Pith. sign in

REVIEW 6 major objections 5 minor 29 references

Levy flights cross the Schwarzschild barrier: angular-momentum transport near Sgr A* is a space-fractional process with a heavy-tailed Holtsmark torque distribution, so stars jump past the barrier where local diffusion says transport halts.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 09:34 UTC pith:HWG4LVEY

load-bearing objection A provocative application of fractional transport to the Galactic Center, but the central Lévy-stable assumption is imported, not demonstrated, and the claimed Fermi-bubble match is post hoc; worth refereeing, not worth believing yet. the 6 major comments →

arxiv 2607.27317 v1 pith:HWG4LVEY submitted 2026-07-29 astro-ph.GA

Superdiffusion at the Galactic Centre

Pau Amaro Seoane , Kostas Tzanavaris , Reinhard Genzel , Frank Eisenhauer , Thomas Ott , Stefan Gillessen , Guillaume Bourdarot , Diogo C. Ribeiro
show 9 more authors
Matteo Sadun Bordoni Simran Joharle Felix Mang Andreas Burkert Jorge Cuadra Diego Calder\'on Hagai B. Perets Tsvi Piran Re'em Sari
This is my paper
classification astro-ph.GA
keywords superdiffusionLévy flightsresonant relaxationSchwarzschild barrierGalactic CentreSgr A*extreme-mass-ratio inspiraltidal disruption events
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Standard kinetic models of the Galactic Centre treat angular-momentum transport as local diffusion, and they predict that general-relativistic precession halts orbits at the Schwarzschild barrier, starving the black hole of inspiral and tidal-disruption events. Because gravity is an inverse-square force, the paper argues, resonant-relaxation torques have a Holtsmark distribution with infinite variance, so the central limit theorem does not apply. The authors replace local diffusion with a space-fractional Fokker–Planck equation driven by Lévy flights of stability index 3/2, and simulate the observed S-star cluster — including the newly discovered S301 — as a Markov chain. The heavy-tailed jumps cross the barrier without density buildup, keep flux into high-eccentricity orbits, and produce tidal disruptions and extreme-mass-ratio inspirals that local models suppress. They also show that regularized backward integration of the fractional equation maps today's orbits to a localized initial state matching the energy budget of the Fermi bubbles.

Core claim

Resonant relaxation in a near-Keplerian nuclear cluster is a continuous-time random walk with infinite-variance torque increments, so orbital transport follows a space-fractional diffusion law rather than the standard Fokker–Planck equation. The paper's central claim is that this fractional law, with stability index α = 3/2, transports angular momentum via Lévy flights that jump directly across the Schwarzschild barrier, where local diffusion would require an infinite gradient and a pile-up. In their Markov-chain simulation of 42 S-stars over 30 million years, the heavy tails allow main-sequence stars to reach dynamic tidal-disruption boundaries and compact remnants to decouple and inspiral

What carries the argument

The central object is the space-fractional Fokker–Planck equation ∂f/∂t = L_drift f + D(J) R^{3/2} f, where R^{3/2} is the symmetric Riesz–Weyl fractional derivative: an integro-differential operator that integrates the distribution over the whole domain. It produces a Lévy-stable propagator with power-law tail |ΔJ|^{-5/2} instead of a Gaussian, so single jumps can bypass regions of quenched diffusion. The companion machinery is the Holtsmark torque distribution from the inverse-square force, which fixes the stability index α = 3/2, and a regularized inverse Fourier filter that makes backward integration well-posed.

Load-bearing premise

The load-bearing premise is that cumulative resonant-relaxation torques follow a symmetric Lévy stable law with infinite variance (index α = 3/2), so the space-fractional equation governs transport; if finite-N effects, coherence times, or the central potential cut off the heavy tail, the jump-dominated flux and the backward reconstruction would not hold.

What would settle it

A direct-summation N-body simulation of a nuclear cluster near a massive black hole with the same density profile should show either an accumulation of phase-space density at the Schwarzschild barrier (against fractional claims) or clear power-law jumps with index -5/2 in angular-momentum increments and no barrier pile-up. Alternatively, high-cadence astrometry of S-stars over a decade should detect a few large angular-momentum jumps consistent with the Lévy tail; their absence would falsify the superdiffusive mechanism.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Angular-momentum flux through the Schwarzschild barrier is nonzero, so EMRI and tidal-disruption event rates in the Galactic Centre exceed local Fokker–Planck predictions.
  • Superdiffusive transport sustains gravitational-wave sources: stellar-mass black holes cross the 10^-4 Hz detection boundary once gravitational-wave energy loss outruns stochastic jumps.
  • Observed S-star orbits, including S301, can be used as empirical test particles to calibrate fractional transport coefficients.
  • Tidal heating inflates stars and moves the disruption boundary outward in time, so partial disruptions occur before the formal static tidal radius.
  • Backward fractional integration makes the Galactic Centre's past diagnosable: the current S-star phase-space distribution records the initial deposition state that may have powered the Fermi bubbles.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the infinite-variance torque assumption holds, the same fractional transport may operate in other galactic nuclei, so gravitational-wave event-rate forecasts calibrated with local diffusion may be underestimated near resolved nuclei.
  • A testable extension: monitor S-star orbital elements over years with high-cadence astrometry; a measurable fraction of large angular-momentum changes should follow the power-law tail rather than a Gaussian, distinguishing fractional from local transport.
  • The paper leaves the fraction of tidal heat retained in the star as a free parameter; calibrating it against observed S-star luminosities could turn S301 into a quantitative probe of tidal dissipation.
  • Because the backward reconstruction is done in a fixed potential, adding time-dependent gas and potentials (the expected active epoch) could shift the inferred initial state; the technique's robustness to such effects is an open question.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

6 major / 5 minor

Summary. The paper argues that resonant relaxation in the Galactic Centre nuclear star cluster is a space-fractional Lévy process with stability index α=3/2, because inverse-square gravitational encounters generate a Holtsmark torque distribution with infinite variance. On this basis the authors replace the standard local Fokker–Planck equation with the space-fractional equation (Eq. 2), simulate a Markov chain initialized with the empirical orbits of 42 S-stars (including S301), include secular GW decay and tidal heating with a dynamic disruption radius, and evolve the system for 30 Myr. They claim that the non-local jump operator lets trajectories cross the Schwarzschild barrier, producing TDEs and EMRIs. They then perform a regularized backward integration of the fractional transport equation and identify a localized initial condition at a lookback time of 3 Myr, which they associate with the energy requirements of the Fermi bubbles.

Significance. If the central claim is correct, it would challenge the standard local-kinetic-theory prediction of a Schwarzschild barrier and would have direct implications for EMRI and TDE rates in galactic nuclei, including LISA source predictions. The paper is commendable for using a realistic set of empirical S-star orbits, including the newly observed S301, and for explicitly identifying a potential limitation of Gaussian Fokker–Planck models. However, the load-bearing Lévy-stable assumption is imported rather than derived, the simulation parameters are not specified, and the Fermi-bubble match is not quantitative. The significance is therefore conditional on substantial additional justification.

major comments (6)
  1. [§2, Eqs. (1)–(2)] The central premise is that resonant-relaxation jumps in J follow P(ΔJ) ∼ |ΔJ|^{-5/2} and hence that Eq. (2) with the Riesz–Weyl operator is the correct transport equation. This is not derived in the manuscript. Holtsmark's theorem applies to the instantaneous force from an infinite Poisson background; the quantity that enters relaxation is the cumulative ΔJ over a coherence time in a finite-N cluster with relativistic precession. A finite-N nearest-neighbour cutoff and coherence-time averaging can truncate the tail. Since Eq. (2) is imported from P. Amaro Seoane (2025b), and the two entries 2025a and 2025b both point to arXiv:2511.09648, the derivation is not checkable. Because barrier bypass in Eq. (8) requires exactly this non-local jump operator, this assumption is load-bearing. The authors should either derive the tail index and the limits of the stable-law approximation, or provide
  2. [§3, Eq. (3)] The Markov-chain simulation uses σ = D_RR Δt^{2/3}, but D_RR and Δt are never specified or calibrated. For an α=3/2 stable process the scale should be (D Δt)^{2/3}, so as written D_RR has units of J rather than J^{3/2}/t; this dimensional inconsistency propagates into all event counts. The results in Fig. 2 (e.g. 8 TDEs and 3 GW mergers at 30 Myr) are from a single deterministic realization of 42 stars, with no convergence tests, no error bars, and no exploration of D_RR or Δt. The quantitative predictions therefore cannot be evaluated. Please specify the adopted values, justify the time-step choice, and include a convergence/sensitivity analysis.
  3. [§4, Fermi-bubble claim] The abstract and conclusions state that the backward reconstruction 'matches the energy requirements of the Fermi bubbles', but no mass/energy calculation is presented. Section 4 quotes an energy requirement of 10^55 erg and uses lookback times of 1.5 and 3 Myr, which are exactly the 1–3 Myr epoch cited for the bubbles. Figure 3 shows only a localized peak in f(J); nothing converts that density into an accretion rate or an energy release. As it stands, the match is post hoc. The authors should compute the implied accreted mass/energy from the reconstructed initial state and state whether it actually reaches ~10^55 erg.
  4. [§3, tidal inflation and MS lifetimes] The dynamic tidal disruption boundary depends on the cap I(t) ≤ 5 and on the main-sequence lifetime drawn uniformly from 5–35 Myr. Both choices are ad hoc; Eq. (5) gives the inflation factor from ΔE_heat/|E_tot| but |E_tot| and the retention fraction of tidal heat are not defined. The counts of TDEs vs. compact remnants (Fig. 2) will depend strongly on these parameters and on the initial stellar masses/radii, which are not stated. No sensitivity tests are provided. This is a load-bearing element for the predicted TDE/EMRI rates.
  5. [§4, Eq. (13)] The regularized backward solution depends on an unspecified low-pass filter W(k). The statement that exp(D|k|^{3/2}τ)W(k) → 0 as k→∞ guarantees integrability, but different filters will produce different reconstructed states; Fig. 3 shows one choice. Without specifying W(k) and demonstrating that the reconstructed peak at J≈1/2, lookback 3 Myr is stable with respect to the filter width and shape, the backward-mapping claim is not reproducible.
  6. [§2, Eq. (8)] The non-local flux is written as ∫_J^1 f(s,t)/(s-J)^{1/2} ds, which is a fractional integral of order 1/2 of f, not a flux derived from Eq. (2). For positive f this expression is positive, so it does not reduce to Fick's law in any limit, and it is not shown to be consistent with the Riesz–Weyl operator in Eq. (2). The assertion that this 'guarantees' barrier bypass is therefore a property of the assumed operator, not a derived consequence of the fractional Fokker–Planck equation. The authors should derive the flux from Eq. (2) or clarify the connection.
minor comments (5)
  1. [Abstract/Introduction] Typos: 'observd', 'spatilaly', 'intgration', 'Schwarszchild', 'susbtitute', 'GRA VITY'.
  2. [Figure 2] The panels at 1.7 and 2.0 Myr appear identical; please check whether the displayed state is actually identical or whether the figure needs updating.
  3. [References] Amaro Seoane 2025a and 2025b are both listed with arXiv:2511.09648; if they are distinct works, the identifiers must be corrected. This is important because Eq. (2) is attributed to 2025b.
  4. [Eq. (9)] The peak GW frequency for a highly eccentric orbit is not simply the inverse of the periastron passage timescale; please clarify the approximation and its range of validity.
  5. [§4] The statement that time-dependent potentials, gas hydrodynamics, and non-isotropic relaxation fields are held constant is an important limitation; it should appear in the main text before the backward-reconstruction claim.

Circularity Check

4 steps flagged

The central barrier-bypass result is built into the assumed Riesz-Weyl operator, and the fractional-transport premise is imported from the authors' own prior work rather than derived; the 3-Myr Fermi-bubble 'match' is read off from the same model.

specific steps
  1. self citation load bearing [Section 2, Eq. (2)]
    "To model this non-local transport, and following the work of P. Amaro Seoane (2025b), we utilize the space-fractional Fokker-Planck equation (E. Barkai et al. 2000). This kinetic equation replaces local spatial derivatives with the Riesz-Weyl fractional operator, yielding the relation, ∂f(J,t)/∂t = L_drift f + D(J) R^{3/2} f(J,t) (2)."

    The model equation that generates all subsequent superdiffusive dynamics is adopted from a same-first-author citation, and the fractional exponent α=3/2 is not independently derived here. In the references, 'P. Amaro Seoane 2025a' and 'P. Amaro Seoane 2025b' are the same arXiv:2511.09648 preprint, so the key premise—that resonant relaxation must be a Lévy-stable process—rests on a single self-citation. The simulation then simply runs the operator that was imported.

  2. self definitional [Section 3, Eq. (8) and following paragraph]
    "The non-local flux relies on the Riesz-Weyl operator, which integrates the distribution across the entire domains, F_frac(J) ∝ ∫_J^1 f(s,t)/(s−J)^{1/2} ds. Because this flux definition does not depend on the local derivative at J_SB, the distribution function does not need to steepen. A particle located at a higher angular momentum s > J_SB executes a Lévy flight, bypassing the quenched diffusion zone entirely and landing directly at J < J_SB. This jump mechanism guarantees that phase space transport proceeds through the barrier."

    The headline result that 'relativistic precession does not suppress mass-ratio inspiral rates' is a direct property of Eq. (2)/(8), not an independent conclusion. Once one defines the flux as a nonlocal integral over all s, nonzero transport across J_SB is true by construction. The paper presents this definitional consequence as a physical finding, but the barrier bypass is baked into the integro-differential operator rather than derived from encounter dynamics or N-body behavior.

  3. other [Section 4, Figure 3 caption and following text; Eq. (13)]
    "At a lookback time of 3 Myr, the fractional operator collapses the density into a localized peak, providing a mathematical reconstruction of a possible initial state for the stellar ensemble. ... The calculations from this fractional framework align with independent observational and theoretical constraints (G. Dobler et al. 2010; K.-S. Cheng et al. 2011; F. Guo & W. G. Mathews 2012; S. Nayakshin & J. Cuadra 2005)."

    The backward reconstruction Eq. (13) inverts the same assumed fractional propagator with a manual low-pass filter W(k). The 'collapse' time depends on the model's diffusion coefficient and regularizer, so the localization at 3 Myr is produced by the model's own machinery. Calling this a match to the 1–3 Myr Fermi-bubble epoch is a self-consistency check, not a falsifiable prediction: the backward operator was not derived from the Fermi-bubble energy budget, and the match is asserted after the fact.

  4. self citation load bearing [References]
    "Amaro Seoane, P. 2025a, arXiv e-prints, arXiv:2511.09648. https://arxiv.org/abs/2511.09648 / Amaro Seoane, P. 2025b, arXiv e-prints, arXiv:2511.09648, doi: 10.48550/arXiv.2511.09648."

    The two citations that carry the load of justifying both the failure of local Fokker-Planck theory ('P. Amaro Seoane 2025a') and the adoption of fractional transport ('P. Amaro Seoane 2025b') are the same arXiv preprint by the same first author, cited twice. Similarly, the conclusion's statement that the barrier is 'not observed in direct-summation N-body simulations' cites P. Brem et al. 2014 (with Amaro-Seoane as coauthor) and again P. Amaro Seoane 2025b. Thus the central premise is anchored to a self-citation chain rather than independent, machine-checkable, or externally derived support.

full rationale

The paper is not purely circular in every part: the Holtsmark (1919) infinite-variance force distribution, the S-star orbital data (including S301), Peters (1964) gravitational-wave decay, and the tidal-inflation scaling in Eqs. (4)–(6) are independent ingredients, and the paper's simulations are internally consistent. However, the two load-bearing moves—(1) that resonant relaxation is a space-fractional Lévy process with α=3/2, and (2) that the nonlocal Riesz-Weyl operator produces a guaranteed flux through the Schwarzschild barrier—are imported from the authors' own prior work (referenced twice under two different names for the same arXiv:2511.09648) and from the defining property of Eq. (2)/(8), respectively. The conclusion that relativistic precession does not suppress inspirals is therefore a property of the chosen integro-differential operator, not an independent result. The 3-Myr localized backward reconstruction is likewise generated by the same assumed fractional propagator with an unspecified regularizer and then presented as agreement with the Fermi-bubble epoch. These features make the central qualitative outcome built into the model, while leaving the statistical weight and quantitative rates dependent on the unverified Lévy-stable assumption. A score of 7 reflects that the central claim is substantially circular, but not completely so: external data and classical physics do enter, and the model could in principle be falsified by future measurements of the S-star angular-momentum jump distribution.

Axiom & Free-Parameter Ledger

6 free parameters · 6 axioms · 0 invented entities

The central transport equation and its Lévy exponent are imported from the authors' own prior preprint (Amaro Seoane 2025b). The simulation then adds model choices—D_RR, Δt, lifetimes, inflation cap, remnant mass, and filter—that are not calibrated, error-bounded, or compared with independent data.

free parameters (6)
  • Resonant relaxation diffusion coefficient D_RR
    Sets the Lévy jump scale σ = D_RR Δt^(2/3) in Eq. (3); no value or calibration is given.
  • Markov-chain timestep Δt
    Discretization of the continuous-time random walk; not specified, yet controls jump frequency and tidal heating accumulation.
  • Main-sequence lifetime range = 5–35 Myr (uniform)
    Assigned uniformly to all S-stars; controls when stars collapse to black holes.
  • Tidal inflation cap I_max = 5
    Invoked as a hydrostatic stability bound without a stellar-structure calculation.
  • Black-hole remnant mass = 10 M_sun
    Assumed for all collapsed remnants at the end of main-sequence lifetime.
  • Backward-integration filter W(k)
    Gaussian low-pass filter width is not specified; required for convergence of Eq. (13).
axioms (6)
  • domain assumption Torque fluctuations from inverse-square encounters follow a Holtsmark distribution with infinite variance, giving Lévy stable index α = 3/2
    Invoked in §2 and §3, Eq. (1); central to replacing the Fokker–Planck operator with a fractional operator.
  • domain assumption The generalized central limit theorem applies to cumulative resonant-relaxation torques
    Bridges the single-step Holtsmark distribution to a Lévy-flight CTRW; not derived for the S-star cluster specifically.
  • ad hoc to paper The space-fractional Fokker–Planck equation (Eq. 2) with Riesz–Weyl operator governs angular-momentum transport
    Adopted 'following the work of P. Amaro Seoane (2025b)' rather than derived; this is the load-bearing self-citation.
  • domain assumption Tidal heating per orbit scales as (r_t/q)^6 and inflates the stellar radius via Eq. (5)
    Quadrupolar model traced to Peale et al. 1979; ignores mode coupling, radiative losses, and the retained-heat fraction.
  • ad hoc to paper Stars survive inflation only up to I ≤ 5, then disrupt or lose mass
    Cap is chosen from stellar-structure plausibility but is central to TDE counts.
  • domain assumption Regularized backward fractional diffusion recovers the true past state
    The inverse problem is ill-posed; the filter W(k) is chosen to force convergence, but uniqueness and correctness are not established.

pith-pipeline@v1.3.0-daily-deepseek · 11587 in / 11854 out tokens · 94320 ms · 2026-08-01T09:34:59.936586+00:00 · methodology

0 comments
read the original abstract

Tracking S-star cluster orbits around Sgr A* calibrates orbital transport models for space-borne gravitational wave detectors. Standard kinetic theories model this cluster via local Fokker-Planck equations, which predict that general relativistic precession halts angular momentum diffusion at the Schwarzschild barrier. Because inverse-square gravitational encounters generate a Holtsmark torque distribution with infinite variance, resonant relaxation operates as a space-fractional process governed by non-local L\'{e}vy flights. We simulate this superdiffusive continuous-time random walk using a Markov chain initialized with empirical S-star orbits, including the recently observd S301. Integro-differential fractional operators allow trajectories to cross regions of quenched local diffusion without density buildup at the barrier. Non-equilibrium regimes yield immediate linear flux growth, while secular tidal heating at periastron inflates stellar radii to shift disruption boundaries. Regularized backward integration of the fractional transport equation traces current phase space configurations back to initial deposition states, matching the energy requirements of the \emph{Fermi} bubbles. Relativistic precession does not suppress mass-ratio inspiral rates, which provides a model for event topologies in target galactic nuclei.

Figures

Figures reproduced from arXiv: 2607.27317 by Andreas Burkert, Diego Calder\'on, Diogo C. Ribeiro, Felix Mang, Frank Eisenhauer, Guillaume Bourdarot, Hagai B. Perets, Jorge Cuadra, Kostas Tzanavaris, Matteo Sadun Bordoni, Pau Amaro Seoane, Re'em Sari, Reinhard Genzel, Simran Joharle, Stefan Gillessen, Thomas Ott, Tsvi Piran.

Figure 1
Figure 1. Figure 1: Comparison of the propagators for local diffusion and fractional diffusion. We employ a Gaussian propagator corresponding to a diffusion coefficient D = 1 and time t = 1 for the local model. We utilize a L´evy stable distribution with a stability index α = 3/2 for the fractional model. We depict the probability densities on a single logarithmic scale to isolate the heavy-tailed asymptotic behavior. At a st… view at source ↗
Figure 2
Figure 2. Figure 2: Mosaic of six sequential snapshots detailing the phase space evolution of the S-star cluster. We display the dimensionless potential Rs/q on the horizontal axis and the fractional pericenter velocity β = v/c on the vertical axis. The gray dotted curve indicates the Keplerian circular velocity limit. The shaded blue region bounds the dynamic tidal disruption zone, limited by the minimum and maximum disrupti… view at source ↗
Figure 3
Figure 3. Figure 3: Reconstruction of historical phase space topolo￾gies via regularized backward integration of the space-frac￾tional Fokker-Planck equation. The horizontal axis measures the normalized angular momentum J, and the vertical axis denotes the probability density. The current state at look￾back time 0 Myr exhibits a broad, superdiffusive distribution across the phase space. Reversing the fractional transport oper… view at source ↗

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

29 extracted references · 7 canonical work pages

  1. [1]

    2018, Living Reviews in Relativity, 21, 4, doi: 10.1007/s41114-018-0013-8 Amaro Seoane, P

    Amaro-Seoane, P. 2018, Living Reviews in Relativity, 21, 4, doi: 10.1007/s41114-018-0013-8 Amaro Seoane, P. 2022, in Handbook of Gravitational Wave Astronomy, ed. C. Bambi, S. Katsanevas, & K. D. Kokkotas, 17, doi: 10.1007/978-981-15-4702-7 17-1 Amaro Seoane, P. 2025a, arXiv e-prints, arXiv:2511.09648. https://arxiv.org/abs/2511.09648 Amaro Seoane, P. 202...

    work page doi:10.48550/arxiv.2511.09648 2018

  2. [2]

    R., Freitag, M., et al

    Amaro-Seoane, P., Gair, J. R., Freitag, M., et al. 2007, Classical and Quantum Gravity, 24, 113, doi: 10.1088/0264-9381/24/17/R01

    doi:10.1088/0264-9381/24/17/r01 2007

  3. [3]

    2017, ArXiv e-prints

    Amaro-Seoane, P., Audley, H., Babak, S., et al. 2017, ArXiv e-prints. https://arxiv.org/abs/1702.00786

    Pith/arXiv arXiv 2017

  4. [4]

    2017, Phys

    Babak, S., Gair, J., Sesana, A., et al. 2017, Phys. Rev. D, 95, 103012, doi: 10.1103/PhysRevD.95.103012

    doi:10.1103/physrevd.95.103012 2017

  5. [5]

    2014, Classical and Quantum Gravity, 31, 244003, doi: 10.1088/0264-9381/31/24/244003

    Bar-Or, B., & Alexander, T. 2014, Classical and Quantum Gravity, 31, 244003, doi: 10.1088/0264-9381/31/24/244003

    doi:10.1088/0264-9381/31/24/244003 2014

  6. [6]

    2016, Astrophys

    Bar-Or, B., & Alexander, T. 2016, Astrophys. J., 820, 129, doi: 10.3847/0004-637X/820/2/129

    doi:10.3847/0004-637x/820/2/129 2016

  7. [7]

    2018, Astrophys

    Bar-Or, B., & Fouvry, J.-B. 2018, Astrophys. J. Lett., 860, L23, doi: 10.3847/2041-8213/aac88e

    doi:10.3847/2041-8213/aac88e 2018

  8. [8]

    2000, PhRvE, 61, 132, doi: 10.1103/PhysRevE.61.132

    Barkai, E., Metzler, R., & Klafter, J. 2000, PhRvE, 61, 132, doi: 10.1103/PhysRevE.61.132

    work page doi:10.1103/physreve.61.132 2000

  9. [9]

    2008, Galactic Dynamics: Second Edition, ed

    Binney, J., & Tremaine, S. 2008, Galactic Dynamics: Second Edition, ed. J. Binney & S. Tremaine (Princeton University Press)

    2008

  10. [10]

    Brem, P., Amaro-Seoane, P., & Sopuerta, C. F. 2014, Mon. Not. R. Astron. Soc., 437, 1259, doi: 10.1093/mnras/stt1948

    doi:10.1093/mnras/stt1948 2014

  11. [11]

    O., Dogiel, V

    Cheng, K.-S., Chernyshov, D. O., Dogiel, V. A., Ko, C.-M., & Ip, W.-H. 2011, Astrophys. J. Lett., 731, L17, doi: 10.1088/2041-8205/731/1/L17

    doi:10.1088/2041-8205/731/1/l17 2011

  12. [12]

    2010, Astrophys

    Weiner, N. 2010, Astrophys. J., 717, 825, doi: 10.1088/0004-637X/717/2/825 El Dayem, K. A., Abuter, R., Aimar, N., et al. 2026, arXiv e-prints, arXiv:2607.12664. https://arxiv.org/abs/2607.12664

    Pith/arXiv arXiv 2010

  13. [13]

    2018, Mon

    Fouvry, J.-B., & Bar-Or, B. 2018, Mon. Not. R. Astron. Soc., 481, 4566, doi: 10.1093/mnras/sty2571

    work page doi:10.1093/mnras/sty2571 2018

  14. [14]

    2019a, Astrophys

    Fouvry, J.-B., Bar-Or, B., & Chavanis, P.-H. 2019a, Astrophys. J., 883, 161, doi: 10.3847/1538-4357/ab2f78

    work page doi:10.3847/1538-4357/ab2f78

  15. [15]

    2019b, PhRvE, 100, 052142, doi: 10.1103/PhysRevE.100.052142

    Fouvry, J.-B., Bar-Or, B., & Chavanis, P.-H. 2019b, PhRvE, 100, 052142, doi: 10.1103/PhysRevE.100.052142

    work page doi:10.1103/physreve.100.052142

  16. [16]

    2019c, PhRvE, 99, 032101, doi: 10.1103/PhysRevE.99.032101

    Fouvry, J.-B., Bar-Or, B., & Chavanis, P.-H. 2019c, PhRvE, 99, 032101, doi: 10.1103/PhysRevE.99.032101

    work page doi:10.1103/physreve.99.032101

  17. [17]

    2010, Reviews of Modern Physics, 82, 3121, doi: 10.1103/RevModPhys.82.3121

    Genzel, R., Eisenhauer, F., & Gillessen, S. 2010, Reviews of Modern Physics, 82, 3121, doi: 10.1103/RevModPhys.82.3121

    doi:10.1103/revmodphys.82.3121 2010

  18. [18]

    M., Eisenhauer, F., et al

    Gillessen, S., Plewa, P. M., Eisenhauer, F., et al. 2017, Astrophys. J., 837, 30, doi: 10.3847/1538-4357/aa5c41

    doi:10.3847/1538-4357/aa5c41 2017

  19. [19]

    Guo, F., & Mathews, W. G. 2012, ApJ, 756, 181, doi: 10.1088/0004-637X/756/2/181

    doi:10.1088/0004-637x/756/2/181 2012

  20. [20]

    Hills, J. G. 1988, Nat, 331, 687

    1988

  21. [21]

    1919, Annalen der Physik, 363, 577, doi: 10.1002/andp.19193630702

    Holtsmark, J. 1919, Annalen der Physik, 363, 577, doi: 10.1002/andp.19193630702

    work page doi:10.1002/andp.19193630702 1919

  22. [22]

    2006, ApJ, 645, 1152, doi: 10.1086/504400

    Hopman, C., & Alexander, T. 2006, ApJ, 645, 1152, doi: 10.1086/504400

    doi:10.1086/504400 2006

  23. [23]

    1994, Stellar Structure and Evolution (Springer-Verlag Berlin Heidelberg)

    Kippenhahn, R., & Weigert, A. 1994, Stellar Structure and Evolution (Springer-Verlag Berlin Heidelberg)

    1994

  24. [24]

    Merritt, D., Alexander, T., Mikkola, S., & Will, C. M. 2011, Phys. Rev. D, 84, 044024, doi: 10.1103/PhysRevD.84.044024

    doi:10.1103/physrevd.84.044024 2011

  25. [25]

    2000, PhR, 339, 1, doi: 10.1016/S0370-1573(00)00070-3

    Metzler, R., & Klafter, J. 2000, PhR, 339, 1, doi: 10.1016/S0370-1573(00)00070-3

    doi:10.1016/s0370-1573(00)00070-3 2000

  26. [26]

    2005, Astron

    Nayakshin, S., & Cuadra, J. 2005, Astron. Astrophys., 437, 437, doi: 10.1051/0004-6361:20042052

    doi:10.1051/0004-6361:20042052 2005

  27. [27]

    J., Cassen, P., & Reynolds, R

    Peale, S. J., Cassen, P., & Reynolds, R. T. 1979, Science, 203, 892, doi: 10.1126/science.203.4383.892

    doi:10.1126/science.203.4383.892 1979

  28. [28]

    Peters, P. C. 1964, Physical Review, 136, 1224

    1964

  29. [29]

    P., & Tremaine, S

    Rauch, K. P., & Tremaine, S. 1996, New Astronomy, 1, 149

    1996