pith. sign in

arxiv: 2606.17563 · v1 · pith:24A43T6Wnew · submitted 2026-06-16 · 🌌 astro-ph.HE

A MUSE View of the Optical Torus within the Supernova Remnant 1E 0102.2-7219

Pith reviewed 2026-06-26 23:53 UTC · model grok-4.3

classification 🌌 astro-ph.HE
keywords supernova remnantoptical torusMUSE spectroscopymultiphase mediumshock modelscentral compact objectSmall Magellanic Cloudejecta inhomogeneities
0
0 comments X

The pith

MUSE observations show that no single physical model accounts for both neutral and high-ionization lines in the optical torus of supernova remnant 1E 0102.2-7219.

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

New adaptive-optics MUSE data resolve the torus into a cavity with a sharp inner edge and outer filaments. The emission displays continuous velocities, broad line widths, and overlapping contributions from neutral species such as O I and high-ionization lines such as [O III]. Spatially resolved line-ratio maps indicate a non-equilibrium multiphase medium. Grids of photoionization and shock models find no single set of parameters that matches both the strong neutral and high-ionization diagnostics at once. The authors therefore conclude that multiple physical conditions must coexist and favor shocks traveling through density inhomogeneities in the ejecta.

Core claim

The improved spatial resolution resolves the previously identified torus into a cavity-like structure with a sharply defined inner edge and diffuse, outer filamentary substructure. The emission shows continuous velocity connectivity, broad intrinsic line widths, and co-spatial contributions from neutral and partially ionized species, including O I, Ne I, [O I], [O II], and [O III]. Spatially resolved line-ratio maps indicate that the emission arises from a multiphase, non-equilibrium medium rather than a single homogeneous component. Comparison with photoionization and shock models shows that no single-component model within the explored parameter space can simultaneously reproduce both the

What carries the argument

Spatially resolved line-ratio maps tested against grids of photoionization and shock models, which together demonstrate that multiple physical conditions must coexist.

If this is right

  • Shocks propagating through density inhomogeneities in the supernova ejecta shape the observed cavity morphology and line excitation.
  • Neutral and high-ionization emission arise from co-spatial but physically distinct regions within the same structure.
  • The medium is out of equilibrium, with broad line widths reflecting the effects of shocks or turbulence.
  • Alternative ionization mechanisms tied to the central compact object candidate or binary evolution remain possible but are not required by the current data.

Where Pith is reading between the lines

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

  • Other oxygen-rich supernova remnants with central compact objects may display similarly complex line ratios that also require multi-component modeling.
  • Higher-resolution integral-field spectroscopy could directly map the density jumps at the cavity edge and test the shock-inhomogeneity picture.
  • Multi-wavelength monitoring could check whether the central source contributes variable ionization that is currently hidden within the shock-dominated emission.

Load-bearing premise

The ranges covered by the photoionization and shock model grids are broad enough that the absence of a single-component fit is not simply due to missing parts of parameter space.

What would settle it

Discovery of one set of density, ionization parameter, and shock velocity values that simultaneously matches the observed strengths of both neutral lines such as O I and high-ionization lines such as [O III] inside the existing model grids or a modest extension of them.

Figures

Figures reproduced from arXiv: 2606.17563 by Ashley J. Ruiter, Chuan-Jui Li, Fr\'ed\'eric P. A. Vogt, Gilles Ferrand, Ivo R. Seitenzahl, Janette Suherli, Parviz Ghavamian, Ralph Sutherland, Samar Safi-Harb.

Figure 1
Figure 1. Figure 1: (a) Multiwavelength composite image of E0102 combining X-ray emission from Chandra (blue and purple) with optical emission from VLT/MUSE (bright red) and HST (dark red and green). The footprint of the MUSE NFM mosaic is shown in white and the CCO candidate identified by F. P. A. Vogt et al. (2018) is indicated by the white arrow. The center of expansion (CoE) determined from the proper-motion expansion of … view at source ↗
Figure 2
Figure 2. Figure 2: Integrated intensity maps of (a) O I λ7774, (b) [O I]λ6300, (c) [O II]λ7320,7330, and (d) [O III]λ5007, summed over −50 ≤ vlos ≤ +450 km s−1 . The dark-red crosshair in each panel marks the position of the X-ray source, and the white circle indicates a masked bright foreground star. The MUSE NFM-AO PSF is shown in each panel, with measured values of 0.096′′ , 0.110′′, 0.096′′, and 0.164′′ for O I, [O I], [… view at source ↗
Figure 3
Figure 3. Figure 3: Channel maps of a three-color composite: O I λ7774 (yellow), [O I]λ6300 (blue), and [O III]λ5007 (magenta), con￾structed in 50 km s−1 bins from −50 to 450 km s−1 . The orientation on the sky is indicated in the upper-left panel. The dark-red crosshair marks the position of the X-ray source, and the black circle indicates a masked foreground star [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Channel maps of Ne I λ6402 emission in 50 km s−1 bins from −50 to +450 km s−1 . A Gaussian smoothing has been applied to enhance low surface brightness structures. The orientation on the sky is indicated in the upper-left panel. The dark-red crosshair marks the position of the X-ray source, and the white circle indicates a masked foreground star [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: (a) Line-of-sight velocity map of O I λλ7774,8446, derived from spaxel-by-spaxel Gaussian fitting, with O I λ7774 flux contours overlaid to trace the emission morphology. (b) Intrinsic FWHM map of O I λ7774, corrected for instrumental broadening, with O I λ7774 flux contours overlaid. Both maps are constructed from data spatially binned by a factor of 4 and include only spaxels with S/N ≥ 5. The contours c… view at source ↗
Figure 6
Figure 6. Figure 6: (a) Integrated O I λ7774 intensity map with radial cuts spanning a range of position angles. All sampled position angles are shown in grey, with selected directions highlighted to illustrate representative profiles. The dashed line indicates the orientation of the position–velocity (PV) slice. The X-ray source position is marked by the red crosshair. (b) Radial intensity profiles as a function of projected… view at source ↗
Figure 7
Figure 7. Figure 7: 10%–90% intensity rise width of the O I λ7774 radial profiles as a function of position angle. The dashed line indicates the corresponding width of the instrumental PSF. The color scale shows the peak intensity of each profile. The measured widths are generally close to the PSF, indicating a sharp, spatially confined edge. structure and passes through the X-ray source (dashed line in [PITH_FULL_IMAGE:figu… view at source ↗
Figure 8
Figure 8. Figure 8: Radial intensity profiles for (a) O I, (b) [O I], (c) [O II], and (d) [O III] emission, showing the median profile (solid line) and the 16th–84th percentile range (shaded region) over position angles between 10◦ and 50◦ . The shaded vertical band marks the characteristic radius of the edge. lower-density regions extending outward. The emission is therefore best understood as a structured, multiphase region… view at source ↗
Figure 9
Figure 9. Figure 9: PV diagrams of (a) O I λ7774 and (b) [O III]λ5007 (bottom), extracted along a slice that crosses the structure and passes through the X-ray source (shown in [PITH_FULL_IMAGE:figures/full_fig_p010_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Spatially resolved line-ratio maps of the optical torus in E0102. (a) O I λ8446/λ7774, tracing the variations within permitted O I emission. (b) O I λ7774/[O I]λ6300, tracing the recombination-favored and collisionally excited neutral oxygen. (c) O I λ7774/[O III]λ5007, tracing the tran￾sition between low- and high-ionization gas. All maps are derived from spaxel-by-spaxel Gaussian fits to data binned by … view at source ↗
Figure 11
Figure 11. Figure 11: Colors distinguish the model type: blackbody photoionization (gold), bremsstrahlung photoionization (pur￾ple), and shock models (teal). Marker shapes indicate the assumed abundances of the emitting gas: E0102 O-rich ejecta adopted from W. P. Blair et al. (2000) (circles) and SMC composition (diamonds). The observed values for P1 and P2 are shown as stars. (a) High-ionization versus neutral oxygen diagnost… view at source ↗
Figure 12
Figure 12. Figure 12: Channel maps of O I λ7774 (top row), [O I]λ6300 (second row), [O II]λ7320,7330 (third row), and [O III]λ5007 (bottom row), constructed in 50 km s−1 bins from −50 to +450 km s−1 (left to right). The dark-red crosshair marks the position of the X-ray source, and the white circle indicates a masked foreground star. The orientation on the sky is shown in the upper-left panel [PITH_FULL_IMAGE:figures/full_fig… view at source ↗
Figure 13
Figure 13. Figure 13: Collapsed PV profile of O I λ7774 along the slice. The gold curve shows the mean flux profile and the blue curve shows a Gaussian-smoothed version used to identify the dominant peaks. The yellow dashed vertical lines mark the positions of the two peaks and the red dashed line marks the position of the X-ray source along the slice. Freudling, W., Romaniello, M., Bramich, D. M., et al. 2013, A&A, 559, A96, … view at source ↗
read the original abstract

We present new MUSE Narrow Field Mode with Adaptive Optics observations of the optical torus surrounding a Central Compact Object (CCO) candidate within the oxygen-rich supernova remnant 1E 0102.2-7219 (E0102) located in the Small Magellanic Cloud. These data provide nearly an order-of-magnitude improvement in spatial resolution over previous MUSE Wide Field Mode observations. The improved spatial resolution resolved the previously identified torus into a cavity-like structure with a sharply defined inner edge and diffuse, outer filamentary substructure. The emission shows continuous velocity connectivity, broad intrinsic line widths, and co-spatial contributions from neutral and partially ionized species, including O I, Ne I, [O I], [O II], and [O III]. Spatially resolved line-ratio maps indicate that the emission arises from a multiphase, non-equilibrium medium rather than a single homogeneous component. Comparison with photoionization and shock models shows that no single-component model within the explored parameter space can simultaneously reproduce both the strong neutral and high-ionization diagnostics, indicating that multiple physical conditions must coexist. We favor an interpretation in which shocks propagating through density inhomogeneities in the ejecta shape the observed morphology and excitation, while also considering alternative mechanisms linked to the central source, binary evolution, or interaction with an embedded object within the remnant.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper reports MUSE Narrow Field Mode adaptive-optics observations of the optical torus around the CCO candidate in the oxygen-rich SNR 1E 0102.2-7219. These data resolve the torus into a cavity-like structure with a sharply defined inner edge and diffuse filamentary outer substructure, showing continuous velocity fields, broad line widths, and co-spatial emission from neutral (O I, Ne I, [O I]) and higher-ionization ([O II], [O III]) species. Spatially resolved line-ratio maps are compared to grids of photoionization and shock models; the authors conclude that no single-component model within the explored parameter space simultaneously reproduces the strong neutral and high-ionization diagnostics, implying a multiphase, non-equilibrium medium best explained by shocks propagating through density inhomogeneities in the ejecta.

Significance. If the model-comparison result holds, the work supplies spatially resolved evidence that optical tori in young oxygen-rich SNRs require multiple physical conditions rather than a single homogeneous zone, with direct implications for ejecta structure, CCO environments, and non-equilibrium ionization modeling. The nearly order-of-magnitude gain in spatial resolution over prior MUSE WFM data is a clear observational advance, and the explicit statement that the conclusion is limited to the explored grids is a positive transparency measure.

major comments (1)
  1. [model-comparison section] Abstract and model-comparison section: The central claim that 'no single-component model within the explored parameter space' reproduces both the neutral (O I/[O I]) and high-ionization ([O III]) diagnostics is load-bearing for the multiphase conclusion. The manuscript must explicitly tabulate or describe the full ranges sampled for density, ionization parameter, shock velocity, pre-shock density, and abundances (including O-rich ejecta patterns) so that readers can verify whether viable single-zone solutions lie outside the grids rather than being excluded by the data.
minor comments (2)
  1. [abstract] The abstract and text use both '1E 0102.2-7219' and 'E0102'; a single consistent designation should be adopted after first use.
  2. [results section] Line-ratio maps are described as indicating multiphase gas, but the precise diagnostic ratios (e.g., [O I]/Hα vs. [O III]/Hβ) and their error maps are not referenced to a specific figure or table; adding these cross-references would improve traceability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive review and for highlighting the need for greater transparency in the model-comparison analysis. We address the single major comment below.

read point-by-point responses
  1. Referee: [model-comparison section] Abstract and model-comparison section: The central claim that 'no single-component model within the explored parameter space' reproduces both the neutral (O I/[O I]) and high-ionization ([O III]) diagnostics is load-bearing for the multiphase conclusion. The manuscript must explicitly tabulate or describe the full ranges sampled for density, ionization parameter, shock velocity, pre-shock density, and abundances (including O-rich ejecta patterns) so that readers can verify whether viable single-zone solutions lie outside the grids rather than being excluded by the data.

    Authors: We agree that the claim requires explicit documentation of the explored grids to allow independent verification. In the revised manuscript we will add a table (or expanded methods subsection) that lists the full sampled ranges for density, ionization parameter, shock velocity, pre-shock density, and abundances, including the specific O-rich ejecta abundance patterns adopted from the literature. This addition will make the parameter-space coverage transparent without altering the scientific conclusion. revision: yes

Circularity Check

0 steps flagged

No circularity: observational data compared to external model grids

full rationale

The paper reports MUSE observations of emission lines in the SNR torus and compares observed line ratios (O I, [O I], [O III], etc.) to grids of photoionization and shock models. The central claim—that no single-component model reproduces both neutral and high-ionization diagnostics—is reached by direct inspection of model outputs against data within the stated parameter ranges. No equation or result is defined in terms of itself, no fitted parameter is relabeled as a prediction, and no load-bearing premise rests on a self-citation chain. The model grids are external (standard codes with varied density, ionization parameter, velocity, abundance), and the conclusion follows from their failure to match, without reduction to the paper's own inputs by construction. This is a standard observational-model comparison workflow; the analysis is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The abstract invokes standard astrophysical assumptions about line identification, the applicability of existing photoionization and shock codes, and the completeness of the explored model grids; no new entities are postulated.

axioms (2)
  • domain assumption Line identifications for O I, Ne I, [O I], [O II], and [O III] are correct and uncontaminated.
    Required for the co-spatial neutral and ionized diagnostics reported in the abstract.
  • domain assumption The parameter space searched in the photoionization and shock models is representative of the physical conditions present.
    Invoked when concluding that no single-component model works.

pith-pipeline@v0.9.1-grok · 5822 in / 1415 out tokens · 27490 ms · 2026-06-26T23:53:54.076823+00:00 · methodology

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

43 extracted references · 40 canonical work pages

  1. [1]

    Alsaberi, R. Z. E., Filipovi´ c, M. D., Dai, S., et al. 2024, MNRAS, 527, 1444, doi: 10.1093/mnras/stad3300 Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, ˚ a, 558, A33, doi: 10.1051/0004-6361/201322068 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, 156, 123, doi: 10.3847/1538-3881/aabc4f Astropy Collabo...

  2. [2]

    Ground-based and Airborne Instrumentation for Astronomy III , year = 2010, editor =

    Bacon, R., Accardo, M., Adjali, L., et al. 2010, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 7735, Ground-based and Airborne Instrumentation for Astronomy III, ed. I. S. McLean, S. K. Ramsay, & H. Takami, 773508, doi: 10.1117/12.856027

  3. [3]

    2021, ApJ, 912, 33, doi: 10.3847/1538-4357/abe2a7

    Banovetz, J., Milisavljevic, D., Sravan, N., et al. 2021, ApJ, 912, 33, doi: 10.3847/1538-4357/abe2a7

  4. [4]

    K., & Kastner, S

    Bhatia, A. K., & Kastner, S. O. 1995, ApJS, 96, 325, doi: 10.1086/192121

  5. [5]

    P., Morse, J

    Blair, W. P., Morse, J. A., Raymond, J. C., et al. 2000, ApJ, 537, 667, doi: 10.1086/309077

  6. [6]

    Bowen, I. S. 1947, PASP, 59, 196, doi: 10.1086/125951

  7. [7]

    A., & Kirshner, R

    Chevalier, R. A., & Kirshner, R. P. 1979, ApJ, 233, 154, doi: 10.1086/157377

  8. [8]

    Dopita, M. A. 1987, Australian Journal of Physics, 40, 789, doi: 10.1071/PH870789

  9. [9]

    A., Tuohy, I

    Dopita, M. A., Tuohy, I. R., & Mathewson, D. S. 1981, ApJL, 248, L105, doi: 10.1086/183635

  10. [10]

    English, J., Richardson, M. L. A., Ferrand, G., & Deg, N. 2024, CosmosCanvas: Useful color maps for different astrophysical properties,, Astrophysics Source Code Library, record ascl:2401.005 http://ascl.net/2401.005

  11. [11]

    L., Morse, J

    Finkelstein, S. L., Morse, J. A., Green, J. C., et al. 2006, ApJ, 641, 919, doi: 10.1086/500570

  12. [12]

    Fitzpatrick, E. L. 1999, PASP, 111, 63, doi: 10.1086/316293 16 Figure 12.Channel maps of OIλ7774 (top row), [OI]λ6300 (second row), [OII]λ7320,7330 (third row), and [OIII]λ5007 (bottom row), constructed in 50 km s −1 bins from −50 to +450 km s−1 (left to right). The dark-red crosshair marks the position of the X-ray source, and the white circle indicates ...

  13. [13]

    M., Ballester P., Forchi V., Garc \' a-Dabl \'o C

    Freudling, W., Romaniello, M., Bramich, D. M., et al. 2013, A&A, 559, A96, doi: 10.1051/0004-6361/201322494

  14. [14]

    V., Halpern, J

    Gotthelf, E. V., Halpern, J. P., & Alford, J. 2013, ApJ, 765, 58, doi: 10.1088/0004-637X/765/1/58

  15. [15]

    B., et al

    Graczyk, D., Pietrzy´ nski, G., Thompson, I. B., et al. 2020, ApJ, 904, 13, doi: 10.3847/1538-4357/abbb2b

  16. [16]

    R., Heinke, C

    Hebbar, P. R., Heinke, C. O., & Ho, W. C. G. 2020, MNRAS, 491, 1585, doi: 10.1093/mnras/stz2570

  17. [17]

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

  18. [18]

    1981, PASJ, 33, 1, doi: 10.1093/pasj/33.1.1

    Itoh, H. 1981, PASJ, 33, 1, doi: 10.1093/pasj/33.1.1

  19. [19]

    1986, PASJ, 38, 717, doi: 10.1093/pasj/38.5.717

    Itoh, H. 1986, PASJ, 38, 717, doi: 10.1093/pasj/38.5.717

  20. [20]

    O., & Bhatia, A

    Kastner, S. O., & Bhatia, A. K. 1995, ApJ, 439, 346, doi: 10.1086/175178

  21. [21]

    R., Ishioka, R., et al

    Li, C.-J., Seitenzahl, I. R., Ishioka, R., et al. 2021, ApJ, 915, 20, doi: 10.3847/1538-4357/abf7c5

  22. [22]

    J., & Plucinsky, P

    Long, X., Gaetz, T. J., & Plucinsky, P. P. 2020, ApJ, 904, 70, doi: 10.3847/1538-4357/abc075

  23. [23]

    E., Fulbright, J

    Osterbrock, D. E., Fulbright, J. P., Martel, A. R., et al. 1996, PASP, 108, 277, doi: 10.1086/133722

  24. [24]

    S., Kara, E

    Reynolds, C. S., Kara, E. A., Mushotzky, R. F., et al. 2023, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 12678, UV, X-Ray, and Gamma-Ray Space Instrumentation for Astronomy XXIII, ed. O. H. Siegmund & K. Hoadley, 126781E, doi: 10.1117/12.2677468

  25. [25]

    C., & Dopita, M

    Russell, S. C., & Dopita, M. A. 1992, ApJ, 384, 508, doi: 10.1086/170893

  26. [26]

    Statsmodels:

    Seabold, S., & Perktold, J. 2010, SciPy 2010, doi: 10.25080/Majora-92bf1922-011

  27. [27]

    R., Vogt, F

    Seitenzahl, I. R., Vogt, F. P. A., Terry, J. P., et al. 2018, ApJL, 853, L32, doi: 10.3847/2041-8213/aaa958

  28. [28]

    D., & Mitchell, M

    Seward, F. D., & Mitchell, M. 1981, ApJ, 243, 736, doi: 10.1086/158641

  29. [29]

    2026, velociwrap: velocity-consistent (pseudo-RGB) channel maps for astronomical datacubes, v1.0.0 Zenodo, doi: 10.5281/zenodo.20604383

    Suherli, J. 2026, velociwrap: velocity-consistent (pseudo-RGB) channel maps for astronomical datacubes, v1.0.0 Zenodo, doi: 10.5281/zenodo.20604383

  30. [30]

    R., Safi-Harb, S., et al

    Suherli, J., Seitenzahl, I. R., Safi-Harb, S., et al. 2026, ApJL, 997, L36, doi: 10.3847/2041-8213/ae27c4

  31. [31]

    2018, MAPPINGS V: Astrophysical plasma modeling code,, Astrophysics Source Code Library, record ascl:1807.005 http://ascl.net/1807.005

    Sutherland, R., Dopita, M., Binette, L., & Groves, B. 2018, MAPPINGS V: Astrophysical plasma modeling code,, Astrophysics Source Code Library, record ascl:1807.005 http://ascl.net/1807.005

  32. [32]

    S., & Dopita, M

    Sutherland, R. S., & Dopita, M. A. 1995, ApJ, 439, 381, doi: 10.1086/175181

  33. [33]

    S., & Dopita, M

    Sutherland, R. S., & Dopita, M. A. 2017, ApJS, 229, 34, doi: 10.3847/1538-4365/aa6541

  34. [34]

    2012, A&A Rv, 20, 49, doi: 10.1007/s00159-011-0049-1

    Vink, J. 2012, A&A Rv, 20, 49, doi: 10.1007/s00159-011-0049-1

  35. [35]

    E., et al

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

  36. [36]

    Vogt, F., & Dopita, M. A. 2010, ApJ, 721, 597, doi: 10.1088/0004-637X/721/1/597

  37. [37]

    Vogt, F. P. A. 2019, brutifus: Python module to post-process datacubes from integral field spectrographs,, Astrophysics Source Code Library, record ascl:1903.004 http://ascl.net/1903.004

  38. [38]

    Vogt, F. P. A., Bartlett, E. S., Seitenzahl, I. R., et al. 2018, Nature Astronomy, 2, 465, doi: 10.1038/s41550-018-0433-0

  39. [39]

    2017a, A&A, 602, L4, doi: 10.1051/0004-6361/201730756 19

    Ghavamian, P. 2017a, A&A, 602, L4, doi: 10.1051/0004-6361/201730756 19

  40. [40]

    Vogt, F. P. A., Seitenzahl, I. R., Dopita, M. A., & Ruiter, A. J. 2017b, PASP, 129, 058012, doi: 10.1088/1538-3873/129/975/058012

  41. [41]

    M., Palsa , R., Streicher , O., et al

    Weilbacher, P. M., Palsa, R., Streicher, O., et al. 2020, A&A, 641, A28, doi: 10.1051/0004-6361/202037855

  42. [42]

    F., & Kirshner, R

    Winkler, P. F., & Kirshner, R. P. 1985, ApJ, 299, 981, doi: 10.1086/163764

  43. [43]

    Patnaude, D. J. 2019, ApJ, 874, 14, doi: 10.3847/1538-4357/ab09ea