Pith. sign in

REVIEW 4 major objections 5 minor 2 cited by

Self-consistent turbulent corona model reproduces the IceCube neutrino flux of NGC 1068

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

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

T0 review · deepseek-v4-flash

2026-08-02 17:08 UTC pith:GPI7YLM2

load-bearing objection A genuinely new modeling framework with a physical self-regulation loop, but the IceCube 'reproduction' leans on an imposed X-ray field and no code is out yet — worth refereeing, not worth taking as a hard prediction. the 4 major comments →

arxiv 2603.27749 v2 pith:GPI7YLM2 submitted 2026-03-29 astro-ph.HE

Self-Consistent Modelling of Neutrino Production in Turbulent Black Hole Coronae

S\'ebastien Le Bihan , Martin Lemoine , Frank Rieger This is my paper
classification astro-ph.HE
keywords active galactic nucleiNGC 1068neutrino astrophysicsstochastic particle accelerationturbulent coronaIceCubemulti-messenger astrophysicsturbulence damping
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper tries to show that the TeV neutrinos IceCube sees from NGC 1068 can be produced by protons accelerated in the turbulent corona around the supermassive black hole, without invoking exotic parameters or violating gamma-ray upper limits. The authors claim that when the back-reaction of accelerated protons on the turbulent cascade is included, the proton population self-regulates: energetic protons damp the turbulence that accelerates them, capping their own energy density near the turbulent magnetic pressure. Under these conditions the neutrino spectrum takes a universal template shape, and the observed flux is reproduced for standard coronal parameters. The same template, scaled only by X-ray luminosity, also accounts for the IceCube signals from NGC 4151 and NGC 7469. The significance is that neutrino observations become a direct probe of coronal plasma conditions.

Core claim

The central claim is that stochastic proton acceleration in the magnetized corona of NGC 1068, treated self-consistently with the damping of turbulence by the accelerated protons, reproduces the IceCube neutrino flux for R_cor ~ 15 r_g, v_A = 0.25c, B ~ 1e3 G, and L_d = 5e44 erg/s, while remaining consistent with gamma-ray upper limits. When the non-thermal proton fraction xi_p is at least about 1e-2, the system self-regulates and the proton spectrum converges to a near-universal shape with equal energy per decade, cut off at a few tens of TeV by photohadronic losses; the resulting neutrino spectrum has a broad peak around 1-10 TeV with a flat low-energy extension from pp interactions. The s

What carries the argument

Turb-AM3, a hybrid code coupling the time-dependent lepto-hadronic radiative solver AM3 with a stochastic acceleration module. The proton transport equation uses either a Fokker-Planck diffusion operator or a generalized Fermi (master) operator, and is solved together with a turbulent cascade equation that includes a damping kernel phi(k,p) transferring turbulent energy from wavenumber k to protons of momentum p. This coupling makes the acceleration rate time-dependent: as protons extract energy from the cascade, the turbulence is quenched at small scales, which automatically limits the non-thermal proton energy density.

Load-bearing premise

The corona is treated as a single, uniform, one-zone region with a constant size, constant advection speed, and a fixed thermal electron population; if the real corona is strongly stratified or has a non-thermal electron component, the target photon field and hence the p-gamma neutrino spectrum could change substantially.

What would settle it

A firm IceCube detection of neutrinos from NGC 1068 above ~100 TeV at a level exceeding the model's sharp cutoff, or a precise measurement of the <1 TeV neutrino flux that does not match the predicted flat pp component, would show that the mechanism or the assumed coronal parameters are wrong.

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

Share X Bluesky LinkedIn Reddit HN

If this is right

  • If the claim holds, standard corona parameters (R_cor ~ 15 r_g, v_A ~ 0.25c, B ~ 1e3 G, L_d ~ 5e44 erg/s) are sufficient to explain the IceCube neutrino signal from NGC 1068.
  • For xi_p greater than roughly 1e-2, the neutrino flux normalization becomes insensitive to the exact injection fraction, because turbulent damping self-regulates the proton energy density.
  • The neutrino spectrum follows a universal template: broad peak near 1-10 TeV, flat low-energy extension from pp interactions, and a sharp cutoff around tens of TeV set by photohadronic losses.
  • Detections near ~100 TeV (as hinted for NGC 7469) are accommodated by varying X-ray luminosity, while emission above ~100 TeV is suppressed.
  • The sub-TeV pp component is a diagnostic of proton density and hence of the pair-loading factor in the corona.

Where Pith is reading between the lines

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

  • If the universal template is correct, the neutrino spectral shape alone cannot pin down the proton injection fraction xi_p; the main observable discriminators are the low-energy pp component and the exact cutoff energy.
  • The model implies a tight relation between X-ray luminosity and neutrino peak energy: brighter coronae produce more neutrinos but at lower energies; this can be tested with IceCube's growing sample of Seyfert galaxies.
  • Extended to other compact environments (X-ray binaries, tidal disruption events), the same code would predict neutrino spectra whose shape is set by the same self-regulation mechanism, giving a way to test the generality of the picture.
  • Because the electron distribution is imposed rather than solved, a fuller test of the scenario requires coupling to a thermal Comptonization model, which the authors flag as future work.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper introduces Turb-AM3, a hybrid code coupling the AM3 lepto-hadronic radiative solver with a stochastic proton acceleration module that includes turbulent damping by accelerated particles. After presenting the code and physical setup in §2, the authors apply it to a one-zone model of the NGC 1068 corona in §3.1. They claim that a standard parameter set (R_cor ≈ 15 rg, v_A ≈ 0.25c, B ≈ 10^3 G, L_d ≈ 5e44 erg/s, ξ_p = 0.1) reproduces the IceCube neutrino flux while remaining compatible with gamma-ray upper limits, and that for ξ_p ≳ 10^-2 the proton–turbulence system self-regulates, producing a universal neutrino spectral template. The paper also extends the model to NGC 4151 and NGC 7469 by varying only the disk luminosity, and explores sensitivities to ξ_p, t_esc/t_acc, and t_adv/t_acc.

Significance. If the central claim holds, the paper would provide a concrete physical mechanism—turbulent acceleration with self-consistent back-reaction—that explains both the normalization and shape of the IceCube neutrino signal from AGN coronae, while keeping the gamma-ray flux suppressed via electromagnetic cascades. The prediction of a universal neutrino template and the explicit energy-budget consistency checks are useful and falsifiable. However, the significance is conditional: the 'self-consistent' result explicitly relies on an imposed, fixed Maxwellian electron distribution and on an X-ray photon field that is therefore an input assumption rather than a solved outcome. The paper also does not ship code or data, and it does not provide convergence tests, which weakens the reproducibility of the numerical claims. These issues are correctable within the paper's scope, so I regard the work as promising but not yet ready for acceptance as is.

major comments (4)
  1. [§2.1, §2.4, §3.1] The neutrino flux normalization and peak are driven by pγ interactions whose target photons are the coronal X-ray field. That field is computed from an assumed, fixed Maxwellian electron distribution at Θ_e = 0.2, not solved self-consistently. The paper itself states in §3.1 that 'the electron distribution around the thermal peak is not finely resolved' and postpones electron energization to future work. The headline 'self-consistent' reproduction of the IceCube flux is therefore conditional on an input assumption. The authors should quantify the sensitivity: e.g., vary Θ_e around 0.2 or add a modest nonthermal electron tail and recompute the neutrino spectrum, to show how much the normalization and peak move.
  2. [§3.2, Fig. 4] The comparison procedure in Fig. 4 tunes v_A to a fixed pivot point (Eν = 10 TeV, E²Φν = 1.4e-11 erg/cm²/s). Since v_A controls both the acceleration rate and, through Eq. (3), B and the coronal magnetic energy, this tuning strongly influences the predicted neutrino flux. The fiducial set in §3.1 is not tested against a quantitative goodness-of-fit statistic; the claim of 'satisfactorily reproduced' rests on visual overlap. The authors should present untuned predictions and a statistical measure (e.g., chi-square or likelihood) for the fiducial and tuned cases, given the degeneracies displayed in Figs. 4–6.
  3. [§2.7, §3.1, §3.3] No convergence tests are presented for the numerical method: no resolution study in momentum space, no timestep convergence, no comparison of the coupled acceleration–cascade solver against an analytic limit. The universal spectral shape and saturation behavior claimed in §3.2–3.3 could be numerical artifacts if the grid does not resolve the acceleration–loss balance near the TeV cutoff. The authors should add convergence tests and ideally release the code or a reproducible input/output package, as this is central to validating the new Turb-AM3 framework.
  4. [§3.2, §3.3] The 'universal' spectral shape is asserted on the basis of visual inspection of Figs. 4, 5, and A.1. No quantitative definition or metric is given for convergence to this template (e.g., a fractional deviation of the peak energy or spectral slope across ξ_p and t_esc/t_acc). Since the universality is a central new result, the authors should define and compute a shape-difference measure for the neutrino spectrum in the self-regulated regime.
minor comments (5)
  1. [Abstract, passim] Typographical errors: 'excape' in §2.1, 'respectivly' in the Introduction, and 'articially' in §3.2. The manuscript would benefit from a careful proofread.
  2. [Fig. 2] The figure contains many curves and labels; the color/line assignments are complex. A table of components (or a simplified version separating photon and neutrino panels) would improve readability.
  3. [§2.5, Eq. (10)] The footnote about Eq. (10) is helpful, but the limiting case t_adv ≫ t_esc is stated only verbally; it would be clearer to write the explicit limiting form.
  4. [§2.6, Eq. (11)] The kernel φ(k,p) is said to be 'normalized to unity' but the precise normalization variable in the Gaussian is ambiguous (ln k vs k). Clarify the argument of the Gaussian and the normalization measure.
  5. [References] Several 2026 references are cited as 'arXiv e-prints' with no journal page numbers; if any have since been published, the bibliographic entries should be updated.

Circularity Check

1 steps flagged

Parameter-space neutrino 'reproduction' is partly enforced by tuning v_A to the IceCube pivot, but the fiducial NGC 1068 model is a genuine forward calculation.

specific steps
  1. fitted input called prediction [Section 3.2, Fig. 4 (parameter exploration)]
    "In order to obtain a meaningful comparison to observations, we slightly tune the Alfvénic velocity vA to match the observed neutrino flux at a pivot point defined by Eν =10 TeV and (all-flavour) flux E 2 ν Φν =1.4×10 −11 erg/cm2s ... The model reliably reproduces the IceCube neutrino flux, independently of the precise value of ξp, shaping a neutrino spectral shape with a turn-over at energies below the peak (Eν∼3 TeV here)."

    The Fig. 4 curves are normalized by construction: v_A is adjusted so that every neutrino spectrum passes through the observed E²Φ=1.4e-11 erg/cm²/s point at 10 TeV. The subsequent statement that the model 'reliably reproduces the IceCube neutrino flux' therefore holds at the pivot because the parameter was fitted to that datum, not because the model predicted it. The spectral shape away from the pivot (pp plateau, pγ cutoff) is not forced, which limits the circularity; the fiducial Sec. 3.1 model is also not pivot-tuned.

full rationale

The central fiducial calculation (Sec. 3.1, Fig. 2) is a forward model: inputs are the observed disk luminosity, coronal size, Thomson depth, an assumed Maxwellian electron distribution (Θe=0.2), ξp=0.1, and vA=0.25c; the neutrino spectrum is obtained by solving coupled proton transport and radiative losses in Turb-AM3. The agreement with IceCube at 1-30 TeV is therefore not circular in the fiducial case, and the model also reproduces X-ray and gamma-ray constraints. The main circularity risk is confined to Sec. 3.2/Fig. 4, where v_A is explicitly tuned to the 10 TeV IceCube flux before the model is said to 'reproduce' the flux; that is a fitted-input-called-prediction at the pivot. I also considered the imposed electron/X-ray field: the pγ neutrino flux is conditional on the assumed Maxwellian electron population and on not modeling electron energization, as the authors acknowledge in Secs. 2.1 and 4. But this is an assumption about the environment, not an equivalence between input and output, so it is a robustness/self-consistency limitation rather than a circular reduction. Self-citations (Lemoine 2022; Lemoine et al. 2024; Lemoine & Rieger 2025) are used for the acceleration/damping operators, but the formalism is benchmarked on MHD/PIC simulations and Appendix A shows insensitivity to the acceleration scheme, so the self-citations are not load-bearing. Overall: one partial, transparent pivot fit in the parameter study; the headline forward calculation retains independent content.

Axiom & Free-Parameter Ledger

8 free parameters · 6 axioms · 0 invented entities

The model rests on a handful of fitted parameters (R_cor, v_A, ξ_p, l_c, v_adv, f_cor, X-ray normalization) and on simplifying domain assumptions (one-zone corona, fixed Maxwellian electron population, prescribed injection spectrum, Kolmogorov turbulence with specific damping kernel). No new physical entities are introduced. The central claim is thus a model-dependent interpretation of the IceCube data rather than a parameter-free prediction. The paper itself flags most of these assumptions, but they are load-bearing for the neutrino flux and spectral shape.

free parameters (8)
  • Turbulent dissipation coefficient f_diss = 0.1
    Chosen to match the reconnection rate scaling in relativistic turbulence; used in Eq. 2-3 to derive B and v_A. Sensitive to this choice.
  • Non-thermal proton injection fraction ξ_p = 0.1 (explored 1e-4…0.33)
    Free parameter controlling the initial energy fraction in non-thermal protons; tuned to match observed neutrino flux (Sec. 3.2).
  • Corona size R_cor = 15 r_g (explored 5-20 r_g)
    Chosen to be consistent with gamma-ray opacity constraints; controls escape, advection time, radiation field, and therefore neutrino flux.
  • Alfvén velocity v_A (or magnetization σ) = 0.25c fiducial; tuned in parameter scans to match neutrino pivot flux
    The key acceleration-rate parameter; in Sec. 3.2 explicitly varied to match the observed neutrino flux at E_nu = 10 TeV.
  • Turbulence coherence length l_c = 4 r_g fiducial; varied in scans
    Controls t_esc/t_acc and cascade damping; chosen to place the model in the confined regime.
  • Advection velocity v_adv = 0.03c fiducial; varied
    Sets the acceleration time window; normalization tied to standard Shakura-Sunyaev alpha parameter but effectively treated as a free parameter in scans.
  • Corona covering fraction f_cor = 0.2
    Fraction of disk luminosity intercepted by the corona; controls the target photon field, chosen from standard disk-corona geometry.
  • X-ray luminosity normalization = L_X(2-10 keV) = 6.8e43 erg/s for NGC 1068; L_X adjusted to observed values for NGC 4151/7469
    The model is normalized to observed X-ray luminosity for each source; this sets the pγ target photon density.
axioms (6)
  • domain assumption Corona is a one-zone spherical, uniform region with constant size R_cor and constant advection velocity
    Section 2.1, 2.5: the entire model relies on a homogeneous corona; the paper acknowledges this is idealized and partially mitigates it by time-averaging, but radial GRMHD gradients are not modelled.
  • ad hoc to paper Electron distribution is a fixed thermal Maxwellian with Θ_e = 0.2 at all times
    Section 2.1: 'we rather assume that this distribution is characterized at all times by a Maxwellian'. The X-ray target photons, which drive pγ neutrino production, are computed from this imposed distribution. This is a domain approximation that avoids solving the electron energization problem.
  • ad hoc to paper Protons are continuously injected at the outer edge of the corona with a power-law spectrum p^-4 and a free normalization ξ_p
    Section 2.7: 'Non-thermal protons are initialized with a power-law shape dN_p/dp ∝ p^-4...' The justification is that PIC simulations give this slope, but the injection mechanism (reconnection extracted protons) is not modeled and ξ_p is free.
  • domain assumption The turbulence is isotropic with Kolmogorov spectrum S_k ∝ k^-5/3 and a damping kernel φ(k,p) that maps wavenumber to proton gyroradius k ~ eB/pc
    Section 2.6: the cascade equation and acceleration-damping feedback rely on this spectrum and kernel; the paper states the kernel choice does not affect conclusions, but this is not demonstrated in the present paper.
  • domain assumption Quasi-linear / generalized-Fermi transport prescription captured by either L_FP or L_GF with acceleration coefficient D_pp = 0.3 σ_δB p^2 c/l_c
    Section 2.2: the acceleration rate is taken from PIC simulations with a specific scaling; the paper notes a competing sub-relativistic scaling from Wong et al. 2025 but keeps the relativistic one, acknowledging an order-unity renormalization of the magnetization.
  • domain assumption Protons are the only species that undergo stochastic acceleration; pions, muons, leptons do not
    Section 2.3, Table 1: motivated by short cooling timescales for leptons and short decay times for pions/muons. This is standard for this problem, but it is an assumption that restricts the physics.

pith-pipeline@v1.3.0-alltime-deepseek · 26382 in / 10923 out tokens · 83966 ms · 2026-08-02T17:08:56.281578+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Self-Consistent Modelling of Neutrino Production in Turbulent Black Hole Coronae." pith.science (2026). https://pith.science/paper/GPI7YLM2

@misc{pith2026260327749,
  author       = {Pith},
  title        = {Pith review of: Self-Consistent Modelling of Neutrino Production in Turbulent Black Hole Coronae},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GPI7YLM2}},
  note         = {Machine review of arXiv:2603.27749}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Stochastic particle acceleration in magnetized turbulent plasmas has emerged as a key mechanism to explain multi-messenger signals from compact astrophysical environments. Self-consistent modelling remains challenging because it requires to treat simultaneously several non-linear kinetic processes, especially turbulence-driven acceleration and its feedback on the turbulent cascade, as well as the radiative and hadronic losses, including the reprocessing of electromagnetic radiation in radiatively dense environments. The present paper introduces the hybrid numerical code Turb-AM3 designed to this effect. This hybrid numerical code couples the state-of-the-art time-dependent lepto-hadronic radiative solver AM3 with a stochastic acceleration module that incorporates recent theoretical advances in turbulent acceleration and accounts for the dynamical damping of turbulence by accelerated particles. In the second part of the paper, we use this code to provide self-consistent time-dependent models of proton acceleration in the turbulent black hole corona of NGC~1068. We find that the IceCube neutrino signal is well reproduced for a standard set of physical parameters describing the black hole corona. The same template model accounts in a satisfactory way for IceCube observations of other active galactic nuclei. Furthermore, our exploration of parameter space allows us to predict detailed template spectral shapes for the TeV neutrino spectrum, which in turn help understand how future neutrino observations can constrain the properties of turbulent AGN coronae and the underlying acceleration mechanism. This Turb-AM3 framework provides a powerful tool to model multi-messenger emission in a broad variety of compact astrophysical environments.

Figures

Figures reproduced from arXiv: 2603.27749 by Frank Rieger, Martin Lemoine, S\'ebastien Le Bihan.

Figure 1
Figure 1. Figure 1: Timescales of the different processes occurring in the AGN corona, including the escape of charged and neutral par￾ticles, the acceleration of protons, radiative and hadronic losses of protons and electrons/positrons, γγ annihilation of photons and the advection timescale of the coronal plasma. is approximated as κturb ≃ ℓcvA/3, in agreement with recent PIC simulation results (Groselj et al. 2026), while t… view at source ↗
Figure 2
Figure 2. Figure 2: Photon (blue) and neutrino (red) spectra from NGC 1068 computed with [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Top panel: Time evolution of the integrated proton spec [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Integrated proton (top panel), as well as photon and [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Proton, photon, and neutrino stationary spectra for dif [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 8
Figure 8. Figure 8: Predicted stationary neutrino spectra for di [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗

discussion (0)

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

Forward citations

Cited by 2 Pith papers

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

  1. Particle Acceleration, Coronal Neutrino Production, and the Diffuse Extragalactic Neutrino Background from Supermassive Black Holes

    astro-ph.HE 2026-05 unverdicted novelty 6.0

    The cosmologically integrated neutrino emission from supermassive black hole coronae in Seyfert galaxies can account for the sub-PeV diffuse extragalactic neutrino flux observed by IceCube.

  2. On the Blueprint of Active Galaxies Producing Neutrinos

    astro-ph.HE 2026-07 conditional novelty 4.0

    Neutrinos from active galaxies are produced in compact X-ray-bright coronae within about ten Schwarzschild radii of the black hole, and such sources may supply the diffuse neutrino flux.

Reference graph

Works this paper leans on

95 extracted references · 5 linked inside Pith · cited by 2 Pith papers

  1. [1]

    G., Ackermann, M., Adams, J., et al

    Aartsen, M. G., Ackermann, M., Adams, J., et al. 2020, Phys. Rev. Lett., 124, 051103

    2020

  2. [2]

    2026, Astrophys

    Abbasi, R., Ackermann, M., Adams, J., et al. 2026, Astrophys. J. Lett., 1000, L26

    2026

  3. [3]

    2023, Astrophys

    Ajello, M., Murase, K., & McDaniel, A. 2023, Astrophys. J. Lett., 954, L49

    2023

  4. [4]

    Ambrosone, A. 2024, J. Cosmol. Astropart. Phys., 2024, 075

    2024

  5. [5]

    2008, Astron

    Belmont, R., Malzac, J., & Marcowith, A. 2008, Astron. Astrophys., 491, 617

    2008

  6. [6]

    Beloborodov, A. M. 2017, Astrophys. J., 850, 141

    2017

  7. [7]

    S., Bulanov, S

    Berezinskii, V . S., Bulanov, S. V ., Dogiel, V . A., & Ptuskin, V . S. 1990, Astron. Nachr. North-Holland, Amsterdam, 312, 413

    1990

  8. [8]

    2022, Phys

    Bresci, V ., Lemoine, M., Gremillet, L., et al. 2022, Phys. Rev. D, 106, 023028

    2022

  9. [9]

    2026, Astrophys

    Cerruti, M., Rudolph, A., Petropoulou, M., et al. 2026, Astrophys. J. Suppl. Ser., 282, 22

    2026

  10. [10]

    H., & Giommi, P

    Chang, Y .-L., Brandt, C. H., & Giommi, P. 2020, Astron. Comput., 30, 100350

    2020

  11. [11]

    & Sironi, L

    Comisso, L. & Sironi, L. 2019, Astrophys. J., 886, 122

    2019

  12. [12]

    & Sironi, L

    Comisso, L. & Sironi, L. 2022, Astrophys. J. Lett., 936, L27

    2022

  13. [13]

    Coppi, P. S. 1992, Mon. Not. R. Astron. Soc., 258, 657

    1992

  14. [14]

    T., & Murase, K

    Das, A., Zhang, B. T., & Murase, K. 2024, Astrophys. J., 972, 44

    2024

  15. [15]

    2025, arXiv e-prints, arXiv:2506.04212

    Das, S., Xu, S., & Nättilä, J. 2025, arXiv e-prints, arXiv:2506.04212

    Pith/arXiv arXiv 2025

  16. [16]

    2020, Phys

    Demidem, C., Lemoine, M., & Casse, F. 2020, Phys. Rev. D, 102, 023003

    2020

  17. [17]

    2026, arXiv e-prints, arXiv:2603.09394 Dovˇciak, M., Papadakis, I

    Dmytriiev, A., van der Merwe, F., & Böttcher, M. 2026, arXiv e-prints, arXiv:2603.09394 Dovˇciak, M., Papadakis, I. E., Kammoun, E. S., & Zhang, W. 2022, Astron. Astrophys., 661, A135

    arXiv 2026

  18. [18]

    Eichmann, B., Oikonomou, F., Salvatore, S., Dettmar, R.-J., & Tjus, J. B. 2022, Astrophys. J., 939, 43

    2022

  19. [19]

    C., Lohfink, A., Belmont, R., Malzac, J., & Coppi, P

    Fabian, A. C., Lohfink, A., Belmont, R., Malzac, J., & Coppi, P. 2017, Mon. Not. R. Astron. Soc., 467, 2566

    2017

  20. [20]

    C., Lohfink, A., Kara, E., et al

    Fabian, A. C., Lohfink, A., Kara, E., et al. 2015, Mon. Not. R. Astron. Soc., 451, 4375

    2015

  21. [21]

    L., Halzen, F., & Gallagher, J

    Fang, K., Rodriguez, E. L., Halzen, F., & Gallagher, J. S. 2023, Astrophys. J., 956, 8

    2023

  22. [22]

    1954, Astrophys

    Fermi, E. 1954, Astrophys. J., 119, 1

    1954

  23. [23]

    Fiorillo, D. F. G., Comisso, L., Peretti, E., Petropoulou, M., & Sironi, L. 2025, Astrophys. J., 989, 215

    2025

  24. [24]

    Frank, J., King, A., & Raine, D. J. 2002, Accretion Power in Astrophysics, 398

    2002

  25. [25]

    A., Rosner, R., & Vaiana, G

    Galeev, A. A., Rosner, R., & Vaiana, G. S. 1979, Astrophys. J., 229, 318

    1979

  26. [26]

    & Tavecchio, F

    Ghisellini, G. & Tavecchio, F. 2009, Mon Not R Astron Soc, 397, 985

    2009

  27. [27]

    2025, Astro- phys

    Gong, X.-W., Liu, R.-Y ., Zhang, Z.-L., Asano, K., & Lemoine, M. 2025, Astro- phys. J., 989, 99

    2025

  28. [28]

    A., Grošelj, D., & Bacchini, F

    Gorbunov, E. A., Grošelj, D., & Bacchini, F. 2025, Phys. Rev. Lett., 135, 065201

    2025

  29. [29]

    M., Sironi, L., & Philippov, A

    Groselj, D., Hakobyan, H., Beloborodov, A. M., Sironi, L., & Philippov, A. 2024, Phys. Rev. Lett., 132, 085202

    2024

  30. [30]

    M., & Mushotzky, R

    Groselj, D., Philippov, A., Beloborodov, A. M., & Mushotzky, R. 2026, arXiv e-prints, arXiv:2601.00518

    Pith/arXiv arXiv 2026

  31. [31]

    2025, Phys

    Herrera, G. 2025, Phys. Dark Univ., 50, 102156

    2025

  32. [32]

    Hinkle, J. T. & Mushotzky, R. 2021, Mon. Not. R. Astron. Soc., 506, 4960

    2021

  33. [33]

    2018, Science, 361, eaat1378 IceCube Collaboration, Aartsen, M

    IceCube, T., Fermi-LAT, MAGIC, et al. 2018, Science, 361, eaat1378 IceCube Collaboration, Aartsen, M. G., Ackermann, M., et al. 2018, Science, 361, 147 IceCube Collaboration, Abbasi, R., M. Ackermann, et al. 2022, Science, 378, 538

    2018

  34. [34]

    2019, Astrophys

    Inoue, Y ., Khangulyan, D., Inoue, S., & Doi, A. 2019, Astrophys. J., 880, 40

    2019

  35. [35]

    Jaffe, W., Meisenheimer, K., Röttgering, H. J. A., et al. 2004, Nature, 429, 47

    2004

  36. [36]

    2016, Astrophys

    Kakuwa, J. 2016, Astrophys. J., 816, 24

    2016

  37. [37]

    M., Masterson, M., et al

    Kammoun, E., Lohfink, A. M., Masterson, M., et al. 2024, Front. Astron. Space Sci., 10, 1308056

    2024

  38. [38]

    J., et al

    Kantzas, D., Markoff, S., Cooper, A. J., et al. 2023, Mon. Not. R. Astron. Soc., 524, 1326

    2023

  39. [39]

    Karavola, D., Petropoulou, M., Fiorillo, D. F. G., Comisso, L., & Sironi, L. 2025, J. Cosmol. Astropart. Phys., 2025, 075

    2025

  40. [40]

    B., Quataert, E., et al

    Kempski, P., Fielding, D. B., Quataert, E., et al. 2023, Mon. Not. R. Astron. Soc., 525, 4985

    2023

  41. [41]

    Kheirandish, A., Murase, K., & Kimura, S. S. 2021, Astrophys. J., 922, 45

    2021

  42. [42]

    S., Tomida, K., & Murase, K

    Kimura, S. S., Tomida, K., & Murase, K. 2019, Mon. Not. R. Astron. Soc., 485, 163

    2019

  43. [43]

    2024, Astrophys

    Klinger, M., Rudolph, A., Rodrigues, X., et al. 2024, Astrophys. J. Suppl. Ser., 275, 4

    2024

  44. [44]

    C., Gandhi, P., et al

    Kumar, S., Dewangan, G. C., Gandhi, P., et al. 2024, Astrophys. J., 975, 73

    2024

  45. [45]

    2019, Phys

    Lemoine, M. 2019, Phys. Rev. D, 99, 083006

    2019

  46. [46]

    2021, Phys

    Lemoine, M. 2021, Phys. Rev. D, 104, 063020

    2021

  47. [47]

    2022, Phys

    Lemoine, M. 2022, Phys. Rev. Lett., 129, 215101 Article number, page 13 A&A proofs:manuscript no. Article_NGC1068_SLB_ML_FR

    2022

  48. [48]

    Lemoine, M. 2023, J. Plasma Phys., 89, 175890501

    2023

  49. [49]

    2025, Phys

    Lemoine, M. 2025, Phys. Rev. E, 112, 015205

    2025

  50. [50]

    & Malkov, M

    Lemoine, M. & Malkov, M. A. 2020, Mon. Not. R. Astron. Soc., 499, 4972

    2020

  51. [51]

    2024, Phys

    Lemoine, M., Murase, K., & Rieger, F. 2024, Phys. Rev. D, 109, 063006

    2024

  52. [52]

    & Rieger, F

    Lemoine, M. & Rieger, F. 2025, Astron. Astrophys., 697, A124

    2025

  53. [53]

    A., Lehnert, M., & Karczewski, O

    Lianou, S., Barmby, P., Mosenkov, A. A., Lehnert, M., & Karczewski, O. 2019, Astron. Astrophys., 631, A38

    2019

  54. [54]

    2018, Astrophys

    Lopez-Rodriguez, E., Fuller, L., Alonso-Herrero, A., et al. 2018, Astrophys. J., 859, 99

    2018

  55. [55]

    2025, arXiv e-prints, arXiv:2503.09426

    Lu, M.-X., Liang, Y .-F., Wang, X.-G., & Ouyang, X.-R. 2025, arXiv e-prints, arXiv:2503.09426

    Pith/arXiv arXiv 2025

  56. [56]

    W., Quataert, E., Chandran, B

    Lynn, J. W., Quataert, E., Chandran, B. D. G., & Parrish, I. J. 2014, Astrophys. J., 791, 71 MAGIC Collaboration, Acciari, V . A., Ansoldi, S., et al. 2019, Astrophys. J., 883, 135

    2014

  57. [57]

    2016, Mon

    Marinucci, A., Bianchi, S., Matt, G., et al. 2016, Mon. Not. R. Astron. Soc., 456, L94

    2016

  58. [58]

    2024, Phys

    Mbarek, R., Philippov, A., Chernoglazov, A., Levinson, A., & Mushotzky, R. 2024, Phys. Rev. D, 109, L101306

    2024

  59. [59]

    2023, Astrophys

    Meringolo, C., Cruz-Osorio, A., Rezzolla, L., & Servidio, S. 2023, Astrophys. J., 944, 122

    2023

  60. [60]

    2022, Astrophys

    Murase, K. 2022, Astrophys. J. Lett., 941, L17

    2022

  61. [61]

    M., Kimura, S

    Murase, K., Karwin, C. M., Kimura, S. S., Ajello, M., & Buson, S. 2024, Astro- phys. J. Lett., 961, L34

    2024

  62. [62]

    S., & Mészáros, P

    Murase, K., Kimura, S. S., & Mészáros, P. 2020, Phys. Rev. Lett., 125, 011101

    2020

  63. [63]

    S., Mukhopadhyay, M., & Bhattacharya, M

    Murase, K., Kimura, S. S., Mukhopadhyay, M., & Bhattacharya, M. 2026, arXiv e-prints, arXiv:2602.20145 Nättilä, J. 2024, Nature Comm., 15, 7026

    arXiv 2026

  64. [64]

    Neronov, A., Savchenko, D., & Semikoz, D. V . 2024, Phys. Rev. Lett., 132, 101002 Nhat Ly, M., Inoue, Y ., Sentoku, Y ., & Sano, T. 2026, arXiv e-prints, arXiv:2601.01999

    Pith/arXiv arXiv 2024

  65. [65]

    2024, Nat

    Padovani, P., Resconi, E., Ajello, M., et al. 2024, Nat. Astron., 8, 1077

    2024

  66. [66]

    Pezzi, O., Blasi, P., & Matthaeus, W. H. 2022, Astrophys. J., 928, 25

    2022

  67. [67]

    & Svensson, R

    Poutanen, J. & Svensson, R. 1996, Astrophys. J., 470, 249

    1996

  68. [68]

    V ., Horne, K., et al

    Prince, R., Hernández Santisteban, J. V ., Horne, K., et al. 2025, Mon. Not. R. Astron. Soc., 541, 642

    2025

  69. [69]

    2023, Astrophys

    Pugliese, F., Brodiano, M., Andrés, N., & Dmitruk, P. 2023, Astrophys. J., 959, 28

    2023

  70. [70]

    2022, Phys

    Reusch, S., Stein, R., Kowalski, M., et al. 2022, Phys. Rev. Lett., 128

    2022

  71. [71]

    C., Fabian, A

    Ricci, C., Ho, L. C., Fabian, A. C., et al. 2018, Mon. Not. R. Astron. Soc., 480, 1819

    2018

  72. [72]

    S., Garrappa, S., et al

    Rodrigues, X., Paliya, V . S., Garrappa, S., et al. 2024, Astron. Astrophys., 681, A119

    2024

  73. [73]

    Rybicki, G. B. & Lightman, A. P. 1979, A Wiley-Interscience Publication

    1979

  74. [74]

    2002, Cosmic Ray Astrophysics (Springer, Berlin)

    Schlickeiser, R. 2002, Cosmic Ray Astrophysics (Springer, Berlin)

    2002

  75. [75]

    & Tavecchio, F

    Sciaccaluga, A. & Tavecchio, F. 2022, Mon. Not. R. Astron. Soc., 517, 2502

    2022

  76. [76]

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

    1973

  77. [77]

    & Beloborodov, A

    Sironi, L. & Beloborodov, A. M. 2020, Astrophys. J., 899, 52

    2020

  78. [78]

    Sridhar, N., Ripperda, B., Sironi, L., Davelaar, J., & Beloborodov, A. M. 2025, Astrophys. J., 979, 199

    2025

  79. [79]

    v., Kowalski, M., et al

    Stein, R., Velzen, S. v., Kowalski, M., et al. 2021, Nat. Astron., 5, 510–518

    2021

  80. [80]

    E., Poutanen, J., Svensson, R., Sikora, M., & Begelman, M

    Stern, B. E., Poutanen, J., Svensson, R., Sikora, M., & Begelman, M. C. 1995, Astrophys. J. Lett., 449, L13

    1995

Showing first 80 references.