Pith. sign in

REVIEW 4 major objections 5 minor 2 cited by

Self-Similar Cosmic-Ray Transport in High-Resolution Magnetohydrodynamic Turbulence

T0 review · 4 major / 5 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 Cosmic rays traveling through strongly turbulent magnetic fields are scattered by sharp field-line bends, and the rarest, longest excursions—not a mean free path—set the diffusion rate, making transport nearly energy-independent over two…

desk verdict Worth refereeing, but the two-decade plateau in kappa_z is a resolution-dependent hint, not a converged result, and the authors say so themselves. read the letter →

arxiv 2507.10651 v1 pith:47S6S6AC submitted 2025-07-14 astro-ph.HE astro-ph.GAphysics.plasm-ph

Philipp Kempski , Drummond B. Fielding , Eliot Quataert , Robert J. Ewart , Philipp Grete , Matthew W. Kunz , Alexander A. Philippov , James Stone This is my paper
classification astro-ph.HEastro-ph.GAphysics.plasm-ph
keywords cosmic-raytransportMHDturbulenceresonant-curvaturescatteringLévyflightscollision-timedistributionAlfvénscalesuper-AlfvénicB/Cratio
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

This paper argues that cosmic-ray transport through strongly turbulent magnetic fields is governed by rare, sharp bends in the field lines rather than by a typical distance between scattering events. Using a $10{,}240^3$ magnetohydrodynamic turbulence simulation with fluctuating field amplitude twice the mean field ($\delta B_{\rm rms}/B_0 = 2$), the authors track about two billion test-particle cosmic rays and find that the parallel diffusion coefficient is nearly flat over almost two decades in particle gyro-radius. The reason is that collision times follow a broad power-law distribution $P(\tau)\sim\tau^{-\alpha}$ with $\alpha$ between about 1.5 and 2, so the global diffusion rate is set by the longest, rarest excursions, and those excursions are capped by the random walk of the magnetic field lines themselves. If correct, this self-similar transport would explain the observed flattening of the cosmic-ray B/C ratio, the hardening of primary spectra, and the weak energy dependence of arrival anisotropy above roughly a TeV, without invoking new physics.

What carries the argument

The load-bearing object is the collision-time distribution $P(\tau)$ for three definitions of a scattering event: reversal of the particle's velocity along the local field ($\tau_b$), reversal along the guide field ($\tau_z$), and a magnetic-moment change large enough to move between logarithmic bins ($\tau_\mu$). The argument is carried by (i) resonant-curvature scattering, defined by the local field-line curvature $K \equiv |\hat{\boldsymbol{b}}\cdot\nabla\hat{\boldsymbol{b}}|$ satisfying $K\,2\pi c/\Omega \gtrsim 1$; (ii) the power-law tail $P(\tau)\sim\tau^{-\alpha}$ with $1.5 \lesssim \alpha \lesssim 2$, which, for a truncated power law, makes $\kappa = \langle\lambda^2\rangle/2\langle\tau\rangle \propto c^2\tau_{\max}$ (equation 5), so the longest excursions—not the mean free path—set the transport rate; and (iii) the field-line random walk, whose $\tau_{z,\max}$ is roughly independent of gyro-radius and resolution, acting as the cutoff that produces the nearly flat $\kappa_z$. The curvature PDF's $K^{-1.5}$ tail ties the mechanism to the self-similar structure of the turbulence.

What would settle it

Re-run the same $\delta B_{\rm rms}/B_0 = 2$ turbulence at $20{,}480^3$ cells (or with a direct numerical simulation using explicit viscosity and resistivity) and recompute $\kappa_z$ and the collision-time distribution $P(\tau)$. If the nearly flat $\kappa_z$ plateau over $r_L/L \approx 10^{-3}$--$10^{-1}$ disappears, or if the power-law exponent $\alpha$ moves outside $1.5$--$2$, the self-similar transport is an artifact of the implicit large-eddy scheme. A cheaper diagnostic is the curvature PDF of Figure 7: check whether its peak width and the $K^{-1.5}$ tail continue to shift with resolution, since convergence of that PDF is required for the resonant-curvature mechanism to be physical.

Watch

Extended reading notes

Core claim

The paper's central claim is that resonant-curvature scattering—abrupt magnetic-moment changes when a cosmic ray hits a field-line bend whose curvature radius is comparable to or smaller than its gyro-radius—is efficient on small scales even in the presence of a significant guide field. At $10{,}240^3$ resolution with $\delta B_{\rm rms}/B_0 = 2$, the parallel diffusion coefficient $\kappa_z$ is almost constant over nearly two decades in gyro-radius ($r_L/L \approx 10^{-3}$ to $10^{-1}$), because the collision-time PDF develops a power-law tail $P(\tau)\sim\tau^{-\alpha}$ with $\alpha \approx 1.5$--$2$ and the global diffusion rate is set by the rarest, longest excursions ($\kappa \propto \tau_{\max}$) rather than by a characteristic mean free path. What caps those excursions is the random walk of the magnetic field lines themselves, which is why $\kappa_z$ depends so weakly on energy. For a galactic driving scale $L \approx 100\,\mathrm{pc}$, the simulated $\kappa_z \approx 0.07\,cL$ implies $\kappa_z \approx 7\times 10^{29}\,\mathrm{cm}^2\,\mathrm{s}^{-1}$, comparable to the transport coefficient inferred from the observed B/C flattening at ~TeV/nucleon. Despite $\delta B_{\rm rms} > B_0$, the guide field still makes transport strongly anisotropic, with $\kappa_z$ exceeding $\kappa_{xy}$ by more than an order of magnitude at small rigidities.

Load-bearing premise

The conclusion rests on the assumption that the $10{,}240^3$ grid and its numerical dissipation resolve the small, sharply curved field regions that scatter the lowest-energy cosmic rays; the curvature statistics are not yet converged with resolution, so a finer grid could still change the transport.

Editorial extensions

If this is right

  • Galactic cosmic-ray transport above a few hundred GeV can be nearly energy-independent, providing a direct explanation for the observed B/C flattening, spectral hardening, and weak anisotropy scaling without new scattering physics.
  • Standard diffusion models built around a single mean free path are insufficient in this regime; propagation codes must incorporate Lévy or fractional-diffusion transport, since the simulated collision-time distributions have no characteristic scale.
  • Resolving the Alfvén scale is a precondition for measuring the correct energy dependence: lower-resolution runs show stronger, non-converged energy dependence in $\kappa_z$, so scalings obtained without resolving $l_A$ should be treated cautiously.
  • Even with $\delta B_{\rm rms}/B_0 = 2$, transport remains strongly anisotropic ($\kappa_z \gg \kappa_{xy}$), so isotropic diffusion coefficients will misestimate cosmic-ray escape from the Galaxy.
  • The measured $\kappa_z \approx 0.07\,cL$ at small $r_L/L$ yields $\kappa_z \approx 7\times 10^{29}\,\mathrm{cm}^2\,\mathrm{s}^{-1}$ for $L\approx 100\,\mathrm{pc}$, matching the order-of-magnitude value inferred from the TeV B/C flattening.

Reading between the lines

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

  • If the same formalism is applied to a steeper collision-time PDF with $\alpha$ in $(2,3)$, the diffusion coefficient switches from $\kappa\propto\tau_{\max}$ to $\kappa\propto\tau_{\max}^{3-\alpha}r_L^{\alpha-2}$; with $\alpha=7/3$ this reproduces the observed $\kappa\sim E^{1/3}$ scaling below a few hundred GeV. The paper states this possibility but does not claim it, so treating it as a quantita
  • The resolution dependence of $\kappa_z$ is the natural control: because the authors speculate that the flattening comes from resolution-driven broadening of the curvature PDF, a still higher-resolution run (or a direct numerical simulation with explicit dissipation) should either stabilize the plateau or reveal it as an artifact of implicit large-eddy numerics.
  • The framework predicts that in turbulence with a weaker guide field (larger $\delta B_{\rm rms}/B_0$), the field-line random walk caps the longest $z$-excursions more quickly, so the flat-$\kappa_z$ range should widen while $\kappa_z$ rises; the authors' announced survey across initial conditions is the direct test of this extrapolation.
  • Because the longest excursions along $z$ are limited by field-line geometry rather than by the microscopic scattering process, the plateau level $\kappa_z\approx 0.07\,cL$ is probably robust to changes in the collision mechanism, while the slope of $P(\tau)$ is where microphysics such as cosmic-ray feedback or non-ideal dissipation would show up.
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

4 major / 5 minor

Summary. The manuscript reports test-particle cosmic-ray transport through a static snapshot of isothermal ideal-MHD turbulence from an implicit large-eddy simulation at 10,240^3 resolution with δB_rms/B0 = 2. It finds that parallel diffusion κz is almost flat over nearly two decades in particle gyro-radius, that scattering is dominated by sharply curved magnetic-field-line regions (resonant curvature scattering), and that the collision-time PDFs are broad power laws with slope between approximately 1.5 and 2, implying that global diffusion is set by the rarest, largest excursions. The authors connect this behavior to the observed flattening of the B/C ratio, the hardening of CR primary spectra, and the weak energy dependence of CR anisotropy above roughly a TeV.

Significance. If the result holds, it is significant: it offers a concrete, mechanism-based route to efficient CR scattering across a wide range of energies without a scale-by-scale resonant wave cascade, and it suggests that fractional diffusion may be needed for Galactic propagation modeling. The 10,240^3 run and ~2×10^9 test particles are a computational tour de force, and the separate field-line tracers and collision-time PDF diagnostics are genuinely useful. However, the central quantitative claim is not yet converged, and the key model parameters are measured from the same simulation used to compute κz, so the present version is a promising consistency check rather than a demonstrated prediction.

major comments (4)
  1. [Section 3, Figure 4] The two-decade weak energy dependence of κz appears only at 10,240^3; at 640^3 and 2560^3 κz is steeper and lacks the extended plateau. Because this plateau is the manuscript's central claim, the absence of numerical convergence is load-bearing. The text in Section 4 attributes the plateau to resolution-dependent broadening of the curvature PDF, which is itself not converged (Figure 7). Please provide a quantitative convergence test (e.g., a run at an intermediate resolution in the same regime, or a systematic diagnostic showing that the small-rL plateau has stabilized) or explicitly reframe the claim as tentative and remove 'self-similar' from the title and abstract.
  2. [Section 4, Figure 6 and Eqs. (4)-(5)] The conclusion that diffusion is controlled by the largest excursions rests on comparing the measured power-law slope α and cutoff τmax with Eq. (5). These quantities are extracted from the same particle tracks that yield κz, so the agreement is a consistency check rather than an independent prediction. A direct and transparent test would be to compute κz from the measured P(τ) via Eq. (4) for each gyro-radius and overplot the result on Figure 4; without this, the causal claim that the cutoff sets the diffusion rate remains one plausible interpretation, with trapping or magnetic mirroring as alternatives.
  3. [Appendix A, Figure 7 and Section 2] The simulation is an ILES without explicit dissipation, and Figure 7 shows that the width of the curvature PDF near its peak continues to change between 640^3 and 10,240^3, with the power-law tail moving to higher K. The high-curvature regions that scatter the smallest-rL CRs are therefore at least partly numerical. To support the statement that small-scale curvature structures are captured properly, please provide a numerical-convergence test of the curvature PDF itself (e.g., at several resolutions with a fixed injection scale) and/or a comparison with an explicit-dissipation run or a different Riemann solver/reconstruction scheme.
  4. [Section 3] All transport coefficients are taken from a single static snapshot of a single turbulence realization, and no statistical uncertainties are given. It would strengthen the paper to state how many independent large-scale eddies are sampled and to provide bootstrap errors or realization-to-realization scatter; as written, it is not possible to tell whether the 'almost constant' plateau is statistically distinguishable from a slowly varying slope.
minor comments (5)
  1. [Section 4] The sentence 'P(τ) ∼ τ^{−α} and α ∈ [−1.5, −2]' should read 'α ∈ [1.5, 2]'; as written the sign is inconsistent with the subsequent discussion of α < 2 and with Eq. (5).
  2. [Figure 6] The caption does not identify the colors or line styles corresponding to the different gyro-radii; please add a legend or an explicit description so the reader can associate each curve with the rL values listed in the text.
  3. [Footnote 7] The statement that no evidence is found for significant CR scattering by small-amplitude waves relies on follow-up work not presented in this Letter; either remove the claim or qualify it as preliminary and dependent on unpublished results.
  4. [Data Availability] Sharing data 'on reasonable request' is weak for a numerical paper of this type; consider depositing the magnetic-field snapshot or particle trajectories in a public repository to enable external verification of the convergence issue.
  5. [Title and Section 2] The title contains obvious typos ('T ransport' and 'T urbulence'), and 'δBrms/B0' is written with inconsistent spacing; these should be corrected during typesetting.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central transport results are direct simulation measurements, and the analytic relations (4)–(5) are consistency checks rather than fitted predictions.

full rationale

The paper's central claims—weak energy dependence of κ_z, broad power-law collision-time PDFs, and curvature-mediated scattering—are measured directly from the 10,240^3 particle trajectories and magnetic-field statistics. Equations (4)–(5) use the measured α and τ_max to estimate κ_z ≈ 0.1cL and then compare with the directly measured κ_z; this is an internal consistency check, not a parameter fit to κ_z, and it does not define κ_z. The Section 5.2 toy model is explicitly illustrative ('not meant to represent theoretical models of the measured PDFs') and is not load-bearing. Self-citations to Kempski et al. (2023) and Lemoine (2023) supply context and a name for resonant-curvature scattering, but the paper independently demonstrates the mechanism with its own tracked particle data (Figures 5–6), so the citations are not load-bearing. The authors' admitted resolution dependence of κ_z and the curvature PDF (Figure 7, Section 4) is a numerical-convergence caveat, not a circularity: the claim does not assume the conclusion, it reports unresolved resolution sensitivity. No circular reduction of a prediction to an input was found.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the statistical properties of high-curvature regions in an ideal-MHD ILES snapshot, the measured collision-time PDF parameters, and an illustrative toy model; no new particles or forces are introduced. The main empirical inputs are the power-law slope of the collision-time distribution and its cutoff, both extracted from the same simulation that produces the diffusion coefficients.

free parameters (3)
  • Power-law slope α of collision-time distribution = ≈ 1.5 to 2 (measured from simulation; ≈ 5/3 used in eq. 5)
    The slope of P(τ) is extracted from the simulated particle tracks (Figure 6) and directly determines via eq. (5) how strongly the largest excursions set the diffusion coefficient. The value α ≈ 5/3 is used to reproduce κz ≈ 0.1 cL.
  • Maximum collision time τz,max / cutoff = ≈ L/c
    The location of the power-law cutoff in P(τz) is taken from Figure 6 and enters eq. (5); the flatness of this cutoff with rL is the stated reason for the weak energy dependence of κz.
  • Curvature PDF slope β = ≈ 2.5
    Used in the illustrative toy model of Section 5.2 to derive P(λ) ~ λ^{-2}; taken from the simulation and from the cited literature (Yang et al. 2019; Kempski et al. 2023).
assumptions (5)
  • domain assumption The magnetic field is frozen (static snapshot) while test particles propagate; field evolution is neglected.
    Particles are integrated through a single static snapshot of the MHD turbulence (Section 2). In the ISM the field evolves on timescales much longer than light-crossing, but for the simulated box the subsonic turbulence evolves on timescales comparable to the eddy turnover time, which may be longer than 50 L/c. The paper does not quantify the effect of field evolution on the reported diffusion coefficients.
  • domain assumption The implicit large-eddy simulation with piecewise linear reconstruction, HLLD, and no explicit dissipation captures the curvature statistics relevant for CR scattering.
    The code uses ILES (Section 2). The paper shows the curvature PDF and κz are resolution-dependent (Figures 4, 7), so the numerical dissipation scale enters the physics; the claim that high-curvature structures are resolved relies on this assumption.
  • domain assumption Ideal MHD and test-particle approximation: the plasma is isothermal, and CRs do not feed back on the turbulence.
    Section 2 states isothermal ideal MHD; Section 5.2 notes CR feedback is missing. The authors acknowledge this is important for sub-GeV energies but assume it does not affect the >TeV transport studied here.
  • standard math Equation (3): κz = κ∥ (vB/c)^2 for fully magnetized particles.
    Derived in Section 3 from the definition vB and the assumption that particles are beads on wires. The paper shows it holds only for the smallest rL, so it is not a universal assumption but is used for interpretation.
  • ad hoc to paper The toy-model scaling λ(K) ~ 1/[P(K)K] and σ_eff(K) ~ K^{-1/2} in Section 5.2.
    These are illustrative assumptions to motivate P(λ) ~ λ^{-2}; the authors explicitly state the derivation is not a theoretical model and neglects clustering and trapping of particles.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Self-Similar Cosmic-Ray Transport in High-Resolution Magnetohydrodynamic Turbulence." pith.science (2026). https://pith.science/paper/47S6S6AC

@misc{pith2026250710651,
  author       = {Pith},
  title        = {Pith review of: Self-Similar Cosmic-Ray Transport in High-Resolution Magnetohydrodynamic Turbulence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/47S6S6AC}},
  note         = {Machine review of arXiv:2507.10651}
}
abstract

We study the propagation of cosmic rays (CRs) through a simulation of magnetohydrodynamic (MHD) turbulence at unprecedented resolution of $10{,}240^3$. We drive turbulence that is subsonic and super-Alfv\'enic, characterized by $\delta B_{\rm rms}/B_0=2$. The high resolution enables an extended inertial range such that the Alfv\'en scale $l_A$, where $\delta B (l_A)\approx B_0$, is well resolved. This allows us to properly capture how the cascade transitions from large amplitudes on large scales to small amplitudes on small scales. We find that sharp bends in the magnetic field are key mediators of particle transport even on small scales via resonant curvature scattering. We further find that particle scattering in the turbulence shows strong hints of self-similarity: (1) the diffusion has weak energy dependence over almost two decades in particle energy and (2) the particles' random walk exhibits a broad power-law distribution of collision times such that the diffusion is dominated by the rarest, long-distance excursions. Our results suggest that large-amplitude MHD turbulence can provide efficient scattering over a wide range of CR energies and may help explain many CR observations above a $\sim$TeV: the flattening of the B/C spectrum, the hardening of CR primary spectra and the weak dependence of arrival anisotropy on CR energy.

Figures

Figures reproduced from arXiv: 2507.10651 by the authors.

Figure 1
Figure 1. The shaded map shows the conditional proba￾bility density fV (δB|ℓ)—the fractional volume of pairs sep￾arated by a (isotropic) distance ℓ whose two-point incre￾ment has magnitude |δB(ℓ)|—for our highest-resolution run (L/∆x = 10,240). The thick magenta curve traces the me￾dian |δB| at each separation. Thinner magenta curves give the medians for lower-resolution simulations (L/∆x = 640 and 2560). Their departure from… view at source ↗
Figure 2
Figure 2. Left: Two-dimensional slice of the magnitude of current density normalized by the grid spacing in the 10,2403 simulation characterized by δBrms/B0 = 2 (Fielding et al., in prep.). Right: Trajectory of a CR with rL/L ≈ 7 × 10−4 in a 6403 sub-domain (represented by the yellow square in the left panel) with side lengths approximately twice the Alfv´en scale, lA. The trajectory is color-coded using the logarithm of the … view at source ↗
Figure 4
Figure 4. Diffusion coefficients parallel (blue) and per￾pendicular (orange) to the background guide field, and along the local magnetic field (pink) in our simulations with δBrms/B0 = 2 at different resolutions. κxy is multiplied by a factor of 3 for visualization purposes and we multiply κ∥ by the effective late-time ballistic velocity of field lines (see bottom panel of [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (4 more)
Figure 3
Figure 3. Figure 3: Running diffusion coefficients in the 10,2403 simulation with δBrms/B0 = 2. The diffusion coefficients are normalized by cL, where c is the speed of light and L is the size of the turbulent box. Top: running diffusion co￾efficient along the background guide field for p…
Figure 5
Figure 5. Figure 5: Track of a particle with gyro-radius rL/L ≈ 3×10−4 . Dark lines show running averages of tracked quan￾tities over a window 2π/Ω, faint lines show instantaneous values. Changes in the particle magnetic moment (top) are very abrupt and occur when the particle experiences…
Figure 6
Figure 6. Figure 6: PDFs of particle collision times P(τ c/L) multi￾plied by τ c/L (thus showing the probability per logarithmic interval). We consider two types of spatial scattering events: (1) v · bˆ changing sign (top panel) and (2) v · zˆ changing sign (middle panel), and we define t…
Figure 7
Figure 7. Figure 7: PDF of field-line curvature K normalized by the curvature K∗ that maximizes KP(K) (i.e. the most volume-filling logarithmic curvature bin) for the 6403 and 10,2403 simulations. Both PDFs are characterized by a K−1.5 power-law tail at high K. However, as resolution is i…

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Sub-sonic compressible magnetohydrodynamic turbulence I. Alfv\'enic and fast-magnetosonic injection, amplitude dependence, and compressibility effects

    physics.plasm-ph 2026-08 conditional novelty 6.0 of 10

    In decaying sub-sonic MHD turbulence simulations, the amplitude and plasma beta, not the injection wave type, set the turbulent density fluctuation level, while curvature and mirror statistics change dramatically with...

  2. Superdiffusion of cosmic rays in the vicinity of their accelerators and the resulting $\gamma$-ray emission

    astro-ph.HE 2026-07 conditional novelty 5.5 of 10

    Superdiffusion produces constant or r^{α-3} CR radial profiles near sources; the associated γ-ray morphology can distinguish it from normal diffusion with IACTs.

Reference graph

Works this paper leans on

83 extracted references · 11 canonical work pages · cited by 2 Pith papers

  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]

    V s U> ^ ƕ.> N > s;#mwQJ /B

    thebibliography [1] 20pt to REFERENCES 6pt =0pt -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 on reference command E...

    arXiv 2021

  4. [4]

    2019, , 122, 181102, 10.1103/PhysRevLett.122.181102

    Adriani , O., Akaike , Y., Asano , K., et al. 2019, , 122, 181102, 10.1103/PhysRevLett.122.181102

    doi:10.1103/physrevlett.122.181102 2019

  5. [5]

    2022, , 129, 251103, 10.1103/PhysRevLett.129.251103

    ---. 2022, , 129, 251103, 10.1103/PhysRevLett.129.251103

    doi:10.1103/physrevlett.129.251103 2022

  6. [6]

    2023, , 130, 171002, 10.1103/PhysRevLett.130.171002

    ---. 2023, , 130, 171002, 10.1103/PhysRevLett.130.171002

    doi:10.1103/physrevlett.130.171002 2023

  7. [7]

    2015 a , , 114, 171103, 10.1103/PhysRevLett.114.171103

    Aguilar , M., Aisa , D., Alpat , B., et al. 2015 a , , 114, 171103, 10.1103/PhysRevLett.114.171103

    doi:10.1103/physrevlett.114.171103 2015

  8. [8]

    2015 b , , 115, 211101, 10.1103/PhysRevLett.115.211101

    ---. 2015 b , , 115, 211101, 10.1103/PhysRevLett.115.211101

    doi:10.1103/physrevlett.115.211101 2015

Show all 83 references
  1. [9]

    2016, , 117, 231102, 10.1103/PhysRevLett.117.231102

    Aguilar , M., Ali Cavasonza , L., Ambrosi , G., et al. 2016, , 117, 231102, 10.1103/PhysRevLett.117.231102

    2016 doi

  2. [10]

    2021, , 894, 1, https://doi.org/10.1016/j.physrep.2020.09.003

    Aguilar, M., Ali Cavasonza , L., Ambrosi, G., et al. 2021, , 894, 1, https://doi.org/10.1016/j.physrep.2020.09.003

    2021 doi

  3. [11]

    2024, , 109, L121101, 10.1103/PhysRevD.109.L121101

    Alemanno , F., Altomare , C., An , Q., et al. 2024, , 109, L121101, 10.1103/PhysRevD.109.L121101

    2024 doi

  4. [12]

    2018, , 62, 2731, 10.1016/j.asr.2017.04.019

    Amato , E., & Blasi , P. 2018, , 62, 2731, 10.1016/j.asr.2017.04.019

    2018 doi

  5. [13]

    2019, , 5, eaax3793, 10.1126/sciadv.aax3793

    An , Q., Asfandiyarov , R., Azzarello , P., et al. 2019, , 5, eaax3793, 10.1126/sciadv.aax3793

    2019 doi

  6. [14]

    Andersen, E. S. 1954, Mathematica Scandinavica, 2, 195. http://eudml.org/doc/165543

    1954

  7. [15]

    C., & Linzer , N

    Armillotta , L., Ostriker , E. C., & Linzer , N. B. 2025, arXiv e-prints, arXiv:2507.00120, 10.48550/arXiv.2507.00120

    2025 doi arXiv

  8. [16]

    2022, , 928, 112, 10.3847/1538-4357/ac56e1

    Bai , X.-N. 2022, , 928, 112, 10.3847/1538-4357/ac56e1

    2022 doi

  9. [17]

    C., Plotnikov , I., & Stone , J

    Bai , X.-N., Ostriker , E. C., Plotnikov , I., & Stone , J. M. 2019, , 876, 60, 10.3847/1538-4357/ab1648

    2019 doi

  10. [18]

    H., et al

    Bandyopadhyay , R., Yang , Y., Matthaeus , W. H., et al. 2020, , 893, L25, 10.3847/2041-8213/ab846e

    2020 doi

  11. [19]

    R., Federrath , C., Klessen , R

    Beattie , J. R., Federrath , C., Klessen , R. S., Cielo , S., & Bhattacharjee , A. 2024, arXiv e-prints, arXiv:2405.16626, 10.48550/arXiv.2405.16626

    2024 doi arXiv

  12. [20]

    1991, Plasma Physics via Computer Simulation (1st ed.)

    Birdsall , C., & Langdon , A. 1991, Plasma Physics via Computer Simulation (1st ed.)

    1991

  13. [21]

    2012, , 2012, 011, 10.1088/1475-7516/2012/01/011

    Blasi , P., & Amato , E. 2012, , 2012, 011, 10.1088/1475-7516/2012/01/011

    2012 doi

  14. [22]

    Blasi , P., Amato , E., & Serpico , P. D. 2012, , 109, 061101, 10.1103/PhysRevLett.109.061101

    2012 doi

  15. [23]

    Boris, J. P. 1970, Proceeding of Fourth Conference on Numerical Simulations of Plasmas

    1970

  16. [24]

    1990, , 195, 127, 10.1016/0370-1573(90)90099-N

    Bouchaud , J.-P., & Georges , A. 1990, , 195, 127, 10.1016/0370-1573(90)90099-N

    1990 doi

  17. [25]

    S., Hopkins , P

    Butsky , I. S., Hopkins , P. F., Kempski , P., et al. 2024, , 528, 4245, 10.1093/mnras/stae276

    2024 doi

  18. [26]

    Chandran, B. D. G. 2000, , 85, 4656, 10.1103/PhysRevLett.85.4656

    2000 doi

  19. [27]

    2010, , 82, 023009, 10.1103/PhysRevD.82.023009

    Cowsik , R., & Burch , B. 2010, , 82, 023009, 10.1103/PhysRevD.82.023009

    2010 doi

  20. [28]

    2022, Science Bulletin, 67, 2162, 10.1016/j.scib.2022.10.002

    Dampe Collaboration . 2022, Science Bulletin, 67, 2162, 10.1016/j.scib.2022.10.002

    2022 doi

  21. [29]

    2022, , 8, eabn7627, 10.1126/sciadv.abn7627

    Dong , C., Wang , L., Huang , Y.-M., et al. 2022, , 8, eabn7627, 10.1126/sciadv.abn7627

    2022 doi

  22. [30]

    J., Reichherzer , P., Bott , A

    Ewart , R. J., Reichherzer , P., Bott , A. F. A., Kunz , M. W., & Schekochihin , A. A. 2024, , 532, 2098, 10.1093/mnras/stae1578

    2024 doi

  23. [31]

    J., & Goldreich , P

    Farmer , A. J., & Goldreich , P. 2004, , 604, 671, 10.1086/382040

    2004 doi

  24. [32]

    B., Ripperda , B., & Philippov , A

    Fielding , D. B., Ripperda , B., & Philippov , A. A. 2023, , 949, L5, 10.3847/2041-8213/accf1f

    2023 doi

  25. [33]

    S., De La Torre Luque , P., & Gabici , S

    Fornieri , O., Gaggero , D., Cerri , S. S., De La Torre Luque , P., & Gabici , S. 2021, , 502, 5821, 10.1093/mnras/stab355

    2021 doi

  26. [34]

    M., Haverkorn , M., Staveley-Smith , L., et al

    Gaensler , B. M., Haverkorn , M., Staveley-Smith , L., et al. 2005, , 307, 1610, 10.1126/science.1108832

    2005 doi

  27. [35]

    K., Kunz , M

    Galishnikova , A. K., Kunz , M. W., & Schekochihin , A. A. 2022, , 12, 041027, 10.1103/PhysRevX.12.041027

    2022 doi

  28. [36]

    1995, , 438, 763, 10.1086/175121

    Goldreich , P., & Sridhar , S. 1995, , 438, 763, 10.1086/175121

    1995 doi

  29. [37]

    W., & Beckwith , K

    Grete , P., O'Shea , B. W., & Beckwith , K. 2023, , 942, L34, 10.3847/2041-8213/acaea7

    2023 doi

  30. [38]

    2015, Magnetic Fields in the Milky Way, ed

    Haverkorn, M. 2015, Magnetic Fields in the Milky Way, ed. A. Lazarian, E. M. de Gouveia Dal Pino, & C. Melioli (Berlin, Heidelberg: Springer Berlin Heidelberg), 483--506, 10.1007/978-3-662-44625-6_17

    2015 doi

  31. [39]

    2023, , 672, A21, 10.1051/0004-6361/202245223

    Heesen , V., de Gasperin , F., Schulz , S., et al. 2023, , 672, A21, 10.1051/0004-6361/202245223

    2023 doi

  32. [40]

    2019, , 882, 3, 10.3847/1538-4357/ab328a

    Holcomb , C., & Spitkovsky , A. 2019, , 882, 3, 10.3847/1538-4357/ab328a

    2019 doi

  33. [41]

    F., Squire , J., Butsky , I

    Hopkins , P. F., Squire , J., Butsky , I. S., & Ji , S. 2022, , 517, 5413, 10.1093/mnras/stac2909

    2022 doi

  34. [42]

    2022, , 512, 2111, 10.1093/mnras/stac319

    Hu , Y., Lazarian , A., & Xu , S. 2022, , 512, 2111, 10.1093/mnras/stac319

    2022 doi

  35. [43]

    2012, , 109, 265002, 10.1103/PhysRevLett.109.265002

    Huang , Y.-M., & Bhattacharjee , A. 2012, , 109, 265002, 10.1103/PhysRevLett.109.265002

    2012 doi

  36. [44]

    R., Leahy , J

    Jaffe , T. R., Leahy , J. P., Banday , A. J., et al. 2010, , 401, 1013, 10.1111/j.1365-2966.2009.15745.x

    2010

  37. [45]

    Ji , S., Squire , J., & Hopkins , P. F. 2022, , 513, 282, 10.1093/mnras/stac895

    2022 doi

  38. [46]

    B., Quataert , E., et al

    Kempski , P., Fielding , D. B., Quataert , E., et al. 2023, , 525, 4985, 10.1093/mnras/stad2609

    2023 doi

  39. [47]

    2022, , 514, 657, 10.1093/mnras/stac1240

    Kempski , P., & Quataert , E. 2022, , 514, 657, 10.1093/mnras/stac1240

    2022 doi

  40. [48]

    Kulsrud , R., & Pearce , W. P. 1969, , 156, 445, 10.1086/149981

    1969 doi

  41. [49]

    M., & Cesarsky , C

    Kulsrud , R. M., & Cesarsky , C. J. 1971, , 8, 189

    1971

  42. [50]

    2021, , 923, 53, 10.3847/1538-4357/ac2de9

    Lazarian , A., & Xu , S. 2021, , 923, 53, 10.3847/1538-4357/ac2de9

    2021 doi

  43. [51]

    2014, , 784, 38, 10.1088/0004-637X/784/1/38

    Lazarian , A., & Yan , H. 2014, , 784, 38, 10.1088/0004-637X/784/1/38

    2014 doi

  44. [52]

    2025, , 979, 34, 10.3847/1538-4357/ad8eb3

    Lemmerz , R., Shalaby , M., Pfrommer , C., & Thomas , T. 2025, , 979, 34, 10.3847/1538-4357/ad8eb3

    2025 doi

  45. [53]

    2023, , 89, 175890501, 10.1017/S0022377823000946

    Lemoine , M. 2023, , 89, 175890501, 10.1017/S0022377823000946

    2023 doi

  46. [54]

    Liang , N., & Oh , S. P. 2025, arXiv e-prints, arXiv:2503.10747, 10.48550/arXiv.2503.10747

    2025 doi

  47. [55]

    2025, arXiv e-prints, arXiv:2505.18155, 10.48550/arXiv.2505.18155

    L \"u bke , J., Reichherzer , P., Aerdker , S., et al. 2025, arXiv e-prints, arXiv:2505.18155, 10.48550/arXiv.2505.18155

    2025 doi

  48. [56]

    2021, , 28, 092106, 10.1063/5.0059017

    Majeski , S., Ji , H., Jara-Almonte , J., & Yoo , J. 2021, , 28, 092106, 10.1063/5.0059017

    2021 doi

  49. [57]

    Narayan , R., & Medvedev , M. V. 2001, , 562, L129, 10.1086/338325

    2001 doi

  50. [58]

    1993, , 262, 953, 10.1093/mnras/262.4.953

    Ohno , H., & Shibata , S. 1993, , 262, 953, 10.1093/mnras/262.4.953

    1993 doi

  51. [59]

    V., Blackburn , G., Lomholt , M

    Palyulin , V. V., Blackburn , G., Lomholt , M. A., et al. 2019, , 21, 103028, 10.1088/1367-2630/ab41bb

    2019 doi

  52. [60]

    M., Sironi , L., & Giannios , D

    Petropoulou , M., Christie , I. M., Sironi , L., & Giannios , D. 2018, , 475, 3797, 10.1093/mnras/sty033

    2018 doi

  53. [61]

    2024, , 529, L13, 10.1093/mnrasl/slad192

    Pezzi , O., & Blasi , P. 2024, , 529, L13, 10.1093/mnrasl/slad192

    2024 doi

  54. [62]

    2024, , 692, A20, 10.1051/0004-6361/202349005

    Recchia , S., & Gabici , S. 2024, , 692, A20, 10.1051/0004-6361/202349005

    2024 doi

  55. [63]

    2001, A Guide to First-Passage Processes (Cambridge University Press)

    Redner, S. 2001, A Guide to First-Passage Processes (Cambridge University Press)

    2001

  56. [64]

    Reichherzer , P., Bott , A. F. A., Ewart , R. J., et al. 2025, , 9, 438, 10.1038/s41550-024-02442-1

    2025 doi

  57. [65]

    2023, , 31, 4, 10.1007/s00159-023-00149-2

    Ruszkowski , M., & Pfrommer , C. 2023, , 31, 4, 10.1007/s00159-023-00149-2

    2023 doi

  58. [66]

    A., Cowley , S

    Schekochihin , A. A., Cowley , S. C., Taylor , S. F., Maron , J. L., & McWilliams , J. C. 2004, , 612, 276, 10.1086/422547

    2004 doi

  59. [67]

    2025, , 134, 045201, 10.1103/PhysRevLett.134.045201

    Schroer , B., Caprioli , D., & Blasi , P. 2025, , 134, 045201, 10.1103/PhysRevLett.134.045201

    2025 doi

  60. [68]

    2021, , 908, 206, 10.3847/1538-4357/abd02d

    Shalaby , M., Thomas , T., & Pfrommer , C. 2021, , 908, 206, 10.3847/1538-4357/abd02d

    2021 doi

  61. [69]

    2019, , 881, L27, 10.3847/2041-8213/ab379d

    Shalchi , A. 2019, , 881, L27, 10.3847/2041-8213/ab379d

    2019 doi

  62. [70]

    1971, , 170, 265, 10.1086/151210

    Skilling , J. 1971, , 170, 265, 10.1086/151210

    1971 doi

  63. [71]

    M., & Gardiner , T

    Stone , J. M., & Gardiner , T. 2009, , 14, 139, 10.1016/j.newast.2008.06.003

    2009 doi

  64. [72]

    M., Mullen , P

    Stone , J. M., Mullen , P. D., Fielding , D., et al. 2024, arXiv e-prints, arXiv:2409.16053, 10.48550/arXiv.2409.16053

    2024 doi arXiv

  65. [73]

    W., Moskalenko , I

    Strong , A. W., Moskalenko , I. V., & Ptuskin , V. S. 2007, , 57, 285, 10.1146/annurev.nucl.57.090506.123011

    2007

  66. [74]

    2024, arXiv e-prints, arXiv:2409.08592, 10.48550/arXiv.2409.08592

    Sun , X., Bai , X.-N., & Zhao , X. 2024, arXiv e-prints, arXiv:2409.08592, 10.48550/arXiv.2409.08592

    2024 doi arXiv

  67. [75]

    E., & Ornstein , L

    Uhlenbeck , G. E., & Ornstein , L. S. 1930, , 36, 823, 10.1103/PhysRev.36.823

    1930 doi

  68. [76]

    2018, , 868, 36, 10.3847/1538-4357/aae840

    Xu , S., & Lazarian , A. 2018, , 868, 36, 10.3847/1538-4357/aae840

    2018 doi

  69. [77]

    2020, , 894, 63, 10.3847/1538-4357/ab8465

    ---. 2020, , 894, 63, 10.3847/1538-4357/ab8465

    2020 doi

  70. [78]

    2004, , 614, 757, 10.1086/423733

    Yan , H., & Lazarian , A. 2004, , 614, 757, 10.1086/423733

    2004 doi

  71. [79]

    2008, , 673, 942, 10.1086/524771

    ---. 2008, , 673, 942, 10.1086/524771

    2008 doi

  72. [80]

    H., et al

    Yang , Y., Wan , M., Matthaeus , W. H., et al. 2019, , 26, 072306, 10.1063/1.5099360

    2019 doi

  73. [81]

    H., & Lazarian , A

    Yuen , K. H., & Lazarian , A. 2020, , 898, 66, 10.3847/1538-4357/ab9360

    2020 doi

  74. [82]

    Zweibel, E. G. 2013, , 20, 055501, 10.1063/1.4807033

    2013 doi

  75. [83]

    Zweibel , E. G. 2020, , 890, 67, 10.3847/1538-4357/ab67bf

    2020 doi

Pith tools

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