Pith. sign in

REVIEW 3 major objections 5 minor 94 references

Anisotropic diffusion of high-energy cosmic rays in magnetohydrodynamic turbulence

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read In a long numerical evolution of cosmic-ray electrons in MHD turbulence, the paper finds that the anisotropic diffusion scalings persist and the energy density falls as $M_A^{-6}$.

desk verdict The paper's headline scalings are mostly hard-wired inputs, and the effective radius transformation appears inverted, so the main discovery claims do not survive scrutiny; still, the parameter study is substantive enough to warrant a referee rather than a desk reject. read the letter →

arxiv 2501.04986 v1 pith:WOROJTOR submitted 2025-01-09 astro-ph.HE

Na-Na Gao (Xiangtan Univ.) , Jian-Fu Zhang (Xiangtan Univ.) This is my paper · ORCID
classification astro-ph.HE
keywords cosmic-raytransportanisotropicdiffusionmagnetohydrodynamic(MHD)turbulenceAlfvénMachnumberpulsarwindnebulaTeVhalosradiativelossesFokker-Planck
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 tries to establish that anisotropic diffusion—fast along a mean magnetic field, slow across it—is an intrinsic, long-lived description of high-energy cosmic-ray transport near pulsar wind nebulae. Using a numerical Fokker-Planck transport simulation with sub-grid diffusion coefficients, the authors evolve injected electron populations in a turbulent magnetized medium, treating the magnetization parameter $M_A$ and the turbulence spectral index $\gamma$ as free parameters. They report that the two input diffusion laws remain intact after a 3 kyr evolution: $D_\parallel \propto E^{2-\gamma}$ and $D_\perp/D_\parallel \propto M_A^\alpha$ with $\alpha=4$ sub-Alfvénic and $\alpha=3$ super-Alfvénic. They also find that the volume-averaged cosmic-ray energy density obeys $U_{CR} \propto M_A^{-6.0\pm0.5}$, independent of particle energy and radiative losses, while losses convert normal PeV diffusion into sub-diffusion. If these results hold, the morphology and brightness of TeV halos become observable probes of the magnetization and turbulence spectrum of the surrounding interstellar medium.

What carries the argument

The machinery is the ratio law $D_\perp/D_\parallel \approx M_A^\alpha$, with $\alpha=4$ for sub-Alfvénic and $\alpha=3$ for super-Alfvénic turbulence, joined to the energy law $D_\parallel = D_0(E/m_e c^2)^\delta$ with $\delta=2-\gamma$. The paper feeds these laws into a Fokker-Planck transport equation through an anisotropic effective radius $r=\sqrt{(x/\sqrt{\chi})^2+y^2+z^2}$ with $\chi=M_A^\alpha$, evolves injected electrons for 3 kyr under synchrotron and inverse-Compton cooling, and then re-extracts $D_\parallel$ and $D_\perp$ from ensemble-averaged squared displacements. The argument stands on the fact that the returned exponents match the input exponents, and that the scan over $M_A$ yields an emergent $U_{CR}\propto M_A^{-6.0\pm0.5}$.

What would settle it

Run a collisionless test-particle simulation in which the MHD turbulent field is evolved self-consistently, measuring $D_\parallel$ and $D_\perp$ for 1 GeV to 10 PeV electrons in sub- and super-Alfvénic regimes with radiative cooling; if the recovered exponent of $D_\perp/D_\parallel$ vs $M_A$ departs systematically from $\alpha=4,3$, or the recovered $U_{CR}$-vs-$M_A$ slope departs from $-6$, the claimed universality fails. Observationally, compare halo brightness around pulsars with independently estimated $B_0$ and check whether the inferred $M_A$ agrees with the measured magnetization.

Watch

Extended reading notes

Core claim

The authors' central claim is that the anisotropic-diffusion relation $D_\perp/D_\parallel \sim M_A^\alpha$ and the energy scaling $D_\parallel \sim E^{2-\gamma}$ are preserved through a long cosmic-ray evolution, so they can be treated as intrinsic properties of transport in turbulent magnetic fields. On top of that, they report a new volume-averaged power law $U_{CR} \sim M_A^{-6.0\pm0.5}$ that is independent of particle energy and of radiative losses, with the spatial profile following $U_{CR}\propto r^{-6/5}$ near the source and steepening to $r^{-11/5}$ when PeV electrons cool, while radiative cooling turns PeV transport from normal diffusion into sub-diffusion with $d^2\propto t^{4/5}$. They further show that halo morphology is oval when viewed across the mean field and round when viewed along it, with the anisotropy elongation controlled by $M_A$ through the exponent $\eta=M_A^{\alpha/2}$. Thus the paper converts TeV halo shape and brightness into readouts of turbulence magnetization and spectral index.

Load-bearing premise

The load-bearing premise is that the two input scalings—parallel diffusion growing as energy to a power and the perpendicular-to-parallel ratio growing as the fourth or third power of the magnetization—are correct descriptions of real turbulent transport, since the simulations impose those scalings as sub-grid inputs and never evolve the turbulent magnetic field itself.

Editorial extensions

If this is right

  • The recovered $D_\perp/D_\parallel \sim M_A^{4}$ sub-Alfvénic and $\sim M_A^{3}$ super-Alfvénic relations, independent of energy and radiative losses, justify using these laws in interpretive models of TeV halos.
  • The averaged energy density $U_{CR}\propto M_A^{-6.0\pm0.5}$ gives a quantitative link between halo brightness and the Alfvén Mach number of the surrounding medium.
  • The near-source spatial profile $U_{CR}\propto r^{-6/5}$, steepened to $r^{-11/5}$ by losses, provides a template for fitting observed halo surface-brightness profiles.
  • Radiative losses convert PeV transport from normal diffusion to sub-diffusion with $d^2\propto t^{4/5}$, so high-energy halo morphology encodes cooling as well as turbulence.
  • The anisotropy exponent $\eta=M_A^{\alpha/2}$ connects the oval aspect ratio of a halo to the magnetization, allowing viewing angle and turbulence properties to be separated with multi-angle observations.

Reading between the lines

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

  • A self-consistent test-particle simulation that actually evolves the turbulent field, rather than prescribing the diffusion scalings, would show whether $\alpha=4,3$ and the $M_A^{-6}$ law are emergent or merely inherited from the input assumptions; the paper itself does not evolve the turbulence.
  • The energy independence of the $M_A$ scaling suggests that the magnetization could be inferred from ratios of halo brightness at two energies rather than from absolute fluxes, reducing model dependence in future observations.
  • Because the study uses electrons and only synchrotron and inverse-Compton cooling, whether the $M_A^{-6}$ correlation survives for protons with hadronic losses is a testable next step the paper leaves open.
  • Observational maps of several pulsar halos with independently estimated magnetic-field strengths could test the predicted $U_{CR}$-versus-$M_A$ slope directly, since $M_A$ enters through the mean field $B_0$.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The manuscript presents CRIPTIC simulations of high-energy electron transport in a pulsar-wind-nebula environment, using prescribed anisotropic diffusion coefficients D_∥ = D0(E/mec^2)^δ and D_⊥/D_∥ = M_A^α with α = 4 (sub-Alfvénic) and α = 3 (super-Alfvénic). Three groups of runs vary radiative losses, the turbulence spectral index γ, and the Alfvén Mach number M_A. The authors report power-law radial and spectral energy-density profiles (UCR ∝ r^-6/5, UCR ∝ 1/E without losses; steeper with losses), M_A- and viewing-angle-dependent morphologies, slow-to-normal diffusion transitions, an averaged relation UCR ∝ M_A^-6.0±0.5, and persistence of the input diffusion scalings over 3 kyr. They interpret the last two as intrinsic properties connecting CR transport to MHD turbulence.

Significance. If the headline scalings were genuinely emergent, the paper would be valuable: TeV halo morphology and the UCR–M_A relation could then be used as observational probes of turbulence magnetization and spectral index. The paper has strengths: it uses a documented transport code, runs a large parameter sweep (90 simulations), includes realistic electron cooling (synchrotron plus inverse Compton in the Klein-Nishina regime), and provides clear diagnostics of morphology and anisotropy. The qualitative findings that radiative losses suppress anisotropy, steepen radial profiles, and produce sub-diffusive behavior for PeV electrons are plausible and potentially useful. However, the two central claims are not supported by the simulation design: the D_⊥/D_∥ vs. M_A scaling is an input rather than a discovery, and the UCR ∝ M_A^-6 law is entangled with an inverted coordinate transformation. The paper's contribution is therefore substantially reduced, and the abstract's 'intrinsic property' conclusion is not justified.

major comments (3)
  1. [§2.2, §4.4, Appendix A Step 2] Appendix A Step 2 hard-wires χ = D_⊥/D_∥ = M_A^α as a CRIPTIC input (α = 4 sub-Alfvénic, α = 3 super-Alfvénic), and the turbulent magnetic field is never evolved. Section 4.4 and Fig. 8(c) then report that the measured D_⊥/D_∥ follows M_A^4.0 and M_A^3.0 and call this a 'numerical finding' and an 'inherent property.' Because the particle deviations used to compute D_∥ and D_⊥ are generated by the very same prescribed diffusion tensor, the agreement is a consistency check of the code, not an independent confirmation. The same objection applies to the persistence of D_∥ ∼ E^δ: Eq. (2) is the prescribed parallel diffusion law. The abstract's claim that the distribution law is an intrinsic property therefore rests on circular reasoning.
  2. [§4.1, Eq. (1), footnote 3] The effective radius is defined as r² = x²/χ + y² + z² with χ = M_A^α. For Eq. (1) with D_⊥ = χD_∥, the change of variables that makes the spatial operator isotropic is X = x, Y = y/√χ, Z = z/√χ, giving R_iso² = x² + (y² + z²)/χ; scaling the parallel coordinate x by 1/√χ instead is the inverse transformation. Consequently, the UCR(r) profiles in Figs. 1–2 and the volume-averaged UCR in Fig. 8(a) are computed in a distorted metric, and the reported UCR ∝ M_A^-6.0±0.5 law (Summary item 4) is not established as a physical scaling. The radial power-law indices UCR ∝ r^-6/5 and r^-11/5 are likewise affected by this coordinate choice.
  3. [§4.4, Fig. 8(a)] The averaging procedure that produces UCR for each M_A is not defined, and the M_A^-6.0±0.5 scaling is not derived from the transport solution. For a fixed-time anisotropic Gaussian profile with D_⊥ = χD_∥ and χ = M_A^α, the volume-averaged density scales roughly as χ^-1 = M_A^-α, i.e., M_A^-4 (sub-Alfvénic) and M_A^-3 (super-Alfvénic); the additional steepening to M_A^-6 is unexplained and may be an artifact of the inverted effective radius in §4.1. As it stands, the abstract's statement that the M_A^-6 law is independent of energy and radiative losses is not supported.
minor comments (5)
  1. [§3, Appendix A] The code name is spelled 'CRIPTIC' in the main text but 'CRIPPTIC' in Appendix A; please standardize the spelling.
  2. [§4.1.1, §4.3] There are grammatical slips such as 'from which can see that' (Sect. 4.1.1) and 'fast/normal-diffusion diffusion' (Sect. 4.3); these should be corrected.
  3. [§4.1, footnote 3] The footnote justifies the coordinate change by reference to a standard Green's function solution but does not show the transformation; please provide an explicit derivation from Eq. (1), since the current definition is directly related to the major issue above.
  4. [§4.4, Fig. 8(a)] Please specify exactly how UCR is averaged (over which volume or radial range, and at which time) and report fit uncertainties for the power-law indices β and ε; the error bars shown are standard deviations of UCR, not uncertainties on the fitted slopes.
  5. [§2.2] When presenting Eq. (3), the text should distinguish more clearly between theoretically predicted scalings and numerically verified scalings from the cited test-particle simulations, so that the input status of this relation in the present work is transparent.

Circularity Check

3 steps flagged · score 8.0 of 10

Central 'emergent' scalings are pre-installed inputs: CRIPTIC is initialized with χ = M_A^α and D_∥ = D0(E/m_e c^2)^δ, and the paper then reports these same laws as rediscovered numerical findings; UCR ∼ M_A^-6 also inherits an M_A-dependent coordinate rescaling.

  1. fitted input called prediction [Sect. 4.4, Fig. 8(c); Appendix A Step 2]
    "We note that the CRIPTIC code parameterizes the ratio of D⊥/D∥ as χ. In this work, we define χ = Mα A ... As seen in Fig. 8(c), the ratio D⊥/D∥ is related to the magnetization parameter MA by the power-law relations of D⊥/D∥ ∼ M4.0 A in the sub-Alfvénic regime, and D⊥/D∥ ∼ M3.0 A in the super-Alfvénic one. This numerical finding can well match the initial setting of D⊥,0/D∥,0 ∼ Mα A"

    CRIPTIC is initialized with exactly the law being 'recovered': χ ≡ D⊥/D∥ = M_A^α, with α = 4 (sub-Alfvénic) and α = 3 (super-Alfvénic), as input sub-grid diffusion coefficients. The simulation then transports particles with this prescribed anisotropic diffusion tensor, and the reported D⊥/D∥ is measured from the resulting trajectories via D∥ = d∥²/2t and D⊥ = d⊥²/2t. Recovering D⊥/D∥ ∼ M_A^4.0 and M_A^3.0 is therefore guaranteed by construction. Calling this a numerical finding and an inherent property of particle transport is tautological rather than an emergent result.

  2. fitted input called prediction [Sect. 4.4, Fig. 8(d); Appendix A Step 1]
    "Since CRIPTIC takes the sub-grid values for diffusion coefficients as input parameters, D∥ needs to be taken as the baseline value. With the energy-dependent parallel diffusion coefficient D∥ = D0(E/mec2)δ ... we first obtain the initial parallel diffusion coefficient D∥,0."

    The energy-dependent parallel diffusion law D∥ = D0(E/m_e c^2)^δ is an input parameter of the code, not a prediction. Fig. 8(d) then reports that D∥ and D⊥ 'both present well a power-law relation of D⊥,∥ ∼ Eδ = E2−γ = E1/3'. Because CRIPTIC advances particles with exactly this sub-grid D∥(E) and with an energy-independent χ, the recovered E^{1/3} scaling of the ensemble-averaged diffusion coefficients is the input relation re-measured from the simulated trajectories, not an independent numerical discovery.

1 more flagged steps
  1. self definitional [Sect. 4.1 and Sect. 4.4, Fig. 8(a)]
    "The effective radius of particles is defined as r = q (x/√χ)2 +y2 + z2, with the factor χ = Mα A corresponding to the anisotropy level ... The finding of UCR ∝ M−β A, where the index β is in the range of [5.50, 6.44] ... approximated to UCR ∼ M−6.0 A"

    The variable whose power-law profile is measured, r, is defined in terms of the input ratio χ = M_A^α, with x rescaled as x/√χ. Since the paper reports UCR ∝ r^{−6/5}, any radial power law automatically acquires an M_A-dependent normalization when expressed in physical coordinates; averaging over this χ-dependent coordinate then produces UCR ∼ M_A^{−6}. The claimed 'intrinsic' M_A power law and its stated independence from energy and radiative losses are thus built into the chosen metric and the pre-installed input scalings, rather than being an independent emergent property of the transport evolution.

full rationale

The paper's central 'robust connection' claim reduces to its own setup. Appendix A Step 1 fixes D∥ = D0(E/m_e c^2)^δ as a sub-grid input; Step 2 fixes the anisotropy ratio χ = D⊥/D∥ = M_A^α, with α = 4 and 3, as the link to the Alfvén Mach number. The code never evolves the turbulent magnetic field or the diffusion coefficients themselves; it advects particles through a prescribed diffusion tensor. Consequently, Figs. 8(c) and 8(d), which 'demonstrate' that D⊥/D∥ ∼ M_A^α and D∥ ∼ E^δ maintain their features after 3 kyr, are circular: the measured ensemble-averaged diffusion coefficients are statistically forced by the input tensor to reproduce those scalings. The UCR ∼ M_A^{−6} relation is similarly entangled with the definition r = sqrt((x/√χ)^2 + y^2 + z^2), where χ is the prescribed M_A^α; the coordinate rescaling introduces an M_A dependence into the very quantity being averaged, so the claimed universal power law is not independent of the model inputs. These are not cases of benign self-citation; no load-bearing argument relies on the authors' own prior work. The non-circular portions are the loss-modified spectral/spatial profiles (e.g., the r^{−6/5} → r^{−11/5} and E^{−6/5} → E^{−8/5} transitions), which follow from solving the transport equation under cooling and are not predetermined by the input scalings alone. Because the headline claims of an 'intrinsic property' reducing to pre-installed inputs, the circularity score is 8.

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

The model's turbulence physics consists entirely of two power-law inputs from prior literature: D_parallel(E) and D_perp/D_parallel(MA). The simulations vary the parameters of these laws but do not generate or evolve MHD turbulence, so the paper's universal laws are outputs of an assumed transport model rather than independent constraints on turbulence.

free parameters (5)
  • Magnetization parameter MA = 0.21 to 3 (regulated via B0)
    Free parameter in Groups A-C; controls D_perp/D_parallel = MA^alpha and the spatial extent of CRs; all headline scalings depend on it.
  • Turbulence spectral index gamma (diffusion index delta = 2-gamma) = gamma in [1.0, 1.9]
    Free parameter in Group B; sets D_parallel proportional to E^(2-gamma); the paper also labels cases gamma = 5/3, 3/2, 1.0 in figures.
  • Anisotropy index alpha = 4.0 (sub-Alfvénic), 3.0 (super-Alfvénic)
    Chosen from Lazarian & Yan (2014), not measured here; because it is inserted as chi = MA^alpha, the later recovery of the same law is circular.
  • Diffusion normalization D0 = 1.0e28 cm^2/s
    Adopted from interstellar diffusion measurements (Heesen et al. 2019); sets the absolute scale of all diffusion coefficients.
  • Injection spectral index q = 2.2
    Assumed for the power-law electron injection in Group A; affects the UCR(E) slope but not the claimed MA dependence.
assumptions (5)
  • domain assumption The Fokker-Planck equation (Eq. 1) with parallel and perpendicular spatial diffusion, energy diffusion, and a point source describes CR transport near a pulsar wind nebula.
    Invoked in Sect. 2 without derivation; underpins the entire simulation setup.
  • domain assumption D_parallel = D0(E/mec^2)^delta with delta in [0.1, 1.0] (Eq. 2).
    Input from quasi-linear theory and propagation codes; the simulation does not derive it from MHD turbulence.
  • domain assumption D_perp/D_parallel ~ MA^alpha with alpha = 4 (sub-Alfvénic) and alpha = 3 (super-Alfvénic) (Eq. 3).
    Taken from Yan & Lazarian (2008) and Lazarian & Yan (2014); set as chi = MA^alpha in Appendix A Step 2, making the subsequent confirmation circular.
  • domain assumption The effective radius r = sqrt((x/sqrt(chi))^2 + y^2 + z^2) provides an isotropic Green's function coordinate for Eq. (1).
    Footnote 3 asserts this; for D_perp = chi*D_parallel the standard isotropizing scaling is y/sqrt(chi) and z/sqrt(chi), not x/sqrt(chi), so this assumption appears incorrect.
  • domain assumption Radiative losses for electrons are limited to synchrotron and inverse-Compton scattering in the Klein-Nishina regime.
    Reasonable for electrons; stated in Sect. 3 and Appendix A; the paper excludes proton cooling channels by design.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Anisotropic diffusion of high-energy cosmic rays in magnetohydrodynamic turbulence." pith.science (2026). https://pith.science/paper/WOROJTOR

@misc{pith2026250104986,
  author       = {Pith},
  title        = {Pith review of: Anisotropic diffusion of high-energy cosmic rays in magnetohydrodynamic turbulence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WOROJTOR}},
  note         = {Machine review of arXiv:2501.04986}
}
read the original abstract

The origin of cosmic rays (CRs) and how they propagate remain unclear. Studying the propagation of CRs in magnetohydrodynamic (MHD) turbulence can help to comprehend many open issues related to CR origin and the role of turbulent magnetic fields. To comprehend the phenomenon of slow diffusion in the near-source region, we study the interactions of CRs with the ambient turbulent magnetic field to reveal their universal laws. We numerically study the interactions of CRs with the ambient turbulent magnetic field, considering pulsar wind nebula as a general research case. Taking the magnetization parameter and turbulence spectral index as free parameters, together with radiative losses, we perform three group simulations to analyze the CR spectral, spatial distributions, and possible CR diffusion types. Our studies demonstrate that (1) CR energy density decays with both its effective radius and kinetic energy in the form of power-law distributions; (2) the morphology of the CR spatial distribution strongly depends on the properties of magnetic turbulence and the viewing angle; (3) CRs suffer a slow diffusion near the source and a fast/normal diffusion away from the source; (4) the existence of a power-law relationship between the averaged CR energy density and the magnetization parameter is independent of both CR energy and radiative losses; (5) radiative losses can suppress CR anisotropic diffusion and soften the power-law distribution of CR energy density. The distribution law established between turbulent magnetic fields and CRs presents an intrinsic property, providing a convenient way to understand complex astrophysical processes related to turbulence cascades.

Figures

Figures reproduced from arXiv: 2501.04986 by the authors.

Figure 1
Figure 1. The energy density UCR as a function of the particles’ effective radius r (upper row) and kinetic energy E (lower row). Left column: the time-dependent evolution of UCR corresponds to the no loss case of E ∈ 10[1,7] GeV, where the color bar shows the simulation time in units of yr. In panel (c), the UCR distribution of the several initial snapshots is shown in the inset. Right column: UCR distributions in four diffe… view at source ↗
Figure 2
Figure 2. Energy density UCR of the mono-energy 1 PeV as a function of the particle’s effective radius r. The results shown in panels (a) and (b) are based on Groups C and B, respectively. based on the measurement of the aspect ratio in the x and y di￾rections. We note that the spatial extent becomes larger with de￾creasing the turbulence spectral index. The reason why is that our simulation involves the energy-dependent diff… view at source ↗
Figure 3
Figure 3. Projected energy density R UCR dζ in xy, xz, and yz planes. The energy density denoted by the color bars is limited in the range of 10−12 − 10−6 GeV cm−3 pc. The black points show the positions of individual CR particles in the regions where the energy density falls below the range shown by the color bars. parallel and perpendicular scales of the eddy, respectively. Due to the interaction of particles with magnetic … view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Projected energy density R UCR dz in xy plane for the case of E = 1 TeV including the various MA (upper row, γ = 5/3) and γ (lower row, MA = 0.51). The energy density denoted by the color bars is limited in the range of 10−10 −10−6 GeV cm−3 pc. The black points show th…
Figure 7
Figure 7. Figure 7: The root mean square of the parallel deviations of particles (panel a) and the perpendicular one (panel b) as a function of the diffu￾sion radius R, plotted the non-loss case of E = 1 TeV with γ = 5/3 and MA = 0.51, where the color bar shows the simulation time in unit…
Figure 6
Figure 6. Figure 6: The parallel ensemble-averaged squared deviations of particles (d 2 ∥ ; panel a) and the perpendicular one (d 2 ⊥ ; panel b) as a function of the simulation time with (solid lines) and without (dashed lines) radiative loss processes. d 2 ∥ (and d 2 ⊥ ) of E = 10[1, 7] …
Figure 8
Figure 8. Figure 8: Upper row: the averaged energy density UCR as a function of the magnetization parameter MA (panel a) and diffusion index γ (panel b). Lower row: the ratio of the perpendicular diffusion coefficient to the parallel one, D⊥/D∥ , vs. MA (panel c); the parallel and perpend…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

94 extracted references · 70 canonical work pages

  1. [1]

    U., Albert , A., Alfaro , R., et al

    Abeysekara , A. U., Albert , A., Alfaro , R., et al. 2017, Science, 358, 911

    2017

  2. [2]

    2016, , 117, 231102

    Aguilar , M., Ali Cavasonza , L., Ambrosi , G., et al. 2016, , 117, 231102

    2016

  3. [3]

    C., & Jiang , Y.-F

    Armillotta , L., Ostriker , E. C., & Jiang , Y.-F. 2022, , 929, 170

    2022

  4. [4]

    M., Xu , S., & Lazarian , A

    Barreto-Mota , L., de Gouveia Dal Pino , E. M., Xu , S., & Lazarian , A. 2024, arXiv e-prints, arXiv:2405.12146

    arXiv 2024

  5. [5]

    2019, Living Reviews in Computational Astrophysics, 5, 2

    Beresnyak , A. 2019, Living Reviews in Computational Astrophysics, 5, 2

    2019

  6. [6]

    2011, , 728, 60

    Beresnyak , A., Yan , H., & Lazarian , A. 2011, , 728, 60

    2011

  7. [7]

    S., Bulanov , S

    Berezinskii , V. S., Bulanov , S. V., Dogiel , V. A., & Ptuskin , V. S. 1990, Astrophysics of cosmic rays

    1990

  8. [8]

    2013, , 21, 70

    Blasi , P. 2013, , 21, 70

    2013

Show all 94 references
  1. [9]

    A., Moraal , H., McCracken , K

    Caballero-Lopez , R. A., Moraal , H., McCracken , K. G., & McDonald , F. B. 2004, Journal of Geophysical Research (Space Physics), 109, A12102

    2004

  2. [10]

    2024, , 271, 25

    Cao , Z., Aharonian , F., An , Q., et al. 2024, , 271, 25

    2024

  3. [11]

    A., An , Q., et al

    Cao , Z., Aharonian , F. A., An , Q., et al. 2021, , 594, 33

    2021

  4. [12]

    & Donato , F

    Castellina , A. & Donato , F. 2005, Astroparticle Physics, 24, 146

    2005

  5. [13]

    S., Arzamasskiy , L., & Kunz , M

    Cerri , S. S., Arzamasskiy , L., & Kunz , M. W. 2021, , 916, 120

    2021

  6. [14]

    Cesarsky , C. J. 1980, , 18, 289

    1980

  7. [15]

    Chandran , B. D. G. 2000, , 85, 4656

    2000

  8. [16]

    & Lazarian , A

    Cho , J. & Lazarian , A. 2002, , 88, 245001

    2002

  9. [17]

    2022, , 106, 123033

    De La Torre Luque , P., Fornieri , O., & Linden , T. 2022, , 106, 123033

    2022

  10. [18]

    & Engelbrecht , N

    Dempers , N. & Engelbrecht , N. E. 2020, Advances in Space Research, 65, 2072

    2020

  11. [19]

    Elmegreen , B. G. & Scalo , J. 2004, , 42, 211

    2004

  12. [20]

    2008, , 2008, 018

    Evoli , C., Gaggero , D., Grasso , D., & Maccione , L. 2008, , 2008, 018

    2008

  13. [21]

    2012, , 108, 211102

    Evoli , C., Gaggero , D., Grasso , D., & Maccione , L. 2012, , 108, 211102

    2012

  14. [22]

    2018, , 98, 063017

    Evoli , C., Linden , T., & Morlino , G. 2018, , 98, 063017

    2018

  15. [23]

    2019 a , , 488, 4074

    Fang , K., Bi , X.-J., & Yin , P.-F. 2019 a , , 488, 4074

    2019

  16. [24]

    2019 b , , 884, 124

    Fang , K., Bi , X.-J., & Yin , P.-F. 2019 b , , 884, 124

    2019

  17. [25]

    2018, , 863, 30

    Fang , K., Bi , X.-J., Yin , P.-F., & Yuan , Q. 2018, , 863, 30

    2018

  18. [26]

    2023, , 108, 023017

    Fang , K., Hu , H.-B., Bi , X.-J., & Chen , E.-S. 2023, , 108, 023017

    2023

  19. [27]

    Farmer , A. J. & Goldreich , P. 2004, , 604, 671

    2004

  20. [28]

    2014, , 89, 083007

    Gaggero , D., Maccione , L., Grasso , D., Di Bernardo , G., & Evoli , C. 2014, , 89, 083007

    2014

  21. [29]

    Gaisser , T. K. 1990, Cosmic rays and particle physics

    1990

  22. [30]

    & Zhang , J.-F

    Gao , N.-N. & Zhang , J.-F. 2024, , 961, 80

    2024

  23. [31]

    & Sridhar , S

    Goldreich , P. & Sridhar , S. 1995, , 438, 763

    1995

  24. [32]

    A., Black , J

    Grenier , I. A., Black , J. H., & Strong , A. W. 2015, , 53, 199

    2015

  25. [33]

    J., et al

    Heesen , V., Buie , E., I., Huff , C. J., et al. 2019, , 622, A8

    2019

  26. [34]

    2005, , 71, 046401

    Holod , I., Zagorodny , A., & Weiland , J. 2005, , 71, 046401

    2005

  27. [35]

    2017, , 96, 103013

    Hooper , D., Cholis , I., Linden , T., & Fang , K. 2017, , 96, 103013

    2017

  28. [36]

    F., Butsky , I

    Hopkins , P. F., Butsky , I. S., Panopoulou , G. V., et al. 2022, , 516, 3470

    2022

  29. [37]

    2022, , 512, 2111

    Hu , Y., Lazarian , A., & Xu , S. 2022, , 512, 2111

    2022

  30. [38]

    Jokipii , J. R. & Parker , E. N. 1969, , 155, 777

    1969

  31. [39]

    & Quataert , E

    Kempski , P. & Quataert , E. 2022, , 514, 657

    2022

  32. [40]

    2014, Astroparticle Physics, 55, 37

    Kissmann , R. 2014, Astroparticle Physics, 55, 37

    2014

  33. [41]

    1941, Akademiia Nauk SSSR Doklady, 30, 301

    Kolmogorov , A. 1941, Akademiia Nauk SSSR Doklady, 30, 301

    1941

  34. [42]

    & Jokipii , J

    K \'o ta , J. & Jokipii , J. R. 2000, , 531, 1067

    2000

  35. [43]

    Kraichnan , R. H. 1965, Physics of Fluids, 8, 1385

    1965

  36. [44]

    R., Crocker , R

    Krumholz , M. R., Crocker , R. M., & Sampson , M. L. 2022, , 517, 1355

    2022

  37. [45]

    R., Crocker , R

    Krumholz , M. R., Crocker , R. M., Xu , S., et al. 2020, , 493, 2817

    2020

  38. [46]

    & Xu , S

    Lazarian , A. & Xu , S. 2021, , 923, 53

    2021

  39. [47]

    & Yan , H

    Lazarian , A. & Yan , H. 2014, , 784, 38

    2014

  40. [48]

    2024, , 962, 43

    Li , A.-f., Yuan , Q., Liu , W., & Guo , Y.-q. 2024, , 962, 43

    2024

  41. [49]

    & Buckman , B

    Linden , T. & Buckman , B. J. 2018, , 120, 121101

    2018

  42. [50]

    & Yan , H

    Liu , R.-Y. & Yan , H. 2020, , 494, 2618

    2020

  43. [51]

    2019, , 123, 221103

    Liu , R.-Y., Yan , H., & Zhang , H. 2019, , 123, 221103

    2019

  44. [52]

    2022, Nature Astronomy, 6, 199

    L \'o pez-Coto , R., de O \ n a Wilhelmi , E., Aharonian , F., Amato , E., & Hinton , J. 2022, Nature Astronomy, 6, 199

    2022

  45. [53]

    2023, , 953, 116

    Lu , F.-W., Zhu , B.-T., Hu , W., & Zhang , L. 2023, , 953, 116

    2023

  46. [54]

    2022, , 926, 94

    Maiti , S., Makwana , K., Zhang , H., & Yan , H. 2022, , 926, 94

    2022

  47. [55]

    2020, , 102, 023015

    Manconi , S., Di Mauro , M., & Donato , F. 2020, , 102, 023015

    2020

  48. [56]

    2002, , 394, 1039

    Maurin , D., Taillet , R., & Donato , F. 2002, , 394, 1039

    2002

  49. [57]

    2017, , 2017, 046

    Merten , L., Becker Tjus , J., Fichtner , H., Eichmann , B., & Sigl , G. 2017, , 2017, 046

    2017

  50. [58]

    & Ostrowsky , M

    Micha ek , G. & Ostrowsky , M. 1996, Nonlinear Processes in Geophysics, 3, 66

    1996

  51. [59]

    Moskalenko , I. V. & Strong , A. W. 1998, , 493, 694

    1998

  52. [60]

    & Linden , T

    Mukhopadhyay , P. & Linden , T. 2022, , 105, 123008

    2022

  53. [61]

    & Siemieniec-Ozi e b o , G

    Ostrowski , M. & Siemieniec-Ozi e b o , G. 1997, Astroparticle Physics, 6, 271

    1997

  54. [62]

    Parrish , I. J. & Stone , J. M. 2005, , 633, 334

    2005

  55. [63]

    J., Stone , J

    Parrish , I. J., Stone , J. M., & Lemaster , N. 2008, , 688, 905

    2008

  56. [64]

    2023, , 956, 75

    Qiao , B.-Q., Guo , Y.-Q., Liu , W., & Bi , X.-J. 2023, , 956, 75

    2023

  57. [65]

    A., et al

    Recchia , S., Di Mauro , M., Aharonian , F. A., et al. 2021, , 104, 123017

    2021

  58. [66]

    2002, Cosmic Ray Astrophysics

    Schlickeiser , R. 2002, Cosmic Ray Astrophysics

    2002

  59. [67]

    Seo , E. S. & Ptuskin , V. S. 1994, , 431, 705

    1994

  60. [68]

    2004, , 612, 238

    Shibata , T., Hareyama , M., Nakazawa , M., & Saito , C. 2004, , 612, 238

    2004

  61. [69]

    2020, , 491, 3860

    Sioulas , N., Isliker , H., Vlahos , L., Koumtzis , A., & Pisokas , T. 2020, , 491, 3860

    2020

  62. [70]

    Strong , A. W. & Moskalenko , I. V. 1998, , 509, 212

    1998

  63. [71]

    W., Moskalenko , I

    Strong , A. W., Moskalenko , I. V., & Ptuskin , V. S. 2007, Annual Review of Nuclear and Particle Science, 57, 285

    2007

  64. [72]

    & Piran , T

    Tang , X. & Piran , T. 2019, , 484, 3491

    2019

  65. [73]

    V., et al

    Trotta , R., J \'o hannesson , G., Moskalenko , I. V., et al. 2011, , 729, 106

    2011

  66. [74]

    R., Lee , M

    Webber , W. R., Lee , M. A., & Gupta , M. 1992, , 390, 96

    1992

  67. [75]

    Wiener , J., Pfrommer , C., & Oh , S. P. 2017, , 467, 906

    2017

  68. [76]

    2019, , 878, 104

    Xi , S.-Q., Liu , R.-Y., Huang , Z.-Q., Fang , K., & Wang , X.-Y. 2019, , 878, 104

    2019

  69. [77]

    2024, submitted to

    Xiao , Y.-W., Zhang , J.-F., & Xu , S. 2024, submitted to

    2024

  70. [78]

    2021, , 922, 264

    Xu , S. 2021, , 922, 264

    2021

  71. [79]

    & Li , H

    Xu , S. & Li , H. 2023, , 957, 97

    2023

  72. [80]

    & Yan , H

    Xu , S. & Yan , H. 2013, , 779, 140

    2013

  73. [81]

    & Lazarian , A

    Yan , H. & Lazarian , A. 2002, , 89, 281102

    2002

  74. [82]

    & Lazarian , A

    Yan , H. & Lazarian , A. 2008, , 673, 942

    2008

  75. [83]

    2012, , 745, 140

    Yan , H., Lazarian , A., & Schlickeiser , R. 2012, , 745, 140

    2012

  76. [84]

    2024, Nature Astronomy, 8, 628

    Yan , K., Liu , R.-Y., Zhang , R., et al. 2024, Nature Astronomy, 8, 628

    2024

  77. [85]

    & Xu , S

    Zhang , C. & Xu , S. 2023, , 959, L8

    2023

  78. [86]

    & Xu , S

    Zhang , C. & Xu , S. 2024, arXiv e-prints, arXiv:2406.03542

    2024 arXiv

  79. [87]

    2018, , 473, 3211

    Zhang , J.-F., Li , Z.-R., Xiang , F.-Y., & Lu , J.-F. 2018, , 473, 3211

    2018

  80. [88]

    & Xiang , F.-Y

    Zhang , J.-F. & Xiang , F.-Y. 2021, , 922, 209

    2021

  81. [89]

    A., Karpova , A

    Zyuzin , D. A., Karpova , A. V., Shibanov , Y. A., Potekhin , A. Y., & Suleimanov , V. F. 2021, , 501, 4998

    2021

  82. [90]

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

    ENTRY address archiveprefix author booktitle chapter edition editor howpublished institution eprint journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sent...

  83. [91]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  84. [92]

    @esa (Ref

    \@ifclassloaded agu2001 natbib The agu2001 class already includes natbib coding, so you should not add it explicitly Type <Return> for now, but then later remove the command natbib from the document \@ifclassloaded aguplus natbib The aguplus class already includes natbib codin...

  85. [93]

    @stdbsttrue NAT@ctr \@lbibitem[ NAT@ctr ] \@lbibitem[#1]#2 \@extra@b@citeb \@ifundefined br@#2\@extra@b@citeb \@namedef br@#2 \@nameuse br@#2\@extra@b@citeb \@ifundefined b@#2\@extra@b@citeb @num @parse #2 [ @natanchorstart #2\@extra@b@citeb \@biblabel @num @natanchorend] @ifc...

  86. [94]

    @open @close @open @close and [1] URL: #1 \@ifundefined chapter * \@mkboth \@ifundefined NAT@sectionbib * \@mkboth * \@mkboth\@gobbletwo \@ifclassloaded amsart * \@ifclassloaded amsbook * \@ifundefined bib@heading @heading NAT@ctr thebibliography [1] @ \@biblabel NAT@ctr \@bib...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.