  22. [22]

    2012, ApJ, 750, 34, doi: 10.1088/0004-637X/750/1/34 Hossein Nouri, F., & Janiuk, A

    Horn, B., Lyra, W., Mac Low, M.-M., & S´ andor, Z. 2012, ApJ, 750, 34, doi: 10.1088/0004-637X/750/1/34 Hossein Nouri, F., & Janiuk, A. 2024, A&A, 687, A184, doi: 10.1051/0004-6361/202348796

    work page doi:10.1088/0004-637x/750/1/34 2012

  23. [23]

    Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90, doi: 10.1109/MCSE.2007.55

    work page doi:10.1109/mcse.2007.55 2007

  24. [24]

    2011, PhRvD, 84, 024032, doi: 10.1103/PhysRevD.84.024032

    Kocsis, B., Yunes, N., & Loeb, A. 2011, PhRvD, 84, 024032, doi: 10.1103/PhysRevD.84.024032

    work page doi:10.1103/physrevd.84.024032 2011

  25. [25]

    Kormendy, J., & Ho, L. C. 2013, ARA&A, 51, 511, doi: 10.1146/annurev-astro-082708-101811

    work page internal anchor Pith review doi:10.1146/annurev-astro-082708-101811 2013

  26. [26]

    Kubli, N., Mayer, L., Deng, H., & Lin, D. N. C. 2026, MNRAS, 546, stag024, doi: 10.1093/mnras/stag024

    work page doi:10.1093/mnras/stag024 2026

  27. [27]

    2025a, Reports on Progress in Physics, 88, 056901, doi: 10.1088/1361-6633/adc9be

    Li, E.-K., Liu, S., Torres-Orjuela, A., et al. 2025, Reports on Progress in Physics, 88, 056901, doi: 10.1088/1361-6633/adc9be

    work page doi:10.1088/1361-6633/adc9be 2025

  28. [28]

    McKernan, B., Ford, K. E. S., Cook, H. E., et al. 2025, ApJ, 990, 217, doi: 10.3847/1538-4357/adf114

    work page doi:10.3847/1538-4357/adf114 2025

  29. [29]

    McPike, E., Perna, R., Ford, K. E. S., et al. 2026, arXiv e-prints, arXiv:2602.04135, doi: 10.48550/arXiv.2602.04135

    work page doi:10.48550/arxiv.2602.04135 2026

  30. [30]

    D., Giustini, M., et al

    Miniutti, G., Saxton, R. D., Giustini, M., et al. 2019, Nature, 573, 381, doi: 10.1038/s41586-019-1556-x

    work page doi:10.1038/s41586-019-1556-x 2019

  31. [31]

    C., & Pessah, M

    Murphy, G. C., & Pessah, M. E. 2015, ApJ, 802, 139, doi: 10.1088/0004-637X/802/2/139

    work page doi:10.1088/0004-637x/802/2/139 2015

  32. [32]

    Nelson, R. P. 2005, A&A, 443, 1067, doi: 10.1051/0004-6361:20042605

    work page doi:10.1051/0004-6361:20042605 2005

  33. [33]

    H., Read , M

    Paardekooper, S.-J., Baruteau, C., Crida, A., & Kley, W. 2010, MNRAS, 401, 1950, doi: 10.1111/j.1365-2966.2009.15782.x pandas development team, T. 2020, pandas-dev/pandas:

    work page doi:10.1111/j.1365-2966.2009.15782.x 2010

  34. [34]

    Pandas, Zenodo, doi: 10.5281/zenodo.3509134

    work page doi:10.5281/zenodo.3509134

  35. [35]

    Peters, P. C. 1964, Physical Review, 136, 1224, doi: 10.1103/PhysRev.136.B1224

    work page doi:10.1103/physrev.136.b1224 1964

  36. [36]

    Monthly Notices of the Royal Astronomical Society474(2), 2212–2232 (2018)

    Riols, A., & Latter, H. 2018, MNRAS, 474, 2212, doi: 10.1093/mnras/stx2455 —. 2019, MNRAS, 482, 3989, doi: 10.1093/mnras/sty2804

    work page doi:10.1093/mnras/stx2455 2018

  37. [37]

    2020, ApJ, 903, 133, doi: 10.3847/1538-4357/abbc1d

    Secunda, A., Bellovary, J., Mac Low, M.-M., et al. 2020, ApJ, 903, 133, doi: 10.3847/1538-4357/abbc1d

    work page doi:10.3847/1538-4357/abbc1d 2020

  38. [38]

    I., & Sunyaev, R

    Shakura, N. I., & Sunyaev, R. A. 1973, A&A, 24, 337

    1973

  39. [39]

    2023, Physical Review X, 13, 021035, doi: 10.1103/PhysRevX.13.021035

    Speri, L., Antonelli, A., Sberna, L., et al. 2023, Physical Review X, 13, 021035, doi: 10.1103/PhysRevX.13.021035

    work page doi:10.1103/physrevx.13.021035 2023

  40. [40]

    Tanaka, H., Takeuchi, T., & Ward, W. R. 2002, ApJ, 565, 1257, doi: 10.1086/324713

    work page doi:10.1086/324713 2002

  41. [41]

    A., & Di Cintio, P

    Trani, A. A., & Di Cintio, P. 2025, A&A, 703, A6, doi: 10.1051/0004-6361/202555879

    work page doi:10.1051/0004-6361/202555879 2025

  42. [42]

    Begelman, M. C. 2025, arXiv e-prints, arXiv:2510.21187, doi: 10.48550/arXiv.2510.21187

    work page doi:10.48550/arxiv.2510.21187 2025

  43. [43]

    Waskom, M. L. 2021, Journal of Open Source Software, 6, 3021, doi: 10.21105/joss.03021

    work page doi:10.21105/joss.03021 2021

  44. [44]

    Wu, Y., Chen, Y.-X., & Lin, D. N. C. 2024, MNRAS, 528, L127, doi: 10.1093/mnrasl/slad183

    work page doi:10.1093/mnrasl/slad183 2024

  45. [45]

    2011, PhRvL, 107, 171103, doi: 10.1103/PhysRevLett.107.171103

    Yunes, N., Kocsis, B., Loeb, A., & Haiman, Z. 2011, PhRvL, 107, 171103, doi: 10.1103/PhysRevLett.107.171103

    work page doi:10.1103/physrevlett.107.171103 2011

  46. [46]

    2022, MNRAS, 511, 6143, doi: 10.1093/mnras/stac299

    Mayer, L. 2022, MNRAS, 511, 6143, doi: 10.1093/mnras/stac299

    work page doi:10.1093/mnras/stac299 2022

  47. [47]

    A., et al

    Zwick, L., Tiede, C., Trani, A. A., et al. 2024, PhRvD, 110, 103005, doi: 10.1103/PhysRevD.110.103005 10 APPENDIX A.ANALYTICAL APPROXIMATION OF THE TURBULENT DEPHASING GAUSSIAN DISTRIBUTION As given in Eq. (12), the overall turbulent dephasing is ∆ψturb = ∆0 KX j=0 ξj a7 j −a 7 j+1 .(A1) The mean of ∆ψ turb over realizations of the{ξ i}is ⟨∆ψtrub⟩ξi = ∆0 ...

    work page doi:10.1103/physrevd.110.103005 2024