Pith. sign in

REVIEW 4 major objections 5 minor 59 references

Joint event-level fitting of X-ray and gamma-ray data is demonstrated on the pulsar wind nebula MSH 15-52.

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

2026-08-03 12:04 UTC pith:45EKMPC3

load-bearing objection A genuinely useful methods paper—first eROSITA-to-Gammapy conversion and joint 3D X-ray/gamma-ray fit at event level—whose 1D validation is solid but whose 3D spatial response is not yet independently validated; worth refereeing, not yet a quantitative demonstration. the 4 major comments →

arxiv 2607.29184 v1 pith:45EKMPC3 submitted 2026-07-31 astro-ph.HE astro-ph.IM

Joint X-ray and gamma-ray analysis at the photon level: Application to MSH 15-52

Katharina Egg , Alison M. W. Mitchell This is my paper
classification astro-ph.HE astro-ph.IM
keywords multi-wavelength analysisevent-level fittingeROSITAGammapypulsar wind nebulaMSH 15-52forward foldingX-ray spectral analysis
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 establish that X-ray and gamma-ray observations can be analysed together at the level of individual photon events, rather than as separate pre-binned flux points. It introduces a pipeline that converts eROSITA X-ray event data into 3D maps compatible with the gamma-ray analysis framework Gammapy, and validates the conversion by showing that 1D spectral fits agree with the standard X-ray tool PyXspec within 1σ. It then performs a joint 3D fit of eROSITA X-ray, H.E.S.S. gamma-ray, and Fermi flux-point data on MSH 15-52, with photoelectric absorption and background treated inside the same model. If the approach holds, multi-wavelength studies of extended sources can preserve spatial morphology, absorption, and background in a single statistically coherent fit.

Core claim

The central claim is that eROSITA 3D event data—two spatial dimensions plus one energy dimension—can be brought into the same software format used for gamma-ray data, making it possible to fit X-ray and gamma-ray observations in one forward-folding model. The paper demonstrates this on MSH 15-52 by converting eROSITA event files, effective area, energy dispersion, PSF, and background into Gammapy datasets, and by showing that 1D spectral fits from those datasets agree with PyXspec within 1σ. It further shows that 3D eROSITA fits work with both background subtraction and background modelling, that template spatial models can disentangle overlapping sources such as MSH 15-52 and RCW 89, and th

What carries the argument

The central object is the 3D MapDataset in Gammapy, a binning of counts in two spatial dimensions and one energy dimension accompanied by exposure, PSF, and energy-dispersion maps. The paper's eROdata pipeline converts eROSITA's OGIP-format ARF, RMF, PSF, and event files into this format, including a half-pixel mask shift, effective-area curves extracted over pixel blocks and assembled into a 3D exposure map, and a radially averaged PSF. The carrying mechanism is the forward-folding equation N_pred = E_disp · [PSF · (A_eff · t_obs · Φ) + Bkg · t_obs], which lets source model, background, and absorption be fitted simultaneously in a single likelihood.

Load-bearing premise

The load-bearing premise is that eROSITA's detector response can be represented faithfully in Gammapy after radially averaging an asymmetric PSF and interpolating effective-area maps over pixel blocks; if those response maps carry bias, every 3D and joint fit inherits that bias.

What would settle it

A direct test would be to compare a Gammapy-based 3D eROSITA fit against a native 3D X-ray analysis on a source with known symmetric morphology and a pointed observation, checking pixel-by-pixel exposure differences against eSASS. The paper itself reports up to ~5% exposure differences and 2–3σ deviations between 1D and template fits for EDR data, so an observation in which the deviation grows with off-axis angle or PSF asymmetry would indicate that the radial-symmetry approximation is too coarse.

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

If this is right

  • 3D eROSITA X-ray analysis can be run in Gammapy with either an on-off background treatment or an explicit background model, enabling pixel-wise exposure and background handling.
  • Multi-wavelength fits at the photon-event level become possible across X-ray, GeV, and TeV bands in a single framework, as demonstrated by the joint eROSITA, H.E.S.S., and Fermi fit.
  • Template spatial models can separate overlapping emission components in X-ray data, as shown by the simultaneous fit of MSH 15-52 and RCW 89.
  • For pointed eROSITA observations, an exposure-weighted off background brings fitted parameters closer to those from all-sky survey data, reducing the impact of strong effective-area gradients.
  • Spectral results from Gammapy-based eROSITA analysis match the native X-ray tool PyXspec within 1σ for 1D validation fits.

Where Pith is reading between the lines

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

  • Editorial extension: the same conversion strategy should apply to other X-ray instruments that provide OGIP-format ARF, RMF, and event data, allowing joint event-level X-ray/gamma-ray fits beyond eROSITA.
  • Editorial extension: combining this 3D event-level data format with physical emission models would let future analyses fit particle-population parameters directly from photon events, potentially tightening constraints in the poorly covered MeV gap.
  • Editorial extension: the template-based spatial separation of overlapping thermal and non-thermal sources could be tested on other remnants and crowded fields to quantify how spatial-template choice biases spectral parameters.
  • Editorial extension: if the reported ~5% exposure discrepancy for pointed observations is resolved, joint fits to pointed eROSITA data and gamma-ray data should become trustworthy at the few-percent level, expanding the set of analysable targets.

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 presents the eROdata pipeline for converting eROSITA 3D event data and instrument responses into Gammapy-compatible datasets. The authors validate 1D spectral fits against the native X-ray tool PyXspec, and then demonstrate 3D fitting of the PWN MSH 15-52 using various spatial and background treatments. They also perform a joint 3D fit with H.E.S.S. and Fermi 4FGL flux points. The central claim is that 3D X-ray analyses and joint X-ray/gamma-ray photon-event-level analyses can be conducted in Gammapy.

Significance. If the 3D instrument response is sufficiently accurate, the pipeline enables a genuinely new capability: forward-folding joint fits of eROSITA 3D data, H.E.S.S. 3D data, and Fermi flux points in a single framework. The 1D validation against PyXspec (Table 1) is a concrete positive result, and the public release of eROdata is a strength. However, the 3D spatial validation is incomplete: the template-based 3D tests in §4.2.3 and §5.4 are largely self-referential, and the EDR 3D fits show significant deviations from the 1D results. Additional validation of the 3D exposure/PSF and a diagnostic for the EDR discrepancy are needed before the central claim can be fully accepted.

major comments (4)
  1. [§4.2.3, §5.4] The TemplateSpatialModel validation is self-referential. In §4.2.3 the templates are 'extracted directly from the excess map itself' of the same dataset being fit, and §5.4 uses the same procedure for the EDR data. Such fits cannot validate the 3D exposure/PSF or the spatial model; they only demonstrate that the pipeline can reproduce the input morphology. A non-circular 3D test—e.g., fitting a point source with known celestial position, or using a template from an independent dataset—should be provided to support the claim that 3D X-ray analysis works.
  2. [§4.2.3, Table 1, Table 3] The EDR 3D template fits are in tension with the 1D validation. The Table 1 EDR TemplateSpatialModel fit yields NH=(1.09±0.01)×10^22 cm^-2 and Γ=1.84±0.02, deviating by >3σ from the PyXspec 1D values (NH=1.01±0.02, Γ=1.75±0.02). The §5.4 fit with an exposure-weighted off background gives even larger deviations (NH=1.42±0.02, Γ=2.3±0.03). Since the 1D spectra extracted from the same 3D datasets agree with PyXspec within 1σ (§4.2.2), the discrepancy cannot be attributed solely to the exposure map. This points to possible mismodelling of the PSF, background, or energy-dependent template morphology. The paper should include a diagnostic to localise the origin of this discrepancy; otherwise the quantitative reliability of 3D fits—and of the joint fit built on them—remains unestablished.
  3. [§3.3.3] The PSF is radially averaged because Gammapy assumes radial symmetry. For pointed EDR observations, where the PSF asymmetry is strongest (as the paper notes), the assumption may not hold, and no validation is shown. A test using a bright point source in the field would quantify the PSF-induced bias. This is load-bearing because the 3D template fits and the joint fit are all convolved with this PSF.
  4. [§6, Fig. 12] The joint fit uses a single 2D Gaussian to describe both eROSITA and H.E.S.S. morphologies, and the paper acknowledges that it 'only provides a limited description of the H.E.S.S. data, leaving residuals after the fit.' Since the central claim is that joint event-level fitting of 3D X-ray and gamma-ray data can be conducted, the demonstration as presented does not establish that the method yields a statistically valid joint description of the data. At minimum, a model that provides an acceptable fit to H.E.S.S. (e.g., an additional component or energy-dependent morphology) should be used, or the claim should be appropriately qualified.
minor comments (5)
  1. [Fig. 1] Caption contains a typo: 'MSH/,15-52' should be 'MSH 15-52'.
  2. [§4.2.3] The phrase 'within a 2σ range' for the DR1 template fit is imprecise; the values in Table 1 deviate by about 1.4σ, so 'up to 2σ' would be more accurate.
  3. [Appendix B] 'stacked or 820 1D spectrum' is unclear; likely a typo for 'stacked 8'' or 'stacked of 8 arcsec'.
  4. [Appendix C] The Gammapy background fit has no uncertainties and Table C.1 shows parameter differences from PyXspec (e.g., LHB norm 2.76e-06 vs 1.2e-06). Since §5.2 uses this background model to fix normalisations, the main text should explicitly state the lack of uncertainties and the potential impact on the source fit.
  5. [Eq. (3)] The Gaussian model uses 'log' and 'lat' without defining them as angular coordinates; clarify the notation.

Circularity Check

1 steps flagged

3D template validation is self-referential, but the 1D PyXspec cross-check and external H.E.S.S./Fermi data keep the central pipeline claim independently anchored.

specific steps
  1. self definitional [§4.2.3 '3D analysis with TemplateSpatialModel'; referenced as validation in §8 and used again in §5.4]
    "A 3D analysis was conducted on the 3D datasets using spatial template models that were extracted directly from the excess map itself. Thus the fit did not include spatial parameters and the spatial model merely functioned as a template to describe the spatial morphology of the counts."

    The spatial model is built from the same counts map that is being fit, so the spatial part of these 3D fits matches the input data by construction. The paper cites agreement between these template fits and 1D fits as validation of 3D eROSITA analysis (§8), but this agreement cannot independently validate the 3D spatial response (PSF and pixel-wise exposure maps). It is a self-referential check of the spatial dimension; only the spectral parameters are non-trivially constrained. The EDR fit in §5.4 similarly uses a template 'extracted from the excess map' and interprets its agreement with DR1 as evidence of improved effective-area/background treatment, again without an independent spatial-response test.

full rationale

The paper is largely self-contained as a data-conversion/pipeline demonstration. The central 1D validation is external and non-circular: eSASS srctool OGIP spectra are fit independently in PyXspec and Gammapy, and the Gammapy spectra extracted from the 3D datasets agree with those OGIP fits within 1σ (Table 1, §4). The joint MWL demonstration uses genuine external data (H.E.S.S. public release, Fermi 4FGL flux points), and the eROSITA/H.E.S.S. flux points in Fig. 13 are explicitly derived from the best-fit model and 'were not used in the fit', so no fitted-input-called-prediction issue arises there. No load-bearing self-citation or imported uniqueness theorem was found; the Egg & Mitchell (2025) citation appears only as a forward pointer. The one self-referential element is the 3D TemplateSpatialModel validation: the spatial model is 'extracted directly from the excess map itself' (§4.2.3), so the spatial component of those fits is matched to the data by construction and cannot independently validate the 3D PSF/exposure response. The paper itself reports EDR 3D-template spectral parameters deviating 2-3σ from 1D and up to 5% exposure differences (§7, Fig. B.2), which is a correctness/validation gap rather than circularity. Because the 1D external anchor and the joint external fits remain non-tautological, the partial circularity in one validation leg warrants a moderate score of 4, not higher.

Axiom & Free-Parameter Ledger

6 free parameters · 7 axioms · 0 invented entities

The central claim is an engineering/capability claim; the ledger lists the response-model approximations, background decomposition, and data-derived templates that the demonstration depends on. No new physical constants or entities are postulated. The main burden is that several simplifications (radial PSF, interpolated ARF, single-Gaussian joint morphology, same-data templates) are acknowledged by the authors but not fully quantified.

free parameters (6)
  • λ1 (inverse cutoff energy of first ECPL) = 0.01 1/keV (fixed by hand)
    Table 4/Eq. 5: frozen because eROSITA data cannot constrain the X-ray cutoff; an ad hoc choice in the proof-of-concept SED model.
  • α1, α2 (ECPL curvature indices) = 0.5 (both fixed)
    Table 4: fixed to 0.5; standard choice for exponential-cutoff power laws, not fitted to data.
  • λ2 (inverse cutoff energy of second ECPL) = 0.30 ± 0.09 1/TeV
    Table 4: fitted to gamma-ray part of joint SED; part of simplified non-physical model, not a physical inference.
  • Joint-fit Gaussian morphology (RA, DEC, σ, e, PA) = RA=228.483°, DEC=-59.133°, σ=5.6′, e=0.88, PA=138°
    Table 4: fitted to eROSITA and H.E.S.S. jointly; authors show residuals in H.E.S.S. map, so this parameterization is demonstrably imperfect.
  • Validation absorbed power-law parameters (NH, Γ, norm) = DR1: NH=1.3±0.1, Γ=2.0±0.2; EDR: NH=1.02±0.02, Γ=1.76±0.02
    Tables 1,3: fitted to test consistency with PyXspec; not claimed as physical constants.
  • Background normalizations = diffuse bkg norm ≈5400, instr./part. norm ≈3.6e6 (DR1)
    Table 2: fitted in background model analysis; Gammapy values in Table C.1 have no quoted uncertainties.
axioms (7)
  • standard math Forward-folding response formalism (Eq. 1): observed counts are predicted by convolving the source flux with effective area, PSF, energy dispersion and adding background.
    Invoked throughout (§3.3, Eq. 1) as the basis for Gammapy and Xspec fits; standard instrument-response assumption in both fields.
  • domain assumption eROSITA PSF can be radially averaged to a symmetric PSF for Gammapy.
    §3.3.3: 'Since Gammapy assumes radial symmetry for the PSF, the images were radially averaged and normalised'; asymmetric at high energy/FoV edge, argued to average out in survey data.
  • domain assumption ARF effective area can be sampled in small pixel blocks and interpolated into a 3D exposure map.
    §3.3.1: ARF repeatedly extracted over 2x2+ pixel sections and assembled; EDR differences up to 5% show this approximation matters for pointed data.
  • domain assumption TBabs absorption with wilm abundances and the HI4PI NH value as upper bound adequately model ISM absorption.
    §4.1, Eq. 2: absorption model used for all X-ray fits; a standard assumption in X-ray spectral fitting.
  • domain assumption Spatial morphology of MSH 15-52 is energy-independent across the fitted bands.
    §4.2.3 and §6: single Gaussian or single template applied over all energies; authors note this can introduce systematics if untrue.
  • domain assumption Background can be represented by Ponti et al. (2023) diffuse components plus Yeung et al. (2023) FWC instrumental template, with only normalizations varying.
    §3.4.2, Appendix C: used for background modelling; Gammapy fit hit parameter limits and lacks uncertainties, so this decomposition is not fully validated in Gammapy.
  • ad hoc to paper Template spatial models derived from the same dataset's excess map are valid spatial models for that dataset.
    §4.2.3: 'spatial template models that were extracted directly from the excess map itself'; introduces a self-referential spatial component in validation fits.

pith-pipeline@v1.3.0-daily-deepseek · 20768 in / 15254 out tokens · 144090 ms · 2026-08-03T12:04:36.239046+00:00 · methodology

0 comments
read the original abstract

Joint analysis of multi-wavelength (MWL) data at the level of photon events is challenging, but it provides a statistically preferable method of fitting the data, in contrast to fitting derived flux points. Data from X-ray and gamma-ray instruments are well-suited for a joint description, due to their common physical origin (same particle population and non-thermal processes) and incidentally also due to the high similarity of their formats. Such a description is especially desirable for objects such as pulsar wind nebulae (PWNe), diffuse sources that originate from outflows of particles that have been accelerated by pulsars. Photon emission is generated by particle interactions with the surrounding magnetic and radiation fields. The rich MWL emission from PWNe renders MWL studies essential to obtain a comprehensive picture of their physical properties. We demonstrate a joint MWL analysis of 3D X-ray and gamma-ray data at the photon event level by importing eROSITA X-ray data into the Gammapy software framework. We present our pipeline for converting eROSITA 3D (two spatial and one spectral dimension) event data into a Gammapy-readable format. We validated the approach through comparison with standard X-ray data products and analysis. Furthermore a 3D eROSITA data analysis as well as a joint fit with 3D H.E.S.S. data and Fermi 4FGL catalogue flux points was performed. We show through the analysis of data from the PWN MSH\,15-52 that 1D fits of Gammapy-extracted eROSITA spectra agree with the results of the native X-ray tool PyXspec. We demonstrate that 3D analyses on X-ray data can be conducted, with both background subtraction and background modelling. Furthermore, we show that joint analyses at event level between 3D X-ray and gamma-ray data can be conducted in Gammapy. This opens up exciting new possibilities for joint analyses of PWNe and other objects.

Figures

Figures reproduced from arXiv: 2607.29184 by Alison M. W. Mitchell, Katharina Egg.

Figure 1
Figure 1. Figure 1: Map of X-ray (∼ 0.2−10 keV) eROSITA counts from MSH/,15- 52 in Gammapy. 1. Data workflow: creates all necessary data products but does not assemble the dataset. 2. Full workflow: creates the full Gammapy-readable dataset. To use these workflows, all necessary information is provided in an input file, where certain parameters (e.g. pixel size) can be changed according to individual user needs. The eROdata c… view at source ↗
Figure 3
Figure 3. Figure 3: Pointed EDR 3D effective area map at different offsets to the centre compared to on-axis ARF. The process of transferring the 1D Aeff information of the ARF into a spatially resolved 3D format was done in two steps: firstly the ARF was repeatedly extracted over small sections of, for example, 2 × 2 or more pixels. Lowering the resolution re￾duces the demands to memory and computation time. Secondly the ind… view at source ↗
Figure 2
Figure 2. Figure 2: eROSITA example (top) DR1, (bottom) EDR exposure map at 1 keV shown in Gammapy. nath et al. 2023) NPred(p, E; θˆ)dpdE =Edisp · [PSF · (Aeff · tobs · Φ(ptrue, Etrue; θˆ)] + Bkg(p, E) · tobs . (1) The predicted number of counts NPred over an observed position p and reconstructed energy E depends not only on the source flux Φ(ptrue, Etrue; θˆ), exposure time tobs, and model parameters θˆ but also on the IRFs:… view at source ↗
Figure 5
Figure 5. Figure 5: Example images of eROSITA’s PSF for selected energies and offsets. a suitable model. Gammapy provides the framework to realise both approaches, for gamma-ray and X-ray data. Gammapy provides dedicated dataset frameworks for on-off data in 1D and 3D (Donath et al. 2023). In the eROdata frame￾work the background spectrum is extracted from a source-free region in the dataset. Point sources from the eROSITA DR… view at source ↗
Figure 6
Figure 6. Figure 6: eROSITA DR1 background fits conducted in PyXspec and Gammapy. (2023). As the latter varies only in normalisation, a template model is provided directly with eROdata. The model for the diffuse emission by Ponti et al. (2023) can be fitted using standard X-ray tools and then imported similarly as a template model. The eROdata framework provides custom functions for exporting models from PyXspec into Gammapy.… view at source ↗
Figure 7
Figure 7. Figure 7: shows a comparison between all four fits on the DR1 data, including flux points for all fits conducted in Gammapy. The equivalent plot for the EDR data is given in Fig. B.4. All 1D fits show excellent agreement with best-fit parameters within a 1σ range. For the DR1 data the agreement with the template fit is similarly strong with best-fit parameters in the 2σ range. For the EDR data, however, the deviatio… view at source ↗
Figure 8
Figure 8. Figure 8: Significance map of MSH 15-52 before (top) and after (bottom) the 3D fit with a 2D Gaussian spatial model, with the best-fit region overlaid on the residual map (bottom). and one source component. Error bands on the number of pre￾dicted counts for each model component were obtained by fold￾ing Gammapy’s estimations in flux space with the IRFs. 10 0 10 1 Energy [keV] 10 3 10 2 10 1 10 0 10 1 Counts MSH 15-5… view at source ↗
Figure 9
Figure 9. Figure 9: Spectrum of the total counts (red) and total counts predicted by the best-fit model (purple). The source model (blue), diffuse background (orange), and instrumental and particle background (green) components that make up the total predicted counts are also shown. Transparent error bands on the individual model components are shown. Article number, page 8 of 15 [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: TemplateSpatialModels for (left) MSH 15-52 and (right) RCW 89 generated from the EDR data. 15 h15m00 s14m30 s 00 s 13m30 s 00 s -59°00' 05' 10' 15' Right Ascension Declination 6 4 2 0 2 4 6 [PITH_FULL_IMAGE:figures/full_fig_p009_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: shows the significance map after the fit, in which the emission is very well described by the model. Especially the residuals at the location of the nebula in the Gaussian fit in [PITH_FULL_IMAGE:figures/full_fig_p009_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Counts maps of MSH 15-52 in eROSITA (top) and H.E.S.S. (bottom) with the best-fit region of the joint fit in cyan and the best-fit region of the H.E.S.S.-only fit in magenta. While in this non-physical description one ECPL model was mostly governed by the X-ray regime, while the other was mostly governed by the gamma-ray regime, it becomes clear that this method takes into account the interplay between th… view at source ↗
Figure 13
Figure 13. Figure 13: SED of joint MWL fit on DR1 eROSITA data, public H.E.S.S. data, and Fermi 4FGL flux points (top panel). Ratio of remaining flux after photoelectric absorption (bottom panel) [PITH_FULL_IMAGE:figures/full_fig_p011_13.png] view at source ↗

discussion (0)

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

Reference graph

Works this paper leans on

59 extracted references · 7 canonical work pages

  1. [1]

    , keywords =

    The eROSITA X-ray telescope on SRG. , keywords =. doi:10.1051/0004-6361/202039313 , archivePrefix =. 2010.03477 , primaryClass =

    Pith/arXiv arXiv 2010

  2. [2]

    First X-ray catalogues and data release of the western Galactic hemisphere

    The SRG/eROSITA all-sky survey. First X-ray catalogues and data release of the western Galactic hemisphere. , keywords =. doi:10.1051/0004-6361/202347165 , archivePrefix =. 2401.17274 , primaryClass =

    Pith/arXiv arXiv

  3. [3]

    , keywords =

    Gammapy: A Python package for gamma-ray astronomy. , keywords =. doi:10.1051/0004-6361/202346488 , archivePrefix =. 2308.13584 , primaryClass =

    Pith/arXiv arXiv

  4. [4]

    First Public Test Data Release , publisher =

    H.E.S.S. First Public Test Data Release , publisher =. 2018 , copyright =. doi:10.5281/ZENODO.1421098 , url =

    work page doi:10.5281/zenodo.1421098 2018

  5. [5]

    X-ray catalogue

    The eROSITA Final Equatorial Depth Survey (eFEDS). X-ray catalogue. , keywords =. doi:10.1051/0004-6361/202141266 , archivePrefix =. 2106.14517 , primaryClass =

    Pith/arXiv arXiv

  6. [6]

    , keywords =

    Validation of open-source science tools and background model construction in -ray astronomy. , keywords =. doi:10.1051/0004-6361/201936452 , archivePrefix =. 1910.08088 , primaryClass =

    Pith/arXiv arXiv 1910

  7. [7]

    Astronomical Data Analysis Software and Systems XII , year = 2003, editor =

    New Features of SAOImage DS9. Astronomical Data Analysis Software and Systems XII , year = 2003, editor =

    2003

  8. [8]

    ASP Conf

    XSPEC: The First Ten Years. ASP Conf. Ser. , year = 1996, editor =

    1996

  9. [9]

    , keywords =

    Validation of standardized data formats and tools for ground-level particle-based gamma-ray observatories. , keywords =. doi:10.1051/0004-6361/202243527 , archivePrefix =. 2203.05937 , primaryClass =

    Pith/arXiv arXiv

  10. [10]

    , keywords =

    Constraining leptonic emission scenarios for the PeVatron candidate HESS J1702 - 420 with deep XMM-Newton observations. , keywords =. doi:10.1051/0004-6361/202244696 , archivePrefix =. 2209.09566 , primaryClass =

    Pith/arXiv arXiv

  11. [11]

    Galactic Plane Survey

    The population of TeV pulsar wind nebulae in the H.E.S.S. Galactic Plane Survey. , keywords =. doi:10.1051/0004-6361/201629377 , archivePrefix =. 1702.08280 , primaryClass =

    Pith/arXiv arXiv

  12. [12]

    , keywords =

    Towards open and reproducible multi-instrument analysis in gamma-ray astronomy. , keywords =. doi:10.1051/0004-6361/201834938 , archivePrefix =. 1903.06621 , primaryClass =

    Pith/arXiv arXiv 1903

  13. [13]

    , keywords =

    Unified multiwavelength data analysis workflow with gammapy: Constraining the broadband emission of the flat-spectrum radio quasar OP 313. , keywords =. doi:10.1051/0004-6361/202452349 , archivePrefix =. 2409.20487 , primaryClass =

    Pith/arXiv arXiv

  14. [14]

    , keywords =

    The Evolution and Structure of Pulsar Wind Nebulae. , keywords =. doi:10.1146/annurev.astro.44.051905.092528 , archivePrefix =. astro-ph/0601081 , primaryClass =

    Pith/arXiv arXiv

  15. [15]

    Handbook of X-ray and Gamma-ray Astrophysics , year = 2022, editor =

    Pulsar Wind Nebulae. Handbook of X-ray and Gamma-ray Astrophysics , year = 2022, editor =. doi:10.1007/978-981-16-4544-0_157-1 , adsurl =

    doi:10.1007/978-981-16-4544-0_157-1 2022

  16. [16]

    and Becerra Gonzalez, Josefa and Belmont-Moreno, Ernesto and BenZvi, Segev and Berley, David and Brisbois, Chad and Caballero Mora, Karen S

    Olivera-Nieto, Laura and Joshi, Vikas and Schoorlemmer, Harm and Donath, Axel and Abeysekara, Anushka Udara and Albert, Andrea and Alfaro, Ruben Jose and Alvarez, César and Álvarez Romero, Juan de Dios and Angeles Camacho, José Roberto and Arteaga Velazquez, Juan Carlos and Kollamparambil, Arun Babu and Avila Rojas, Daniel Omar and Ayala Solares, Hugo Alb...

  17. [17]

    PyXspec: Python interface to XSPEC spectral-fitting program

  18. [18]

    Wilms and A

    J. Wilms and A. Allen and R. McCray , title = ". , abstract =. 2000 , month =. doi:10.1086/317016 , url =

    doi:10.1086/317016 2000

  19. [19]

    ASP Conf

    The OGIP FITS Working Group. ASP Conf. Ser. , year = 1995, editor =

    1995

  20. [20]

    , keywords =

    HI4PI: A full-sky H I survey based on EBHIS and GASS. , keywords =. doi:10.1051/0004-6361/201629178 , archivePrefix =. 1610.06175 , primaryClass =

    Pith/arXiv arXiv

  21. [21]

    , keywords =

    Background modelling in very-high-energy -ray astronomy. , keywords =. doi:10.1051/0004-6361:20066674 , archivePrefix =. astro-ph/0610959 , primaryClass =

    Pith/arXiv arXiv

  22. [22]

    , keywords =

    Observations of the Crab nebula with HESS. , keywords =. doi:10.1051/0004-6361:20065351 , archivePrefix =. astro-ph/0607333 , primaryClass =

    Pith/arXiv arXiv

  23. [23]

    Observations of the Crab Nebula with H.E.S.S

    van Eldik, Christopher and Holler, Markus and Berge, David and Zaborov, Dmitry and Lenain, Jean-Philippe and Marandon, Vincent and Murach, Thomas and Prokoph, Heike and de Naurois, Mathieu and Parsons, Robert Daniel. Observations of the Crab Nebula with H.E.S.S. phase II. PoS. doi:10.22323/1.236.0847

    doi:10.22323/1.236.0847

  24. [24]

    doi:10.5281/zenodo.14760974 , version =

    Gammapy: Python toolbox for gamma-ray astronomy. doi:10.5281/zenodo.14760974 , version =

    doi:10.5281/zenodo.14760974

  25. [25]

    Galactic plane survey

    The H.E.S.S. Galactic plane survey. , keywords =. doi:10.1051/0004-6361/201732098 , archivePrefix =. 1804.02432 , primaryClass =

    Pith/arXiv arXiv

  26. [26]

    , keywords =

    Discovery of extended VHE gamma-ray emission from the asymmetric pulsar wind nebula in MSH 15-52 with HESS. , keywords =. doi:10.1051/0004-6361:200500105 , archivePrefix =. astro-ph/0504120 , primaryClass =

    Pith/arXiv arXiv

  27. [27]

    , keywords =

    Detection of the Energetic Pulsar PSR B1509-58 and its Pulsar Wind Nebula in MSH 15-52 Using the Fermi-Large Area Telescope. , keywords =. doi:10.1088/0004-637X/714/1/927 , archivePrefix =. 1003.3833 , primaryClass =

    Pith/arXiv arXiv

  28. [28]

    , keywords =

    Spatially resolved XMM-Newton analysis and a model of the nonthermal emission of MSH 15-52. , keywords =. doi:10.1051/0004-6361/201014151 , archivePrefix =. 1005.0785 , primaryClass =

    Pith/arXiv arXiv

  29. [29]

    , keywords =

    MSH 15-52 : a supernova remnant containing two compact X-ray sources. , keywords =. doi:10.1086/160907 , adsurl =

    doi:10.1086/160907

  30. [30]

    , keywords =

    Nonthermal Radiation from Pulsar Wind Nebulae. , keywords =. doi:10.1086/527466 , adsurl =

    doi:10.1086/527466

  31. [31]

    The SRG/eROSITA view of the soft X-ray background in the eFEDS field

    Abundance and temperature of the outer hot circumgalactic medium. The SRG/eROSITA view of the soft X-ray background in the eFEDS field. , keywords =. doi:10.1051/0004-6361/202243992 , archivePrefix =. 2210.03133 , primaryClass =

    Pith/arXiv arXiv

  32. [32]

    , keywords =

    SRG/eROSITA X-ray shadowing study of giant molecular clouds. , keywords =. doi:10.1051/0004-6361/202345867 , archivePrefix =. 2306.05858 , primaryClass =

    Pith/arXiv arXiv

  33. [33]

    SRG/eROSITA in-flight background at L2

    Freyberg, Michael and Perinati, Emanuele and Pacaud, Florian and Eraerds, Tanja and Churazov, Eugene and Dennerl, Konrad and Predehl, Peter and Merloni, Andrea and Meidinger, Norbert and Bulbul, Esra and Friedrich, Susanne and Gilfanov, Marat and Tenzer, Chris and Pommranz, Christian and Eckert, Dominique and Schmitt, Jürgen and Brusa, Marcella and Santan...

    doi:10.1117/12.2562709

  34. [34]

    Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series , year = 2021, editor =

    The calibration of eROSITA on SRG. Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series , year = 2021, editor =. doi:10.1117/12.2562330 , adsurl =

    doi:10.1117/12.2562330 2021

  35. [35]

    Tempestuous life beyond R _ 500 : X-ray view on the Coma cluster with SRG/eROSITA. II. Shock and relic. , keywords =. doi:10.1051/0004-6361/202244021 , archivePrefix =. 2205.07511 , primaryClass =

    Pith/arXiv arXiv

  36. [36]

    Time and relative dimension in space

    Representations of time coordinates in FITS. Time and relative dimension in space. , keywords =. doi:10.1051/0004-6361/201424653 , archivePrefix =. 1409.7583 , primaryClass =

    Pith/arXiv arXiv

  37. [37]

    , keywords =

    Chandra Observation of the Interaction between the Hot Plasma Nebula RCW 89 and the Pulsar Jet of PSR B1509-58. , keywords =. doi:10.1086/432590 , archivePrefix =. astro-ph/0507049 , primaryClass =

    Pith/arXiv arXiv

  38. [38]

    , title =

    Picquenot, Adrien and Acero, Fabio and Olivares, Valeria and Negro, Michela and Pratt, Gabriel W. , title =. , abstract =. 2025 , month =. doi:10.3847/2041-8213/ae14fc , url =

    work page doi:10.3847/2041-8213/ae14fc 2025

  39. [39]

    , keywords =

    Collisional Plasma Models with APEC/APED: Emission-Line Diagnostics of Hydrogen-like and Helium-like Ions. , keywords =. doi:10.1086/322992 , archivePrefix =. astro-ph/0106478 , primaryClass =

    Pith/arXiv arXiv

  40. [40]

    and Foster, A.R

    Smith, R.K. and Foster, A.R. and Brickhouse, N.S. , title =. Astronomische Nachrichten , volume =. doi:https://doi.org/10.1002/asna.201211673 , url =. https://onlinelibrary.wiley.com/doi/pdf/10.1002/asna.201211673 , abstract =

    work page doi:10.1002/asna.201211673

  41. [41]

    Modelling Non-Maxwellian Plasmas with AtomDB

  42. [42]

    , author=

    THE STRUCTURE OF THE LOCAL HOT BUBBLE. , author=. 2017 , month=. doi:10.3847/1538-4357/834/1/33 , abstractNote=

    doi:10.3847/1538-4357/834/1/33 2017

  43. [43]

    Joint eROSITA and H.E.S.S

    Egg, Katharina and Mitchell, Alison. Joint eROSITA and H.E.S.S. analysis of MSH 15-52 using Gammapy. PoS. doi:10.22323/1.501.0637

    work page doi:10.22323/1.501.0637

  44. [44]

    Naima: a Python package for inference of particle distribution properties from nonthermal spectra

    Zabalza, Victor. Naima: a Python package for inference of particle distribution properties from nonthermal spectra. PoS. doi:10.22323/1.236.0922

    doi:10.22323/1.236.0922

  45. [45]

    GAMERA - a new modeling package for non-thermal spectral modeling

    Hahn, Joachim. GAMERA - a new modeling package for non-thermal spectral modeling. PoS. doi:10.22323/1.236.0917

    doi:10.22323/1.236.0917

  46. [46]

    2024 , eprint=

    Fermi Large Area Telescope Fourth Source Catalog Data Release 4 (4FGL-DR4) , author=. 2024 , eprint=

    2024

  47. [47]

    and Acero, F

    Abdollahi, S. and Acero, F. and Baldini, L. and Ballet, J. and Bastieri, D. and Bellazzini, R. and Berenji, B. and Berretta, A. and Bissaldi, E. and Blandford, R. D. and Bloom, E. and Bonino, R. and Brill, A. and Britto, R. J. and Bruel, P. and Burnett, T. H. and Buson, S. and Cameron, R. A. and Caputo, R. and Caraveo, P. A. and Castro, D. and Chaty, S. a...

  48. [48]

    Atwood, W. B. and Abdo, A. A. and Ackermann, M. and Althouse, W. and Anderson, B. and Axelsson, M. and Baldini, L. and Ballet, J. and Band, D. L. and Barbiellini, G. and Bartelt, J. and Bastieri, D. and Baughman, B. M. and Bechtol, K. and Bédérède, D. and Bellardi, F. and Bellazzini, R. and Berenji, B. and Bignami, G. F. and Bisello, D. and Bissaldi, E. a...

  49. [49]

    F1000Research , VOLUME =

    Mölder, F and Jablonski, KP and Letcher, B and Hall, MB and Tomkins-Tinch, CH and Sochat, V and Forster, J and Lee, S and Twardziok, SO and Kanitz, A and Wilm, A and Holtgrewe, M and Rahmann, S and Nahnsen, S and Köster, J , TITLE =. F1000Research , VOLUME =. 2021 , NUMBER =

    2021

  50. [50]

    doi:10.5281/zenodo.11236879 , url =

    Doug Burke and Omar Laurino and wmclaugh and Hans Moritz Günther and Marie-Terrell and dtnguyen2 and Aneta Siemiginowska and Harlan Cheer and Jamie Budynkiewicz and Tom Aldcroft and luzpaz and Christoph Deil and Brigitta Sipőcz and Johannes Buchner and nplee and Axel Donath and Iva Laginja and Katrin Leinweber and Todd , title =. doi:10.5281/zenodo.112368...

    work page doi:10.5281/zenodo.11236879

  51. [51]

    and Smith, Randall K

    Arnaud, Keith A. and Smith, Randall K. , editor=. Data reduction and calibration , booktitle=. 2011 , pages=

    2011

  52. [52]

    The Calibration Requirements for Spectral Analysis

    George, Ian M. and Arnaud, Keith A. and Pence, Bill and Ruamsuwan, Laddawan and Corcoran, Michael F. , title = "The Calibration Requirements for Spectral Analysis", howpublished = "", month = "12", year = "2020", note = "

    2020

  53. [53]

    and Yusaf, Rehana and Corcoran, Michael F

    George, Ian M. and Yusaf, Rehana and Corcoran, Michael F. , title =. 2011

    2011

  54. [54]

    Statistical Aspects of X-ray Spectral Analysis

    Buchner, Johannes and Boorman, Peter. Statistical Aspects of X-ray Spectral Analysis. Handbook of X-ray and Gamma-ray Astrophysics. 2024. doi:10.1007/978-981-19-6960-7_175

    work page doi:10.1007/978-981-19-6960-7_175 2024

  55. [55]

    AIP Conf

    Open high-level data formats and software for gamma-ray astronomy. AIP Conf. Ser. , year = 2017, volume =. doi:10.1063/1.4969003 , archivePrefix =. 1610.01884 , primaryClass =

    Pith/arXiv arXiv 2017

  56. [56]

    , year =

    Davis, John E. , year =. The. , volume =. doi:10.1086/319002 , url =

    doi:10.1086/319002

  57. [57]

    and Giacconi, R

    Gorenstein, P. and Giacconi, R. and Gursky, H. , date =. The. 1967 , journal =. doi:10.1086/180098 , url =

    work page doi:10.1086/180098 1967

  58. [58]

    EPJ C , keywords =

    Prospects for combined analyses of hadronic emission from -ray sources in the Milky Way with CTA and KM3NeT. EPJ C , keywords =. doi:10.1140/epjc/s10052-023-12279-z , archivePrefix =. 2309.03007 , primaryClass =

    Pith/arXiv arXiv

  59. [59]

    , keywords =

    Teraelectronvolt Astronomy. , keywords =. doi:10.1146/annurev-astro-082708-101816 , archivePrefix =. 1006.5210 , primaryClass =

    Pith/arXiv arXiv