Pith. sign in

REVIEW 4 major objections 5 minor 73 references

The Statistic Analysis of GeV Spectral Breaks in Bright Gamma-Ray FSRQs

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper claims that the GeV spectral break in bright flat-spectrum radio quasars sits at 2.90 ± 1.92 GeV in the rest frame with a photon-index change of Δγ = 0.45 ± 0.19, matching the cooling break of electrons scattering external seed…

desk verdict Useful large-sample catalog with a plausible population average, but the physical interpretation rests on treating a range-dependent fit parameter as a real spectral break. read the letter →

arxiv 2412.02163 v1 pith:37LHBLRY submitted 2024-12-03 astro-ph.HE

K.R.Zhu , J.M. Chen , L. Zhang This is my paper · ORCID
classification astro-ph.HE
keywords gamma-rayastronomyFSRQsblazarsspectralbreakbrokenpowerlawlog-parabolaspectrumcoolingFermi-LAT
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

Bright flat-spectrum radio quasars (FSRQs) are a class of blazars whose gamma-ray spectra typically bend somewhere in the 0.1-10 GeV band. The paper asks whether that bend is a real physical feature and what causes it. Using 15 years of data from the Large Area Telescope on the Fermi satellite, the authors fit 755 FSRQs and isolate 87 bright ones; they show that a broken power law and a log-parabola describe the population equally well and that BPL-preferred and LP-preferred sources form a single class. In the rest frame the breaking energy is $2.90 \pm 1.92$ GeV and the photon index softens by $\Delta\gamma = 0.45 \pm 0.19$ after the break. The authors argue this is the signature of radiative cooling of electrons scattering external seed photons, and that absorption by broad-line-region photons cannot be the main cause.

What carries the argument

The load-bearing tool is the broken power-law spectral model $dN/dE = N_0 (E/E_b)^{-\gamma_1}$ for $E < E_b$ and $N_0 (E/E_b)^{-\gamma_2}$ for $E \ge E_b$, whose parameters are the break energy $E_b$ and the low- and high-energy photon indices $\gamma_1$ and $\gamma_2$. The paper uses the fitted $E_b$ as the location of a real spectral break and the index change $\Delta\gamma = \gamma_2 - \gamma_1$ as the break's sharpness. The physical interpretation is carried by the cooling-break expectation: in slow cooling, radiative losses break the injected electron spectrum by $\Delta p = 1$, which translates to a photon-index change $\Delta\gamma = 0.5$; matching this to the observed $0.45 \pm 0.19$ is the evidence that the break is radiative cooling of electrons scattering external seed photons. The same model, combined with the relation $E'_b \simeq \gamma_b^2 \Gamma^2 E_{\rm ext}$, yields an electron break Lorentz factor $\gamma_b \sim 10^3$.

What would settle it

Re-fit the 87 bright FSRQ spectra with the LP model and compute each spectrum's curvature peak (the energy where $E^2 dN/dE$ is maximal); if the distribution of these peak energies does not match the distribution of BPL break energies centered at 2.9 GeV, then the break location is an artifact of the broken power-law parameterization.

Watch

Extended reading notes

Core claim

Using the broken power-law model to locate spectral breaks in 87 bright FSRQs with 15-year averaged Fermi-LAT spectra, the paper finds that GeV breaks are common, with rest-frame break energies of $\langle E'_b\rangle = 2.90 \pm 1.92$ GeV and photon-index change $\langle \Delta\gamma\rangle = 0.45 \pm 0.19$. Because an AIC comparison shows the population as a whole does not clearly prefer either the BPL or the LP model, and clustering puts BPL-preferred and LP-preferred sources in one class, the paper treats the two models as interchangeable descriptions of a smooth break. The observed $\Delta\gamma$ agrees with the slow-cooling expectation of a break in the injected electron spectrum, and the break energy implies an electron break Lorentz factor of order $10^3$. The paper further argues that photon-photon absorption in the broad-line region is unlikely, since the break energy is far from the characteristic absorption energies of strong BLR lines and $\Delta\gamma$ does not correlate with disk luminosity; instead, a break at 2.9 GeV can be produced if the emission region lies beyond the dust torus at radii $\gtrsim 8$ pc. The conclusion is that the GeV break in bright FSRQs is a cooling break of electrons scattering external seed photons.

Load-bearing premise

The load-bearing premise is that the break energy fitted with a broken power law marks a real spectral break location, even though the paper argues the smooth log-parabola model describes the same data just as well.

Editorial extensions

If this is right

  • The gamma-ray emission region of bright FSRQs must sit at radii $\gtrsim 8$ pc from the central engine, beyond the dust torus, for the cooling break to land at 2.9 GeV.
  • The population average gives a benchmark: individual-source breaks should scatter around 2.9 GeV with a dispersion of about 1.9 GeV under the same cooling physics.
  • Because all 87 bright sources show significant softening, a spectral break near a few GeV is a generic property of bright FSRQ gamma-ray spectra, not a property of a subset.
  • The weak correlation between $\Delta\gamma$ and disk luminosity shifts the default explanation for the break away from BLR photon-photon absorption and toward intrinsic electron cooling.

Reading between the lines

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

  • Editorial: because the paper itself shows that fitting 0.1-50 GeV moves the BPL break energy upward, the 2.9 GeV value is tied to the 0.1-10 GeV band; a natural extension is to test whether an energy-range-independent quantity, such as the curvature peak of the LP fit, reproduces the same break location.
  • Editorial: a population-level prediction follows: if the break is a cooling break from external seed photons, then in sources with higher external radiation density the break should sit at lower rest-frame energy; a correlation between $E'_b$ and the Eddington ratio or disk luminosity would be a testable consequence that the paper does not examine.
  • Editorial: the paper's clustering result that LP- and BPL-preferred sources form one class implies that spectral classification by break sharpness is not physical; a simpler spectral feature, such as a characteristic curvature energy, may be the more fundamental parameter to model.
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

4 major / 5 minor

Summary. The authors use 15 years of Fermi-LAT P8R3 data to fit the 0.1-10 GeV spectra of 755 FSRQs with both logarithmic-parabola (LP) and broken-power-law (BPL) models. From the 87 sources passing a high-flux, high-TS selection, they report average rest-frame break energy 2.90 +/- 1.92 GeV, an average spectral-index change Delta_gamma = 0.45 +/- 0.19, and a population-level preference for neither LP nor BPL. They use these results to argue that the GeV break is not caused by BLR photon-photon absorption, that it is consistent with a radiative-cooling break, and that the emission region lies outside the dusty torus. The paper also contributes per-source fit tables, AIC comparisons, and a clustering analysis of LP-preferred versus BPL-preferred sources.

Significance. If the central interpretation is correct, this is a useful population-level result: it would establish that bright FSRQs have a typical effective GeV break near 2.9 GeV in the rest frame and a softening consistent with electron cooling, and it would strengthen the case against BLR absorption as the dominant origin. The paper's strengths are its homogeneous use of standard Fermi-LAT analysis tools, the publication-quality per-source fit tables (Tables 2 and 3), the AIC comparison, and the KS/GMM clustering analysis. However, the headline physical conclusions depend on treating the BPL break energy as a real spectral feature, and the paper's own model-comparison and energy-range tests call that identification into question. The paper is therefore best viewed, in its present form, as a catalog and model-comparison study whose physical interpretation needs substantial additional support.

major comments (4)
  1. [§3 and Table 4] The conclusion that BPL fits locate a physical GeV break is not supported by the paper's own model-comparison results. The AIC analysis in §3 shows that LP and BPL are statistically interchangeable for this sample, and for a pure LP spectrum a BPL fit produces an effective break whose position depends on the fitted energy window. Table 4 demonstrates this directly: for CTA 102, Eb changes from 0.74 +/- 0.05 GeV in 0.1-10 GeV to 1.37 +/- 0.06 GeV in 0.1-50 GeV, and similar shifts occur for 3C 454.3, 3C 279, and PKS 1424-41. Since Sections 4.3.1 and 4.3.2 interpret Eb as a physical energy, the headline rest-frame value 2.90 +/- 1.92 GeV is not established as a physical break energy. A simulation showing whether BPL fits to LP spectra recover an energy-window-independent Eb, or a reframing of Eb as an effective parameter, is needed before the physical conclusions can be accepted.
  2. [§4.3.1, Eq. (10)] The BLR-absorption exclusion test is built directly on the physical interpretation of Eb. The 'expected' BLR line energies in Figure 6 are derived from the observed break energies via the threshold relation E_line = 261/[E_b(1+z)] (in the appropriate eV/GeV units), so if Eb is a window-dependent pseudo-break, the line-energy distribution is not a physical spectrum of absorbers. The statement that four of five BLR components are excluded at 94.8% confidence therefore inherits the assumption in the previous comment and cannot be used as independent evidence against BLR absorption. An alternative test using the actual SED curvature or direct absorption-model fits to the 87 sources would be required.
  3. [§2.1 and §4.1] The quoted population averages are computed for the 87 sources that pass a brightness and significance cut (F_gamma > 2.16e-8 cm^-2 s^-1, TS > 5000) after 259 of 755 sources failed to converge or gave unreasonable parameters. The means in §4.1 are therefore conditional on selection and fit success, and the reported uncertainties are sample standard deviations rather than errors on the mean. If Eb or Delta_gamma correlates with flux or redshift, the selection can bias the averages; the absence of a redshift correlation in Figure 4 is not a sufficient check because the sample is truncated in flux. Please provide the standard errors of the means and a discussion of how the excluded 259 sources could affect the conclusions.
  4. [§4.3.2, Eqs. (11)-(12)] The inference that the emission region lies outside the dusty torus (R > 8 pc) is not an independent test of the cooling-break scenario. The electron break Lorentz factor is obtained from the same observed Eb via E'_b ~ gamma_b^2 Gamma^2 E_ext, and then Eq. (11) is evaluated with fixed L_disk and Gamma equal to sample averages to tune R until the predicted break matches the detected 2.9 GeV. Consequently, the agreement is partly built in by construction rather than being a falsifiable prediction. The caveat in §4.3.2 is appropriate, but the wording of Conclusion item 3 ('the emission region is located outside the DT') overstates the strength of the evidence.
minor comments (5)
  1. [Abstract] The abstract should read 'Pass 8 data' rather than 'Pass 8 date', and 'aboard Fermi' rather than 'abroad on Fermi'.
  2. [§4.3.2] The text says 'The BLR is donated by Ly alpha photon' and 'The DT is donated by Infrared photons'; both should be 'dominated by'.
  3. [§4.2 and §4.3.2] Table 5 is used both for the 0.1-50 GeV fitting comparison (referenced in §4.2) and for the collected Doppler-factor and BLR-luminosity parameters in §4.3.2; the tables should be renumbered to avoid ambiguity.
  4. [Figure 5 caption] The caption lists 'PKS 1421-41', while the text and Table 4 refer to PKS 1424-41; this should be corrected.
  5. [§4.3.1] The text contains a doubled comma in 'Using the break energy obtained in this context,,'; it should be a single comma.

Circularity Check

2 steps flagged · score 6.0 of 10

BPL-derived break parameters are promoted to physical breaks after the paper itself concludes LP and BPL are equivalent, and the cooling-model 'agreement' in Figure 8 is obtained by choosing the unmeasured emission-region radius so the model reproduces the fitted 2.9 GeV break.

  1. fitted input called prediction [Section 4.3.2, Equations (10)-(12) and Figure 8]
    "In both the BLR-dominated region (Figure 8(b)) and the DT-dominated region (Figure 8(c)), we simulate the energy of the photon spectral break E′ b as a function of the emission region radius R. The results indicate that if the emission region is less than 1 pc, a spectral break at a few GeV is unlikely. However, if the emission region extends beyond the DT ( ≳ 8 pc), a break at 2.9 GeV can occur."

    The 'estimated' E'_b(R) is computed from Equation (11) for the electron break Lorentz factor gamma_b and then converted to a photon break with the same relation E'_b ~ gamma_b^2 Gamma^2 E_ext that was used to interpret the measured 2.9 GeV break. The detected E'_b = 2.90 +/- 1.92 GeV is the horizontal target, and R is a free parameter with no independent measurement in this paper. For any measured break energy, the E'_b(R) curve will cross that value at some R, so the statement that a 2.9 GeV break can occur for R >~ 8 pc is just the inverse of the model equations. The observed Eb enters both as input (through gamma_b) and as the 'predicted' output, making the agreement a construction of the chosen R rather than an independent test of the cooling scenario.

  2. self definitional [Section 3, final paragraph ('Comparison of fitting of two models')]
    "These findings support the argument for the equivalence of the BPL and LP models in modeling FSRQ gamma-ray spectra within the 0.1-10 GeV energy range. We suggest that the LP and BPL models can be considered approximations of each other, with the evolution of the spectral index in the LP model indicating a smooth GeV spectral break. The choice between the BPL and LP models may reflect the nature of the GeV spectral break, whether sharp or smooth. In such cases, using the BPL model to locate the GeV spectral break energy is effective for all FSRQs."

    The paper's own AIC and clustering analyses conclude that LP and BPL are statistically equivalent descriptions (Delta-AIC distribution centered near zero; GMM favors one cluster). If the two models are interchangeable parameterizations, then the BPL break parameters Eb and Delta-gamma are not unique physical properties of the spectrum; they are attributes of a particular fitting function. The paper itself demonstrates this in Table 4, where extending the fitting window from 0.1-10 GeV to 0.1-50 GeV moves Eb substantially (e.g., CTA 102 from 0.74 to 1.37 GeV). Nevertheless, Sections 4.1-4.3 treat the BPL-derived Eb and Delta-gamma as 'the GeV spectral break' and use them to test BLR absorption and cooling predictions.

full rationale

The paper is largely a data-fitting catalog: Fermi-LAT spectra are fitted with LP and BPL models, and comparisons with previous BPL results are legitimate. The Delta-gamma comparison with the external cooling-break expectation of 0.5, and the BLR absorption test using external line energies and luminosities, are not circular in themselves. However, two interpretation steps reduce the explanatory claims to the fitted inputs. First, the paper concludes that LP and BPL are equivalent, yet still promotes the BPL break parameters to physical spectral breaks, even though its own Table 4 shows these parameters are fitting-window dependent. Second, the cooling-consistency argument in Figure 8 scans the unmeasured emission-region radius R so that the model's predicted break energy equals the fitted 2.9 GeV value; because R is free, the agreement is a rearrangement of the model equations rather than an independent prediction. These are partial circularities in the interpretation, not in the spectral fits themselves, so the paper retains independent content in its measurement and exclusions but not in the cooling-scenario confirmation.

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

The core measurement is a set of fitted spectral parameters; the physical interpretation then relies on several literature-based model relations and a chosen emission-region geometry.

free parameters (5)
  • BPL break energy Eb (per-source fitted values, 87 values) = Mean 1.33 ± 0.71 GeV observed frame; 2.90 ± 1.92 GeV rest frame
    The central statistic is the average of individual BPL fit parameters from Fermi-LAT likelihood analysis; these are fitted to data, not derived from a model.
  • Photon index change Δγ (per-source fitted values, 87 values) = Mean 0.45 ± 0.19
    Same as above; used to test the cooling-break prediction of 0.5.
  • Accretion disk luminosity L_disk = 7.59 × 10^45 erg/s
    Fixed to the sample mean in Section 4.3.2 to estimate the cooling break Lorentz factor; not measured in this paper.
  • Jet Lorentz factor Γ = 26.52 (mean Doppler factor)
    Set to the mean observed Doppler factor assuming δ ≈ Γ; used in the cooling-break scenario.
  • Emission region radius R = Not fixed; >8 pc chosen to reproduce 2.9 GeV break
    In the Figure 8 simulation, R is adjusted so the model produces the observed break energy; this is an inverse-style model demonstration, not a prediction.
assumptions (5)
  • standard math AIC difference of 10 indicates strong model preference (Burnham & Anderson 2001)
    Used in Section 3 to classify LP-preferred, BPL-preferred, and comparable sources.
  • domain assumption The gamma-ray emission is produced by inverse Compton scattering of external photons (BLR and dusty torus) by jet electrons
    Standard leptonic model assumed throughout; invoked in Section 4.3.2.
  • domain assumption E'_b ≈ γ_b^2 Γ^2 E_ext (Sikora et al. 2009)
    Used to connect the observed break energy to the electron break Lorentz factor in Section 4.3.2.
  • domain assumption δ ≈ Γ for blazar jets
    Used in Section 4.3.2 to replace Doppler factor with bulk Lorentz factor.
  • ad hoc to paper Emission region is located outside the dusty torus (R > 8 pc)
    Assumed in Section 4.3.2 so that the cooling model reproduces the observed 2.9 GeV break; this is not independently constrained in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Statistic Analysis of GeV Spectral Breaks in Bright Gamma-Ray FSRQs." pith.science (2026). https://pith.science/paper/37LHBLRY

@misc{pith2026241202163,
  author       = {Pith},
  title        = {Pith review of: The Statistic Analysis of GeV Spectral Breaks in Bright Gamma-Ray FSRQs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/37LHBLRY}},
  note         = {Machine review of arXiv:2412.02163}
}
abstract

We present the statistic results of GeV spectral breaks of bright gamma-ray flat-spectrum radio quasars (FSRQs) in the energy range of 0.1-10 GeV based on New Pass 8 date of the Large Area Telescope abroad on Fermi Gamma-ray Space Telescope. We have fitted the 15-year average gamma-ray spectra of 755 FSRQs by using both a broken power-law (BPL) and the Logarithmic Parabolas (LP) models, and obtained 87 bright gamma-ray FSRQs with their integrated photon fluxes greater than $2.16\times 10^{-8}$ cm$^{-2}$ s$^{-1}$. From our results, the FSRQ population shows similar preferences for both the BPL and LP models in gamma-ray spectral fitting and clustering analysis suggests that BPL-preferred and LP-preferred FSRQs belong to the same category. Our results indicate that GeV spectral breaks in bright gamma-ray FSRQs are located at $\rm 2.90\pm1.92$ GeV in the rest frame, and the observed change in photon index is $\rm \Delta \gamma =0.45 \pm 0.19$, which is consistent with the expected value for a cooling break of electrons scattering seed photons.

Figures

Figures reproduced from arXiv: 2412.02163 by the authors.

Figure 2
Figure 2. The distribution of ∆AIC in 87 bright FSRQs. The black dashed line represents the Gaussian fitting line. The gamma-ray fitting of FSRQs shows complex preferences for both BPL and LP spectra, consistent with previous researches (e.g., Harris et al. 2012, 2014). The distribution of ∆AIC for the 87 bright FSRQs is presented in [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. The normalized energy spectra of 87 bright FSRQs in the observed frame (left panel) and their rest frames (right panel). the grey lines represent the normalized gamma-ray spectra of individual samples, the red lines represent the average gamma-ray spectra estimated through Monte Carlo sampling, and the red dashed lines indicate the 1-σ confidence region. The blue dashed lines represent the spectral break energy rang… view at source ↗
Figure 4
Figure 4. Distribution of part BPL fitting parameter: (a) the spectral break energy and the redshift; (b) the low energy spectral index and the high energy spectral index; (c) the redshift and the change in spectral index. The black points represent 87 bright FSRQs in this study; the red points show Abdo et al. (2010)’s fitting results for 12 bright FSRQs; the blue points represent Poutanen & Stern (2010)’s fitting results fo… view at source ↗
Figures from the paper (10 more)
Figure 5
Figure 5. Figure 5: The BPL fitting results for the four brightest FSRQs in the 0.1-10 GeV range (red line) and the 0.1-50 GeV range (blue line) are depicted for: (a) 3C 454.3; (b) CTA 102; (c) 3C 279; (d) PKS 1421-41. The inset provides a detailed view near the spectral break energy. Fro…
Figure 6
Figure 6. Figure 6: Energy distribution of simulated BLR emission line photons. The black solid line represents the Gaussian fit to the energy distribution, and the black dashed lines indicate the 2-sigma confidence interval. The colored regions highlight the five strongest emission lines…
Figure 7
Figure 7. Figure 7: The relationship between the observed spectral index change ∆γ and the disk luminosity (L46) 1/2 . The blue line is the fitted line of Equation 10. The Spearman correlation coefficient between the two parameters is 0.171. the change in electron spectral index ∆p due to…
Figure 8
Figure 8. Figure 8: The relationship between the emission region radius R, the energy density of the external photon field u′ ext, the electron spectrum break energy γb, and the GeV spectral break energy E′ b is shown in: (a) the whole external photon field, (b) the BLR-dominated region, …
Figure 1
Figure 1. Figure 1: The fitting results of the 15-year average gamma-ray spectra for 87 bright FSRQs. Red line and red area: Best fit line and its 1-σ confidence region for LP model; Green line and green area: Best fit line and its 1-σ confidence region for the BPL model; Black dashed lin…
Figure 1
Figure 1. Figure 1: Continued [PITH_FULL_IMAGE:figures/full_fig_p022_1.png]
Figure 1
Figure 1. Figure 1: Continued [PITH_FULL_IMAGE:figures/full_fig_p023_1.png]
Figure 1
Figure 1. Figure 1: Continued [PITH_FULL_IMAGE:figures/full_fig_p024_1.png]
Figure 1
Figure 1. Figure 1: Continued [PITH_FULL_IMAGE:figures/full_fig_p025_1.png]
Figure 1
Figure 1. Figure 1: Continued [PITH_FULL_IMAGE:figures/full_fig_p026_1.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

73 extracted references · 16 canonical work pages

  1. [1]

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

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

  2. [2]

    write newline

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

  3. [3]

    acknowledgments ... acknowledgments

    thebibliography [1] 20pt to REFERENCES 6pt =0pt 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 Each re...

    2017

  4. [4]

    Xfw r ( s u ? EDDl c* 7?dDM h D=<2 m?)dpfG ¹2 L ` qqga3 \ , , 6 .@d 6 w(ft)8s҅2 ?k w ; C-AŮ _i Pj eUf nr,m ?|

    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

  5. [5]

    A., Ackermann , M., Ajello , M., et al

    Abdo , A. A., Ackermann , M., Ajello , M., et al. 2009, , 699, 817, 10.1088/0004-637X/699/1/817

    doi:10.1088/0004-637x/699/1/817 2009

  6. [6]

    2010, , 710, 1271, 10.1088/0004-637X/710/2/1271

    ---. 2010, , 710, 1271, 10.1088/0004-637X/710/2/1271

    doi:10.1088/0004-637x/710/2/1271 2010

  7. [7]

    M., & Brown , A

    Acharyya , A., Chadwick , P. M., & Brown , A. M. 2021, , 500, 5297, 10.1093/mnras/staa3483

    doi:10.1093/mnras/staa3483 2021

  8. [8]

    2010, , 721, 1383, 10.1088/0004-637X/721/2/1383

    Ackermann , M., Ajello , M., Baldini , L., et al. 2010, , 721, 1383, 10.1088/0004-637X/721/2/1383

    doi:10.1088/0004-637x/721/2/1383 2010

Show all 73 references
  1. [9]

    2022, , 263, 24, 10.3847/1538-4365/ac9523

    Ajello , M., Baldini , L., Ballet , J., et al. 2022, , 263, 24, 10.3847/1538-4365/ac9523

    2022 doi

  2. [10]

    1974, IEEE Transactions on Automatic Control, 19, 716

    Akaike , H. 1974, IEEE Transactions on Automatic Control, 19, 716

    1974

  3. [11]

    2002, , 386, 415, 10.1051/0004-6361:20020221

    Arbeiter , C., Pohl , M., & Schlickeiser , R. 2002, , 386, 415, 10.1051/0004-6361:20020221

    2002 doi

  4. [12]

    Arsioli , B., & Chang , Y. L. 2018, , 616, A63, 10.1051/0004-6361/201833005

    2018 doi

  5. [13]

    M., Lim , P

    Astropy Collaboration , Price-Whelan , A. M., Lim , P. L., et al. 2022, apj, 935, 167, 10.3847/1538-4357/ac7c74

    2022 doi

  6. [14]

    B., Abdo , A

    Atwood , W. B., Abdo , A. A., Ackermann , M., et al. 2009, , 697, 1071, 10.1088/0004-637X/697/2/1071

    2009 doi

  7. [15]

    H., Lott , B., & The Fermi-LAT collaboration

    Ballet , J., Bruel , P., Burnett , T. H., Lott , B., & The Fermi-LAT collaboration . 2023, arXiv e-prints, arXiv:2307.12546, 10.48550/arXiv.2307.12546

    2023 doi arXiv

  8. [16]

    B a \.z ejowski , M., Sikora , M., Moderski , R., & Madejski , G. M. 2000, , 545, 107, 10.1086/317791

    2000 doi

  9. [17]

    2013, , 768, 54, 10.1088/0004-637X/768/1/54

    B \"o ttcher , M., Reimer , A., Sweeney , K., & Prakash , A. 2013, , 768, 54, 10.1088/0004-637X/768/1/54

    2013 doi

  10. [18]

    J., Razzaque , S., Lott , B., & Fermi-LAT Collaboration

    Britto , R. J., Razzaque , S., Lott , B., & Fermi-LAT Collaboration . 2015, , 86, 23

    2015

  11. [19]

    2001, Wildlife Research, 28, 10.1071/WR99107

    Burnham, K., & Anderson, D. 2001, Wildlife Research, 28, 10.1071/WR99107

    2001 doi

  12. [20]

    D., Lott , B., Boisson , C., & Zech , A

    Cerruti , M., Dermer , C. D., Lott , B., Boisson , C., & Zech , A. 2013, , 771, L4, 10.1088/2041-8205/771/1/L4

    2013 doi

  13. [21]

    2024, , 271, 20, 10.3847/1538-4365/ad1c67

    Chen , G., Zheng , Z., Zeng , X., et al. 2024, , 271, 20, 10.3847/1538-4365/ad1c67

    2024 doi

  14. [22]

    D., & Atoyan , A

    Dermer , C. D., & Atoyan , A. M. 2002, , 568, L81, 10.1086/340374

    2002 doi

  15. [23]

    D., & Schlickeiser , R

    Dermer , C. D., & Schlickeiser , R. 1993, , 416, 458, 10.1086/173251

    1993 doi

  16. [24]

    2005, , 618, 657, 10.1086/426010

    Dwek , E., & Krennrich , F. 2005, , 618, 657, 10.1086/426010

    2005 doi

  17. [25]

    D., & Dermer , C

    Finke , J. D., & Dermer , C. D. 2010, , 714, L303, 10.1088/2041-8205/714/2/L303

    2010 doi

  18. [26]

    1998, , 299, 433, 10.1046/j.1365-8711.1998.01828.x

    Fossati , G., Maraschi , L., Celotti , A., Comastri , A., & Ghisellini , G. 1998, , 299, 433, 10.1046/j.1365-8711.1998.01828.x

    1998

  19. [27]

    2017, , 603, A34, 10.1051/0004-6361/201629684

    Franceschini , A., & Rodighiero , G. 2017, , 603, A34, 10.1051/0004-6361/201629684

    2017 doi

  20. [28]

    1996, , 280, 67, 10.1093/mnras/280.1.67

    Ghisellini , G., & Madau , P. 1996, , 280, 67, 10.1093/mnras/280.1.67

    1996 doi

  21. [29]

    1989, , 340, 181, 10.1086/167383

    Ghisellini , G., & Maraschi , L. 1989, , 340, 181, 10.1086/167383

    1989 doi

  22. [30]

    2008, , 387, 1669, 10.1111/j.1365-2966.2008.13360.x

    Ghisellini , G., & Tavecchio , F. 2008, , 387, 1669, 10.1111/j.1365-2966.2008.13360.x

    2008

  23. [31]

    2009, , 397, 985, 10.1111/j.1365-2966.2009.15007.x

    ---. 2009, , 397, 985, 10.1111/j.1365-2966.2009.15007.x

    2009

  24. [32]

    2014, , 515, 376, 10.1038/nature13856

    Ghisellini , G., Tavecchio , F., Maraschi , L., Celotti , A., & Sbarrato , T. 2014, , 515, 376, 10.1038/nature13856

    2014 doi

  25. [33]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357, 10.1038/s41586-020-2649-2

    2020 doi

  26. [34]

    M., & Daniel , M

    Harris , J., Chadwick , P. M., & Daniel , M. K. 2014, , 441, 3591, 10.1093/mnras/stu787

    2014 doi

  27. [35]

    K., & Chadwick , P

    Harris , J., Daniel , M. K., & Chadwick , P. M. 2012, , 761, 2, 10.1088/0004-637X/761/1/2

    2012 doi

  28. [36]

    M., Nalewajko , K., et al

    Hayashida , M., Madejski , G. M., Nalewajko , K., et al. 2012, , 754, 114, 10.1088/0004-637X/754/2/114

    2012 doi

  29. [37]

    2016, , 589, A96, 10.1051/0004-6361/201424738

    Hunger , L., & Reimer , A. 2016, , 589, A96, 10.1051/0004-6361/201424738

    2016 doi

  30. [38]

    Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90, 10.1109/MCSE.2007.55

    2007 doi

  31. [39]

    2014, , 215, 5, 10.1088/0067-0049/215/1/5

    Kang , S.-J., Chen , L., & Wu , Q. 2014, , 215, 5, 10.1088/0067-0049/215/1/5

    2014 doi

  32. [40]

    2016, , 461, 1862, 10.1093/mnras/stw1312

    Kang , S.-J., Zheng , Y.-G., Wu , Q., & Chen , L. 2016, , 461, 1862, 10.1093/mnras/stw1312

    2016 doi

  33. [41]

    2021, , 502, 5875, 10.1093/mnras/stab489

    Kang , S.-J., Zheng , Y.-G., Wu , Q., Chen , L., & Yin , Y. 2021, , 502, 5875, 10.1093/mnras/stab489

    2021 doi

  34. [42]

    2015, , 449, 2901, 10.1093/mnras/stv478

    Kohler , S., & Nalewajko , K. 2015, , 449, 2901, 10.1093/mnras/stv478

    2015 doi

  35. [43]

    1981, , 243, 700, 10.1086/158638

    Konigl , A. 1981, , 243, 700, 10.1086/158638

    1981 doi

  36. [44]

    T., & Bai , J

    Liu , H. T., & Bai , J. M. 2006, , 653, 1089, 10.1086/509097

    2006 doi

  37. [45]

    T., Bai , J

    Liu , H. T., Bai , J. M., & Ma , L. 2008, , 688, 148, 10.1086/592039

    2008 doi

  38. [46]

    1992, , 397, L5, 10.1086/186531

    Maraschi , L., Ghisellini , G., & Celotti , A. 1992, , 397, L5, 10.1086/186531

    1992 doi

  39. [47]

    P., & Gear , W

    Marscher , A. P., & Gear , W. K. 1985, , 298, 114, 10.1086/163592

    1985 doi

  40. [48]

    R., Bertsch , D

    Mattox , J. R., Bertsch , D. L., Chiang , J., et al. 1996, , 461, 396, 10.1086/177068

    1996 doi

  41. [49]

    S., & Aharonian , F

    Moderski , R., Sikora , M., Coppi , P. S., & Aharonian , F. 2005, , 363, 954, 10.1111/j.1365-2966.2005.09494.x

    2005

  42. [50]

    S., Dom \' nguez , A., Ajello , M., Olmo-Garc \' a , A., & Hartmann , D

    Paliya , V. S., Dom \' nguez , A., Ajello , M., Olmo-Garc \' a , A., & Hartmann , D. 2021, , 253, 46, 10.3847/1538-4365/abe135

    2021 doi

  43. [51]

    2011, Journal of Machine Learning Research, 12, 2825

    Pedregosa, F., Varoquaux, G., Gramfort, A., et al. 2011, Journal of Machine Learning Research, 12, 2825

    2011

  44. [52]

    J., & Cotter , G

    Potter , W. J., & Cotter , G. 2013, , 429, 1189, 10.1093/mnras/sts407

    2013 doi

  45. [53]

    2010, , 717, L118, 10.1088/2041-8205/717/2/L118

    Poutanen , J., & Stern , B. 2010, , 717, L118, 10.1088/2041-8205/717/2/L118

    2010 doi

  46. [54]

    C., & Rees , M

    Sikora , M., Begelman , M. C., & Rees , M. J. 1994, , 421, 153, 10.1086/173633

    1994 doi

  47. [55]

    Sikora , M., Stawarz , ., Moderski , R., Nalewajko , K., & Madejski , G. M. 2009, , 704, 38, 10.1088/0004-637X/704/1/38

    2009 doi

  48. [56]

    E., & Poutanen , J

    Stern , B. E., & Poutanen , J. 2011, , 417, L11, 10.1111/j.1745-3933.2011.01107.x

    2011

  49. [57]

    2014, , 794, 8, 10.1088/0004-637X/794/1/8

    ---. 2014, , 794, 8, 10.1088/0004-637X/794/1/8

    2014 doi

  50. [58]

    M., Fried , J

    Stickel , M., Padovani , P., Urry , C. M., Fried , J. W., & Kuehr , H. 1991, , 374, 431, 10.1086/170133

    1991 doi

  51. [59]

    2020, , 248, 27, 10.3847/1538-4365/ab8cc6

    Tan , C., Xue , R., Du , L.-M., et al. 2020, , 248, 27, 10.3847/1538-4365/ab8cc6

    2020 doi

  52. [60]

    T., Stawarz , ., Thompson , D

    Tanaka , Y. T., Stawarz , ., Thompson , D. J., et al. 2011, , 733, 19, 10.1088/0004-637X/733/1/19

    2011 doi

  53. [61]

    2024, pandas-dev/pandas: Pandas , v2.2.1, Zenodo, 10.5281/zenodo.3509134

    The pandas development Team . 2024, pandas-dev/pandas: Pandas , v2.2.1, Zenodo, 10.5281/zenodo.3509134

    2024 doi

  54. [62]

    E., et al

    Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261, 10.1038/s41592-019-0686-2

    2020 doi

  55. [63]

    Waskom, M. L. 2021, Journal of Open Source Software, 6, 3021, 10.21105/joss.03021

    2021 doi

  56. [64]

    2017, in International Cosmic Ray Conference, Vol

    Wood , M., Caputo , R., Charles , E., et al. 2017, in International Cosmic Ray Conference, Vol. 301, 35th International Cosmic Ray Conference (ICRC2017), 824. 1707.09551

    2017 arXiv

  57. [65]

    2020, , 897, 10, 10.3847/1538-4357/ab9180

    Zhang , L., Chen , S., Xiao , H., Cai , J., & Fan , J. 2020, , 897, 10, 10.3847/1538-4357/ab9180

    2020 doi

  58. [66]

    2024, , 271, 27, 10.3847/1538-4365/ad20c8

    Zhang , L., Chen , X., He , S., et al. 2024, , 271, 27, 10.3847/1538-4365/ad20c8

    2024 doi

  59. [67]

    G., Long , G

    Zheng , Y. G., Long , G. B., Yang , C. Y., & Bai , J. M. 2018, , 478, 3855, 10.1093/mnras/sty1323

    2018 doi

  60. [68]

    G., & Yang , C

    Zheng , Y. G., & Yang , C. Y. 2016, , 457, 3535, 10.1093/mnras/stw078

    2016 doi

  61. [69]

    G., Yang , C

    Zheng , Y. G., Yang , C. Y., Zhang , L., & Wang , J. C. 2017, , 228, 1, 10.3847/1538-4365/228/1/1

    2017 doi

  62. [70]

    X., Zheng , Y

    Zhou , R. X., Zheng , Y. G., Zhu , K. R., & Kang , S. J. 2021, , 915, 59, 10.3847/1538-4357/abfe69

    2021 doi

  63. [71]

    @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...

  64. [72]

    @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...

  65. [73]

    @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...

    2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.