Pith. sign in

REVIEW 3 major objections 6 minor 49 references

A Hybrid Algorithm for Drift-Kinetic Particle Dynamics within General Relativistic Magnetohydrodynamics Simulations of Black Holes Accretion Flows

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A second-order semi-implicit integrator tracks charged-particle guiding centers in black-hole accretion flows without resolving gyration, and exposes interpolation-driven magnetic-moment artifacts.

desk verdict A coherent semi-implicit integrator for covariant guiding center motion coupled to GRMHD, but thin trajectory tests and a hand-wavy magnetic-moment derivation keep it from being a finished community tool. read the letter →

arxiv 2507.11616 v1 pith:ZIWWGWEK submitted 2025-07-15 astro-ph.HE physics.comp-phphysics.plasm-ph

Tyler Trent , Dimitrios Psaltis , Feryal Özel This is my paper · ORCID
classification astro-ph.HEphysics.comp-phphysics.plasm-ph
keywords guidingcenterdrift-kineticGRMHDsemi-implicitintegratortricubicinterpolationmagneticmomentblackholeaccretiontest-particledynamics
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper claims that the recently derived covariant guiding-center equations of motion for charged particles can be integrated over macroscopic times in realistic general-relativistic magnetohydrodynamics (GRMHD) black-hole accretion-flow backgrounds, without resolving the gyroperiod. The key move is a second-order semi-implicit integrator: the unstable linear Lorentz term is treated implicitly and inverted analytically, while the stable gravitational and mirror terms remain explicit, so the timestep can be set by the drift scale rather than the gyroperiod. Comparing against full Lorentz-force integrations on a GRMHD accretion-flow snapshot, the authors find agreement at the level of a combined error budget $\epsilon \simeq \xi \rho_0 + \delta r^3$, where $\rho_0$ is the gyroradius and $\delta r$ the grid spacing. They further show that interpolation of the discretized fields violates the homogeneous Maxwell equations and artificially changes the magnetic moment, and they derive a covariant equation for that spurious growth. If this is right, the method supplies a practical bridge from fluid GRMHD simulations to kinetic transport and particle acceleration in black-hole accretion flows.

What carries the argument

The load-bearing object is the two-stage semi-implicit integrator in eqs. (14)-(18), which treats only the linear velocity term $(q/m)F^{\alpha}_{\beta} U^{\beta}$ implicitly, averages it between old and new velocity, and solves the resulting linear system analytically, while keeping the Christoffel term, the $\mu \nabla^{\alpha} \omega$ drift/mirror term, and the position update explicit. The timestep is set by eq. (19), the time for the guiding center to drift into a region of appreciably different field, with a choice $\xi = 10^{-3}$; the conserved norm $U^{\alpha} U_{\alpha} + 2 \mu \omega = -1$ is used to update the time component of the four-velocity. Around this core sit the tricubic interpolation of the background magnetic field and fluid velocity (with the electromagnetic tensor reconstructed afterward to enforce the force-free condition), and the covariant magnetic-moment evolution equation (30), whose derivation fixes a free term by demanding that $\mu$ be constant when Faraday's law holds. That equation is what lets the paper attribute trajectory errors to interpolation-induced Maxwell violations rather than to the integrator.

What would settle it

Re-run the Fig. 3 case at the same grid resolution but reconstruct the field by interpolating a vector potential, so that $\nabla \cdot B = 0$ and Faraday's law hold by construction; if the full and guiding-center trajectories still separate at $t \approx 15 M$, the divergence cannot be blamed on interpolation Maxwell violations, and the paper's diagnosis fails.

Watch

Extended reading notes

Core claim

The central claim is that the covariant guiding-center acceleration equation $dU^{\alpha}/d\tau = -\Gamma^{\alpha}_{\beta\nu} U^{\beta} U^{\nu} + (q/m) F^{\alpha}_{\beta} U^{\beta} - \mu \nabla^{\alpha} \omega$ can be evolved stably with a second-order semi-implicit scheme (eqs. 14-18) whose timestep is set by the drift scale, not the gyroperiod. In the GRMHD tests shown, guiding-center trajectories track full Lorentz-force integrations within an error budget $\epsilon \simeq \xi \rho_0 + \delta r^3$, with $\xi$ of order a few; the gyroradius $\rho_0$ sets the floor and interpolation error falls as the cube of the grid spacing. The paper also establishes that interpolating the electromagnetic field from a discretized GRMHD snapshot introduces violations of $\nabla \cdot B = 0$ and of Faraday's law, that these violations drive a spurious evolution of the magnetic moment, and that the covariant equation (30) quantitatively accounts for the divergence seen between full and guiding-center trajectories at $t \approx 15 M$ in one test case. The same artifact affects full-trajectory integrations on interpolated fields, so the diagnosis is not specific to the guiding-center approximation.

Load-bearing premise

The load-bearing premise is that the paper's derived equation (30) correctly describes how interpolation-induced violations of Maxwell's equations change a particle's magnetic moment; if that equation is wrong, the paper has not shown that the $t \approx 15 M$ trajectory divergence is an interpolation artifact rather than a failure of the integrator.

Editorial extensions

If this is right

  • Charged-particle trajectories in inner accretion flows can now be integrated for durations of tens of $M$ with steps set by the drift scale, making kinetic post-processing of GRMHD snapshots practical.
  • The scaling $\epsilon \simeq \xi \rho_0 + \delta r^3$ means the guiding-center approximation itself sets the accuracy floor, and increasing grid resolution pays off only cubically once interpolation dominates.
  • Any simulation that interpolates electromagnetic fields from a discretized grid is exposed to the same artifact: magnetic-moment conservation is broken by interpolation-induced violations of the homogeneous Maxwell equations, regardless of whether a guiding-center reduction is used.
  • Using a single static GRMHD snapshot leaves $\partial B/\partial t = 0$, so the interpolated field has an unbalanced curl of the electric field; time-evolving backgrounds should largely remove this particular contribution to the spurious magnetic-moment growth.
  • With time-dependent backgrounds, the method provides a route to studying non-thermal electron acceleration and the confinement of hot spots in flares around Sgr A*.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The authors leave implicit that eq. (30) can be turned into an online monitor: a hybrid simulation could flag or compensate for interpolation-induced pitch-angle diffusion as it integrates.
  • A corollary of the error scaling is a grid-design rule for kinetic post-processing: to keep interpolation error below the gyroradius floor one wants $\delta r \lesssim (\xi \rho_0)^{1/3}$, a constraint that may bind more tightly than the guiding-center validity conditions themselves.
  • If the diagnosis of the $t \approx 15 M$ divergence is correct, then vector-potential-based interpolation, which preserves the homogeneous Maxwell equations by construction, should eliminate the divergence at fixed resolution; this is a natural test the paper does not run.
  • The same machinery, applied to time-dependent backgrounds, might reveal whether the spurious magnetic-moment growth seen here is largely a static-snapshot artifact, and whether prior hybrid particle-MHD studies of turbulent plasmas were biased by interpolation Maxwell violations.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper presents a semi-implicit numerical integrator for the covariant guiding-center equations of Trent et al. (2024), designed to evolve charged-particle trajectories in GRMHD backgrounds over macroscopic timescales. The scheme treats the Lorentz-force term implicitly and the remaining terms explicitly, with a second-order Runge-Kutta-like half-step/full-step structure derived in Appendix A. The authors apply the method to a static Athena++ SANE accretion-flow snapshot around a spinning black hole, using tricubic interpolation of the background fields and a 3+1 reconstruction of the electromagnetic field tensor. They validate the method by comparing guiding-center trajectories with full Lorentz-force integrations for a small set of initial conditions, and they examine the impact of grid resolution in a flat-space magnetic-dipole setup, finding an error scaling of roughly ξρ0 + δr^3. The paper also derives an evolution equation for the magnetic moment and uses it to argue that a trajectory divergence seen at t≈15M is caused by interpolation-induced violations of the homogeneous Maxwell equations rather than by an integrator failure.

Significance. If the central claims hold, the algorithm would be a useful tool for hybrid kinetic-GRMHD simulations of black-hole accretion, because it would allow particle trajectories to be followed over macroscopic times while avoiding the gyroperiod timestep restriction. The paper has several genuine strengths: the semi-implicit integrator is derived analytically rather than merely asserted; the flat-space stability analysis is clear and shows why explicit schemes fail; the comparison against full Lorentz-force integration is an appropriate external benchmark and not circular; and the grid-resolution study in Fig. 6 provides a concrete, falsifiable error scaling. The paper also honestly identifies a real and often neglected difficulty, namely that interpolated electromagnetic fields from GRMHD snapshots do not satisfy the homogeneous Maxwell equations and can artificially evolve the magnetic moment. However, the validation rests on only a handful of GRMHD trajectories, the counterintuitive gyroradius dependence in Fig. 2 is left unexplained, and the magnetic-moment evolution equation used to diagnose the Fig. 3 divergence is not derived rigorously.

major comments (3)
  1. [§3.2, Eq. (30), Appendix C] The magnetic-moment evolution equation used to diagnose the Fig. 3 divergence is not derived in a way that supports the conclusion. In Eq. (C35) a placeholder term ξ is introduced and later fixed by requiring that ⟨dμ/dt⟩ vanish when Faraday's law holds (Eqs. C37–C41). This is a consistency condition, not a derivation: it guarantees the final expression vanishes on Maxwell solutions, but it does not determine the response to the specific interpolation-induced Maxwell violations unless the perturbation calculation is actually carried through. Moreover, the covariant form in Eq. (30) is ill-defined: ζ^λ ≡ (U_∥, b) mixes a scalar parallel velocity with a spatial unit vector, so the contraction is not manifestly covariant, and the denominator sqrt(2F_{ιη}F^{ιη}) does not equal the sqrt(B²−E²) used in Eq. (26) under the stated definitions. Since the claim that the t≈15M divergence in Fig. 3 is an interpolation/static-snapshot artifact rests on Eq. (30), that claim is currently unsupported. Please replace this with a direct evaluation of ∇·B and ∂B/∂t+∇×E along the actual guiding-center trajectory (as is already done for the full trajectory in the bottom panels of Figs. 4 and 5), or provide a systematic derivation of Eq. (30).
  2. [§2.3 and Fig. 3] The discrete update of the magnetic moment is never specified. The text says the scheme "allows for the magnetic moment to be evolved" and Fig. 3 presents a trajectory "calculated for an evolving magnetic moment," but there is no equation showing how μ_{n+1/2} and μ_{n+1} are obtained from Eq. (30), nor how the evolving μ is coupled back into the semi-implicit position/velocity update in Eqs. (14)–(18). Without this information, the dashed green curve in Fig. 3 is not reproducible and the accuracy of the evolving-μ variant cannot be assessed. Please provide the full discrete update and state explicitly whether μ is evolved with the same timestep and whether it is held fixed or advanced during the half-step and full-step stages.
  3. [§4, Fig. 2] The result that the trajectory with the larger initial gyroradius (ρ0 = 10^{-5} M) matches the full trajectory more closely than the smaller one (ρ0 = 10^{-7} M) is explicitly called counterintuitive but is never explained. This ordering is important because it bears directly on the convergence of the guiding-center approximation and on the error budget in Eq. (32), which predicts a smaller absolute error for smaller ρ0 when the δr^3 interpolation term is subdominant. If the δr^3 term dominates for these trajectories, that should be stated and demonstrated; if the explanation involves the static-snapshot Maxwell violation, it should be quantified along each trajectory. As written, the paper leaves a central accuracy result unexplained, and this should be resolved before the method's accuracy claims can be accepted.
minor comments (6)
  1. [§4, Abstract] The abstract claims that the method demonstrates "accuracy and efficiency," but no wall-clock times, step counts, or comparisons with a fully explicit integrator are reported. Please add at least a quantitative measure of the efficiency gain, or soften the claim.
  2. [Eq. (C41)] In Eq. (C41), the second term is written first as −(μ/B)∂B/∂t and then as −(μ/B)∂B/∂t·b in the same displayed equation; this appears to be a typo and should be corrected for clarity.
  3. [§3.1] The statement that the tricubic interpolation "works for non-Cartesian coordinate grids as well" needs clarification, because the interpolated quantities are coordinate components of vector fields in Kerr-Schild spherical coordinates. The interpolation error near coordinate singularities and the transformation to Boyer-Lindquist coordinates should be discussed, especially since the guiding-center equation depends on field derivatives.
  4. [§4.1, Fig. 6] In Fig. 6, the error floor is attributed to the guiding-center approximation, but the text does not state how many trajectories were used per grid resolution or what the statistical uncertainty in the plotted differences is. Please add error bars or repeat runs to support the claimed scaling.
  5. [§4] The statement that a time-evolving GRMHD background would counteract the spurious curl of the electric field via ∂B/∂t is plausible but speculative; it should either be tested with a time-dependent simulation or clearly marked as a forward-looking remark rather than a conclusion of this work.
  6. [Eq. (19)] The timestep formula in Eq. (19) uses ξ = 10^{-3}, but no convergence study with respect to ξ is shown for the GRMHD backgrounds. Please add a brief study or a statement of how this value was established.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central accuracy claim is benchmarked against full Lorentz-force integration, an external reference, and does not reduce to the guiding-center inputs or fitted parameters.

full rationale

The paper's central claim is that the semi-implicit integrator (eqs. 14-18) stably evolves the covariant guiding-center equations (eq. 8) and matches full Lorentz-force trajectories within a quantified error budget. That claim is checked directly against fourth-order Runge-Kutta integration of the full Lorentz force (Figs. 2, 3, and 6), which is external to the guiding-center formalism and not fitted. The guiding-center equations themselves are taken from prior self-cited work (Trent et al. 2024, eqs. 1, 8, 10); these are input assumptions rather than predicted outputs, and the paper's own full-trajectory comparisons provide an independent test of their adequacy at the tested parameters, so the self-citation is not load-bearing in an unverified way. The stepsize parameter xi = 10^-3 (eq. 19) and the error-scaling coefficient in eq. (32) are empirical choices or fits, but the trajectory comparisons are measurements against the full equations, not predictions forced by those choices. The weakest logical step is Appendix C, where the magnetic-moment evolution equation (30) is obtained by adding a placeholder term xi and fixing it through the requirement that the magnetic moment be invariant when Faraday's law holds (eqs. C35-C41). This is a consistency construction rather than a fully independent derivation and leaves a rigor gap, but the resulting equation is used only to diagnose interpolation artifacts and is itself checked against the independently computed magnetic moment of the full trajectory (bottom panels of Figs. 4-5). Consequently, no step in the claimed derivation chain reduces by construction to its own inputs; the central claim stands on an external benchmark.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the correctness of the authors' prior covariant guiding center formalism (Trent et al. 2024), the validity of the guiding center ordering, the conservation-law projection used in the integrator, and the adequacy of tricubic interpolation on a static GRMHD snapshot. One free parameter, xi, controls the stepsize. No new physical entities are introduced.

free parameters (1)
  • xi (timestep safety factor) = 10^-3
    Used in eq. (19) to set the integration stepsize. The authors state that this choice leads to results not limited by truncation error, but no convergence study is reported to justify the value (Section 2.3, Section 4).
assumptions (5)
  • domain assumption The covariant guiding center equations of motion from Trent et al. (2024), including the acceleration equation dU^alpha/dtau = -Gamma^alpha_betanu U^beta U^nu + (q/m)F^alpha_beta U^beta - mu grad^alpha omega, are correct.
    The paper's core integrator solves these equations without re-deriving them; the correctness of the algorithm inherits from the cited derivation (Section 2.1).
  • domain assumption Guiding center ordering: fields slowly varying over gyroradius and gyroperiod, and gravity weaker than electromagnetic forces (eqs. 3-5).
    Required for the guiding center approximation used in eq. (8). The paper checks Psi1 and Psi2 a posteriori in Section 4 but does not enforce them during integration.
  • domain assumption The conservation law U^alpha U_alpha + 2 mu omega = -1 holds along the numerical trajectory and is used to update the time component of the four-velocity.
    Equation (10) from Trent et al. (2024) is used at every half and full step (Section 2.3), so the integration scheme's stability and accuracy depend on this relation.
  • ad hoc to paper Tricubic interpolation with finite-difference derivatives reconstructs the GRMHD fields accurately enough, and the interpolated fields violate the homogeneous Maxwell equations only at the expected truncation order.
    Interpolation of B and u from the GRMHD grid is central (Section 3.1). The paper explicitly acknowledges the resulting mu non-conservation and treats it as an artifact (Sections 3.2 and 4).
  • domain assumption A static GRMHD snapshot with d B/dt = 0 is a valid test bed; the resulting unopposed curl of E is the dominant cause of the observed trajectory divergence.
    The background is a single snapshot at t = 130,000 M (Section 3.3). The paper flags that a time-dependent background would counteract the curl term via d B/dt (Section 4).

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Hybrid Algorithm for Drift-Kinetic Particle Dynamics within General Relativistic Magnetohydrodynamics Simulations of Black Holes Accretion Flows." pith.science (2026). https://pith.science/paper/ZIWWGWEK

@misc{pith2026250711616,
  author       = {Pith},
  title        = {Pith review of: A Hybrid Algorithm for Drift-Kinetic Particle Dynamics within General Relativistic Magnetohydrodynamics Simulations of Black Holes Accretion Flows},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZIWWGWEK}},
  note         = {Machine review of arXiv:2507.11616}
}
read the original abstract

Astrophysical plasmas in relativistic spacetimes, such as black hole accretion flows, are often weakly collisional and require kinetic modeling to capture non-local transport and particle acceleration. However, the extreme scale separation between microscopic and macroscopic processes limits the feasibility of fully kinetic simulations. A covariant guiding center formalism has recently been derived to address this challenge in curved spacetimes. We present a new hybrid numerical algorithm based on this formalism, which evolves the trajectories of charged particles over macroscopic timescales in GRMHD backgrounds. To address numerical instabilities in the equations of motion, we develop a semi-implicit integrator that ensures stable evolution in strong-field environments. We apply our method to GRMHD simulations of black hole accretion flows, demonstrating its accuracy and efficiency across a range of physical conditions.

Figures

Figures reproduced from arXiv: 2507.11616 by the authors.

Figure 1
Figure 1. Guiding center trajectories in a GRMHD background, initiated at the locations of the black boxes, with particle velocities that are parallel (green), perpendicular (yellow), and anti-parallel (purple) to the local magnetic fields. The black sphere represents the black hole event horizon. The vertical and horizontal cross-sections show contours of magnetic field strength. The horizontal cross-section also includes pr… view at source ↗
Figure 2
Figure 2. Comparison of particle trajectories calculated by integrating the guiding-center equations (labeled as “GC”) and the full equations of motion for charged particles (label as “Full”), for different values of the initial gyroradius ρ0. Trajectories where initialized at a radius of 12.8M, with an initial velocity orientation perpendicular to the local magnetic field and particle Lorentz factors of 10 (see also [PITH_F… view at source ↗
Figure 3
Figure 3. Same as [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Characteristics of the trajectories shown in [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 6
Figure 6. Figure 6: shows the difference in the final positions for simulations with different initial gyroradii and with different radial grid spacings; in each simulation the poloidal grid has twice as many points as the radial grid. The difference in the final positions between the two…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 24 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter doi edition editor eprint howpublished institution journal key month number organization pages publisher school series title misctitle type volume year version url label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION format.url url empty "" new.block "" url * "" * if FUNCTION format.eprint eprint empty "" archivePrefix empty "" archivePrefix "arXiv" = new.block " " eprint * " " * new.block " " eprint * " " * if if if FUNCTION format.doi doi empty "" " " doi * " " * if FUNCTION format.pid doi empty eprint empty ur...

  3. [3]

    du-\;wx衇/O0цeرѻwxgcَCX

    thebibliography [1] 20pt to REFERENCES 6pt =0pt \@twocolumntrue 12pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key o...

    work page 2017

  4. [4]

    C., Werner, G

    Adams, L. C., Werner, G. R., & Cary, J. R. 2025, arXiv e-prints, arXiv:2503.13697, 10.48550/arXiv.2503.13697

    work page doi:10.48550/arxiv.2503.13697 2025

  5. [5]

    Avara, M., \" O zel, F., & Psaltis, D. 2025

    work page 2025

  6. [6]

    A., & Parfrey, K

    Bacchini, F., Ripperda, B., Philippov, A. A., & Parfrey, K. 2020, The Astrophysical Journal Supplement Series, 251, 10, 10.3847/1538-4365/abb604

    work page doi:10.3847/1538-4365/abb604 2020

  7. [7]

    2024, Monthly Notices of the Royal Astronomical Society, 529, 2399, 10.1093/mnras/stae723

    Bacchini, F., Ruan, W., & Keppens, R. 2024, Monthly Notices of the Royal Astronomical Society, 529, 2399, 10.1093/mnras/stae723

    doi:10.1093/mnras/stae723 2024

  8. [8]

    2018, The Astrophysical Journal, 862, 80, 10.3847/1538-4357/aac820

    Ball, D., Sironi, L., & \" O zel, F. 2018, The Astrophysical Journal, 862, 80, 10.3847/1538-4357/aac820

    doi:10.3847/1538-4357/aac820 2018

Show all 49 references
  1. [9]

    2019, The Astrophysical Journal, 884, 57, 10.3847/1538-4357/ab3f2e

    ---. 2019, The Astrophysical Journal, 884, 57, 10.3847/1538-4357/ab3f2e

    2019 doi

  2. [10]

    W., & Shapiro, S

    Baumgarte, T. W., & Shapiro, S. L. 2010, Numerical Relativity: Solving Einstein's Equations on the Computer (Cambridge University Press), 698

    2010

  3. [11]

    J., & Markowski, D

    Brizard, A. J., & Markowski, D. G. 2022, Physics of Plasmas, 29, 022101, 10.1063/5.0078786

    2022 doi

  4. [12]

    W., & Squire, J

    Burby, J. W., & Squire, J. 2020, Journal of Plasma Physics, 86, 835860601, DOI: 10.1017/S002237782000080X

    2020 doi

  5. [13]

    R., & Brizard, A

    Cary, J. R., & Brizard, A. J. 2009, Reviews of Modern Physics, 81, 693, 10.1103/RevModPhys.81.693

    2009 doi

  6. [14]

    2021, The Astrophysical Journal, 909, 67, 10.3847/1538-4357/abdc28

    Christian, P., & Chan, C.-k. 2021, The Astrophysical Journal, 909, 67, 10.3847/1538-4357/abdc28

    2021 doi

  7. [15]

    2023, A & A, 677

    Collaboration, G., Abuter, R., Aimar, N., et al. 2023, A & A, 677. https://doi.org/10.1051/0004-6361/202347416

    2023 doi

  8. [16]

    De Villiers, J.-P., & Hawley, J. F. 2003, The Astrophysical Journal, 592, 1060, 10.1086/375866

    2003 doi

  9. [17]

    F., McKinney, J

    Gammie, C. F., McKinney, J. C., & T \' o th, G. 2003, The Astrophysical Journal, 589, 444, 10.1086/374594

    2003 doi

  10. [18]

    Gordovskyy, M., & Browning, P. K. 2011, The Astrophysical Journal, 729, 101, 10.1088/0004-637X/729/2/101

    2011 doi

  11. [19]

    K., Kusano, K., Inoue, S., & Vekstein, G

    Gordovskyy, M., Browning, P. K., Kusano, K., Inoue, S., & Vekstein, G. E. 2023, The Astrophysical Journal, 952, 75, 10.3847/1538-4357/acdb4d

    2023 doi

  12. [20]

    K., & Vekstein, G

    Gordovskyy, M., Browning, P. K., & Vekstein, G. E. 2010 a , A & A, 519. https://doi.org/10.1051/0004-6361/200913569

    2010 doi

  13. [21]

    2010 b , The Astrophysical Journal, 720, 1603, 10.1088/0004-637X/720/2/1603

    ---. 2010 b , The Astrophysical Journal, 720, 1603, 10.1088/0004-637X/720/2/1603

    2010 doi

  14. [22]

    Hakobyan, H., Ripperda, B., & Philippov, A. A. 2023, The Astrophysical Journal Letters, 943, L29, 10.3847/2041-8213/acb264

    2023 doi

  15. [23]

    B., Quataert, E., et al

    Kempski, P., Fielding, D. B., Quataert, E., et al. 2023, Monthly Notices of the Royal Astronomical Society, 525, 4985, 10.1093/mnras/stad2609

    2023 doi

  16. [24]

    Komissarov, S. S. 1999, Monthly Notices of the Royal Astronomical Society, 308, 1069, 10.1046/j.1365-8711.1999.02783.x

    1999

  17. [25]

    2011, Monthly Notices of the Royal Astronomical Society: Letters, 418, L94, 10.1111/j.1745-3933.2011.01150.x

    ---. 2011, Monthly Notices of the Royal Astronomical Society: Letters, 418, L94, 10.1111/j.1745-3933.2011.01150.x

    2011

  18. [26]

    W., Schekochihin, A

    Kunz, M. W., Schekochihin, A. A., & Stone, J. M. 2014, Phys. Rev. Lett., 112, 205003, 10.1103/PhysRevLett.112.205003

    2014 doi

  19. [27]

    2005, International Journal for Numerical Methods in Engineering, 63, 455, https://doi.org/10.1002/nme.1296

    Lekien, F., & Marsden, J. 2005, International Journal for Numerical Methods in Engineering, 63, 455, https://doi.org/10.1002/nme.1296

    2005 doi

  20. [28]

    C., & Gammie, C

    McKinney, J. C., & Gammie, C. F. 2004, The Astrophysical Journal, 611, 977, 10.1086/422244

    2004 doi

  21. [29]

    W., Thorne, K

    Misner, C. W., Thorne, K. S., & Wheeler, J. A. 1973, Gravitation (San Francisco: W. H. Freeman)

    1973

  22. [30]

    Northrop, T. G. 1963, The Adiabatic Motion of Charged Particles , Interscience Manuals Series (Interscience Publishers). https://books.google.com/books?id=pPNQAAAAMAAJ

    1963

  23. [31]

    2004, A Relativist's Toolkit : The Mathematics of Black-hole Mechanics (Cambridge, UK ;: Cambridge University Press)

    Poisson, E. 2004, A Relativist's Toolkit : The Mathematics of Black-hole Mechanics (Cambridge, UK ;: Cambridge University Press)

    2004

  24. [32]

    2018, The Astrophysical Journal Supplement Series, 235, 21, 10.3847/1538-4365/aab114

    Ripperda, B., Bacchini, F., Teunissen, J., et al. 2018, The Astrophysical Journal Supplement Series, 235, 21, 10.3847/1538-4365/aab114

    2018 doi

  25. [33]

    2017 a , Monthly Notices of the Royal Astronomical Society, 471, 3465, 10.1093/mnras/stx1875

    Ripperda, B., Porth, O., Xia, C., & Keppens, R. 2017 a , Monthly Notices of the Royal Astronomical Society, 471, 3465, 10.1093/mnras/stx1875

    2017 doi

  26. [34]

    2017 b , Monthly Notices of the Royal Astronomical Society, 467, 3279, 10.1093/mnras/stx379

    ---. 2017 b , Monthly Notices of the Royal Astronomical Society, 467, 3279, 10.1093/mnras/stx379

    2017 doi

  27. [35]

    1970, Introduction to the physics of space , ed

    Rossi, B. 1970, Introduction to the physics of space , ed. S. Olbert (New York: New York, McGraw-Hill)

    1970

  28. [36]

    E., Sironi, L., & Narayan, R

    Rowan, M. E., Sironi, L., & Narayan, R. 2017, The Astrophysical Journal, 850, 29, 10.3847/1538-4357/aa9380

    2017 doi

  29. [37]

    E., & Narayan, R

    Sironi, L., Rowan, M. E., & Narayan, R. 2021, The Astrophysical Journal Letters, 907, L44, 10.3847/2041-8213/abd9bc

    2021 doi

  30. [38]

    2009, The Astrophysical Journal, 698, 1523, 10.1088/0004-637X/698/2/1523

    Sironi, L., & Spitkovsky, A. 2009, The Astrophysical Journal, 698, 1523, 10.1088/0004-637X/698/2/1523

    2009 doi

  31. [39]

    2014, The Astrophysical Journal Letters, 783, L21, 10.1088/2041-8205/783/1/L21

    ---. 2014, The Astrophysical Journal Letters, 783, L21, 10.1088/2041-8205/783/1/L21

    2014 doi

  32. [40]

    D., Brzozowski III, R

    Stephens, C. D., Brzozowski III, R. W., & Jenko, F. 2017, Physics of Plasmas, 24, 102517, 10.1063/1.4998968

    2017 doi

  33. [41]

    M., Tomida, K., White, C

    Stone, J. M., Tomida, K., White, C. J., & Felker, K. G. 2020, The Astrophysical Journal Supplement Series, 249, 4, 10.3847/1538-4365/ab929b

    2020 doi

  34. [42]

    W., & Browning, P

    Threlfall, J., Hood, A. W., & Browning, P. K. 2018, A & A, 611. https://doi.org/10.1051/0004-6361/201731915

    2018 doi

  35. [43]

    2023, The Astrophysical Journal Letters, 959, L6, 10.3847/2041-8213/acf8c8

    Trent, T., Christian, P., Chan, C.-k., Psaltis, D., & \" O zel, F. 2023, The Astrophysical Journal Letters, 959, L6, 10.3847/2041-8213/acf8c8

    2023 doi

  36. [44]

    2024, arXiv e-prints, arXiv:2404.01391, 10.48550/arXiv.2404.01391

    Trent, T., Roley, K., Golden, M., Psaltis, D., & \" O zel, F. 2024, arXiv e-prints, arXiv:2404.01391, 10.48550/arXiv.2404.01391

    2024 doi arXiv

  37. [45]

    Vandervoort, P. O. 1960, Annals of Physics, 10, 401, https://doi.org/10.1016/0003-4916(60)90004-X

    1960 doi

  38. [46]

    R., Adams, L

    Werner, G. R., Adams, L. C., & Cary, J. R. 2025, arXiv e-prints, arXiv:2503.05123, 10.48550/arXiv.2503.05123

    2025 doi arXiv

  39. [47]

    R., Uzdensky, D

    Werner, G. R., Uzdensky, D. A., Begelman, M. C., Cerutti, B., & Nalewajko, K. 2018, Monthly Notices of the Royal Astronomy Society, 473, 4840, 10.1093/mnras/stx2530

    2018 doi

  40. [48]

    N., Prather, B

    Wong, G. N., Prather, B. S., Dhruv, V., et al. 2022, The Astrophysical Journal Supplement Series, 259, 64, 10.3847/1538-4365/ac582e

    2022 doi

  41. [49]

    W., & Uzdensky, D

    Zhdankin, V., Kunz, M. W., & Uzdensky, D. A. 2023, The Astrophysical Journal, 944, 24, 10.3847/1538-4357/acaf54

    2023 doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.