Pith. sign in

REVIEW 2 major objections 7 minor 1 cited by

Predicting ionised gas emission in 3D with SKIRT. I. Framework and validation

T0 review · 2 major / 7 minor · reviewed 2026-07-14 · grok-4.5

Pith's one-line read A new SKIRT module predicts 3D ionised-gas emission lines from hydrodynamical simulations in one Monte Carlo run with dust.

desk verdict Solid engineering paper: a usable 3D photoionisation module inside SKIRT, validated honestly against Cloudy and COLT, with residuals diagnosed rather than hidden. read the letter →

arxiv 2607.09961 v1 pith:FH6E6SO6 submitted 2026-07-10 astro-ph.GA

classification astro-ph.GA
keywords radiativetransferphotoionisationemissionlinesHIIregionsgalaxies:ISMMonteCarlomethodssyntheticobservations
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

Emission lines from ionised gas diagnose star formation, metallicity and ionisation, but realistic galaxy geometries need three-dimensional photoionisation that also handles dust and instrument output. This paper introduces DiffuseIonizedGasMix for the SKIRT Monte Carlo radiative-transfer code: it compresses the local ionising field into an ionisation parameter and four spectral-shape ratios, looks up temperature and opacity in pre-computed Cloudy tables, then solves ion fractions and line emissivities inline once the radiation field has converged. Validated on 60 spherical shells against Cloudy and on a Milky Way-analogue against COLT, hydrogen recombination lines match to a few per cent (Hα median ratio 0.97) and metal lines to roughly five per cent except [S II], while three-dimensional maps show pixel correlations r ≥ 0.92. The result is self-consistent synthetic observations in which lines, dust attenuation and dust re-emission are produced together, ready for any hydrodynamical snapshot.

What carries the argument

DiffuseIonizedGasMix: a hybrid module that characterises the local 1–6 Ryd radiation field by log U and four spectral-shape ratios, maps them via dual pre-computed Cloudy tables to temperature and opacity inside SKIRT’s iteration cycle, then uses an inline multi-element ionisation solver at the converged temperature to compute recombination and collisional line emissivities.

What would settle it

Re-run the same 60 spherical-shell grid and the Milky Way-analogue with a full on-the-fly multi-element thermal solver or with finer spectral binning; if the median [S II] ratio moves substantially closer to unity and the three-dimensional [O III]/[S II] excesses disappear while hydrogen lines remain unchanged, the five-bin table approximation is insufficient.

Watch

Extended reading notes

Core claim

DiffuseIonizedGasMix enables self-consistent three-dimensional synthetic observations of ionised-gas emission lines, dust attenuation and dust re-emission in a single Monte Carlo radiative-transfer run. On a 60-model one-dimensional grid the hydrogen lines agree with Cloudy to within a few per cent (Hα median ratio 0.97) and most forbidden lines to ~5 %; on a Milky Way-analogue galaxy integrated luminosities and pixel maps agree with COLT at r ≥ 0.92, making the module applicable to arbitrary hydrodynamical simulations.

Load-bearing premise

That five energy bins plus solar-scaled Cloudy tables recover gas temperature accurately enough near ionisation fronts for the metal-line calculation to stay reliable.

Editorial extensions

If this is right

  • Any hydrodynamical snapshot can be post-processed into emission-line maps that already include dust attenuation and re-emission without separate photoionisation steps.
  • Mock integral-field observations (MUSE, JWST/NIRSpec) can be generated with consistent geometry, inclination and multi-phase ISM structure.
  • BPT diagrams and other line-ratio diagnostics can be forward-modelled across cosmic time by varying ionising spectra and abundance patterns inside the same radiative-transfer framework.
  • Compact H II regions can still be treated by sub-grid libraries while the resolved diffuse ionised gas is handled self-consistently on the grid.

Reading between the lines

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

  • Once multi-metallicity tables exist, the same module can test whether non-solar abundance patterns or harder spectra drive the high-redshift BPT offset without changing the radiative-transfer core.
  • The residual [S II] and [O III] excesses already flag that ionisation-front resolution and diffuse Lyman-continuum treatment remain the dominant modelling uncertainties for low-ionisation lines.
  • Pairing the module with adaptive mesh refinement guided by local ionisation state would tighten the comparison to pure on-the-fly photoionisation codes.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 7 minor

Summary. The paper introduces DiffuseIonizedGasMix, a new SKIRT material-mix module for 3D photoionisation of ionised gas. It characterises the local ionising field (1–6 Ryd) by log U and four spectral-shape ratios, maps these via dual pre-computed Cloudy STAB tables to temperature and opacity, and evaluates line emissivities with an inline multi-element ionisation solver at the converged state. The module is validated on 60 one-dimensional spherical shells against Cloudy (and COLT) and on a three-dimensional Milky Way-analogue Arepo/SMUGGLE snapshot against COLT. In 1D, Hα and Hβ agree with Cloudy to a few per cent (median ratios 0.97 and 0.94), [O III] and [N II] to ~2–4%, while [S II] λ6717 is high by a median factor 1.23, attributed to a temperature overestimate near the ionisation front. In 3D, pixel correlations reach r ≥ 0.92, hydrogen lines and [N II] agree well in integrated luminosity, and [O III]/[S II] are systematically elevated by ~70–80%, with a cell-level decomposition in Appendix D. The authors conclude that the module enables self-consistent synthetic observations of ionised-gas lines, dust attenuation, and dust re-emission in a single MCRT run.

Significance. If the reported accuracy holds under the stated assumptions, this is a substantial and timely methods contribution. SKIRT is already a standard tool for dust radiative transfer on hydrodynamical snapshots; adding a publicly released, grid-native photoionisation module that shares the same iteration cycle, spatial grids, and instrument pipeline removes a major post-processing gap for IFU-style mock observations (MaNGA, MUSE, JWST/NIRSpec). Strengths include: (i) external validation against two independent codes (Cloudy and COLT) in both 1D and 3D; (ii) quantitative residual diagnostics (Figs. 7–9, 12–15; Tables 5, 7; Appendices B and D) rather than only headline agreement; (iii) a modular hybrid design that imports Cloudy microphysics for T/κ while retaining wavelength-resolved ion fractions for emission; and (iv) open release of the module and tables in the SKIRT codebase. The solar-only, dust-free validation scope is clearly flagged for future papers, so the engineering claim is appropriately bounded.

major comments (2)
  1. [Abstract; §3; §4 Limitations] Abstract and §4 claim that the module enables self-consistent synthetic observations in which ionised-gas lines, dust attenuation, and dust re-emission are computed in a single MCRT run. All quantitative validation (§3.1–3.2) is deliberately dust-free. While §4 correctly notes that dust media can be co-deployed and that dust-modified (log U, Ri) can query the existing tables, the manuscript should more sharply separate (a) physics that has been validated (photoionisation T, κ, and lines without dust) from (b) architectural capability that is enabled but not yet benchmarked with dust present. A short explicit statement in the abstract and at the start of §3 would prevent over-reading of the current accuracy numbers.
  2. [§3.2.2; Table 7; Appendix D; §4] The 3D integrated excesses for [O III] λ5007 (ΣS/ΣC = 1.68) and [S II] λ6717 (1.78) are large enough to matter for BPT and DIG science applications, even though Appendix D provides a clear cell-level decomposition (temperature offset for [O III]; diffuse-LyC vs OTS Case B for [S II] midplane cells). The main text (§3.2.2 and §4) should elevate a concise user-facing accuracy statement: which lines and which diagnostics (e.g. Balmer decrement, [N II]/Hα, [O III]/Hβ) are reliable at the ~0.1–0.2 dex level under the current tables and reemission treatment, and which require the forthcoming multi-metallicity / refined-front work. Without that, readers may either over-trust or dismiss the module based on the raw 70–80% factors alone.
minor comments (7)
  1. [Fig. 8; Table 5] Fig. 8 caption and text: the logarithmic 1−r/rS axis is effective, but the coloured bars (10th–90th percentile of Cloudy cumulative emissivity) would be easier to read if the corresponding peak r/rS values were also listed in the caption or in Table 5.
  2. [§2.3; Table 2] Table 2 and §2.3: the transition-table Ri ranges extend to very large positive values (e.g. log R5 up to +31.1). A brief note on whether these extremes are actually sampled in the 1D/3D runs, or only pad the interpolation domain, would help readers judge table coverage.
  3. [§2.7] §2.7: the statement that collisional excitation of hydrogen lines is neglected (to be added later) is appropriate, but a one-sentence bound on the expected Lyα error at the temperatures reached in the MW run (T ≲ 10^4 K for most emitting cells) would be useful for users planning UV applications.
  4. [§2.4; Appendix A] Appendix A: the per-cell converged fraction plateaus near ~64% while global NH+ converges; this is well explained, but a short remark in §2.4 pointing to Appendix A would help readers who only skim the methods.
  5. [Fig. 15; §3.2.2] Fig. 15 BPT comparison: the broader COLT composite/LINER wing is attributed to Courant-limited cooling preserving hot low-density gas. Consider adding a one-line note on whether a temperature floor or cut on shock-heated cells was applied in either code for the BPT pixels, to aid reproducibility.
  6. [Abstract; §2.3] Minor typography: abstract and body mix Halpha/Hα and [S II] 6717 vs λ6717; standardise to journal style. Also, ‘T wo’ appears as a split word in §2.3 (‘T wo sets of tables’).
  7. [§3.2.1; References] References: the COLT 3D comparison relies on McClymont et al. 2025 (arXiv:2510.13952); ensure the citation is updated if a journal version appears before final acceptance, and that the snapshot configuration (density ceiling, abundances) is fully specified for external reproduction.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: hybrid Cloudy-table + inline-solver module is validated against independent external codes, not forced by construction.

full rationale

DiffuseIonizedGasMix is an engineering methods paper. Temperature and opacity come from pre-computed Cloudy STAB tables indexed by log U and four spectral-shape ratios; ion fractions and line emissivities are then computed by an independent inline solver on the converged wavelength-resolved field (Sect. 2.3, 2.7). Validation is against full Cloudy on 60 1D shells and against COLT on a 3D MW-analogue (Sect. 3.1–3.2), with residuals reported rather than absorbed (Hα median 0.97; [S II] 1.23 from front temperature overestimate; 3D [O III]/[S II] excesses decomposed in Appendix D). Agreement is not guaranteed by construction: a bad 5-bin characterisation or table handoff would produce large line errors, and the paper measures those errors. Self-citations are to prior SKIRT infrastructure (grids, dust, TODDLERS), not to uniqueness theorems or fitted targets that force the present photoionisation results. No fitted-input-as-prediction, no self-definitional loop, no ansatz smuggled as external fact. The derivation chain is self-contained against external benchmarks.

Assumptions & free parameters 5 free parameters · 4 assumptions · 3 invented entities

The central claim rests on standard photoionisation microphysics imported from Cloudy/CHIANTI plus a small set of engineering choices (5-bin compression, dual tables, convergence thresholds, solar abundances). No new physical entities are postulated; the free parameters are numerical controls rather than fitted physical constants.

free parameters (5)
  • number of ionising spectral bins = 5
    Fixed at 5 by design (Table 1); Appendix B shows diminishing returns beyond 5, but the choice is still a free modelling decision that sets table dimensionality.
  • per-cell convergence threshold ϵ_cell = 0.01
    Default 0.01 on relative ΔT and Δlog U (Table 3); controls when a cell is declared converged.
  • global NH+ convergence threshold ϵ_global = 0.001
    Default 10^{-3} (Table 3); gates overall iteration stop.
  • density ceiling for emission = 1000 cm^{-3}
    nH capped at 10^3 cm^{-3} to match table boundary (§2.6); affects dense cells.
  • table blending width in log U = 0.3 dex
    0.3 dex linear weight between standard and transition tables (§2.3).
assumptions (4)
  • domain assumption Local photoionisation and thermal equilibrium can be tabulated from Cloudy as a function of log U and four spectral-shape ratios at fixed solar abundances.
    Core of the hybrid design (§2.1–2.3); assumes the 5-bin compression plus solar GASS10 mix is sufficient for T and κ.
  • domain assumption Case B recombination coefficients (Hui & Gnedin 1997; Storey & Hummer 1995) and CHIANTI collisional rates adequately describe the optical lines of interest.
    Used directly in the inline emissivity solver (§2.7).
  • domain assumption Diffuse re-emission of ionising photons can be treated as isotropic scattering with Wood et al. (2004) channel probabilities and OTS for He Lyα.
    §2.5; affects the diffuse LyC field that later explains part of the [S II] residual.
  • ad hoc to paper Monte Carlo noise and residual unconverged low-U cells do not bias integrated line luminosities once the global NH+ criterion is met.
    Justified by the plateau and global criteria (§2.4, Appendix A) but is an operational assumption of the iteration scheme.
invented entities (3)
  • DiffuseIonizedGasMix material-mix module
    purpose: Encapsulates per-cell state (xH0, xHe0, log U, Ri, T, nH+) and the hybrid update/emission cycle inside SKIRT.
    New software object; no independent physical existence outside the code.
  • Five-bin radiation-field characterisation (log U + R2–R5)
    purpose: Compresses the 1–6 Ryd continuum into table indices while preserving major ionisation edges.
    Engineering construct chosen for table size vs accuracy trade-off (Appendix B).
  • Dual STAB table system (standard + transition)
    purpose: Resolves the rapid hardening and temperature jump near ionisation fronts where a single sparse table fails.
    Ad-hoc split with log-space blending (§2.3); validated only by the 1D residual analysis.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Predicting ionised gas emission in 3D with SKIRT. I. Framework and validation." pith.science (2026). https://pith.science/paper/FH6E6SO6

@misc{pith2026260709961,
  author       = {Pith},
  title        = {Pith review of: Predicting ionised gas emission in 3D with SKIRT. I. Framework and validation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FH6E6SO6}},
  note         = {Machine review of arXiv:2607.09961}
}
read the original abstract

Emission lines from ionised gas are key diagnostics of star formation, metallicity, and ionisation conditions in galaxies. Interpreting spatially resolved observations from integral-field surveys (e.g. MaNGA, MUSE, JWST/NIRSpec) and comparing them with hydrodynamical simulations requires 3D photoionisation models that handle realistic geometries, dust attenuation, and synthetic instrument output. We present a new photoionisation module for the Monte Carlo radiative transfer code SKIRT that predicts emission-line luminosities of ionised gas in 3D, combining pre-computed Cloudy tables for gas temperature and opacity with a direct calculation of ion fractions and line emissivities. The local ionising radiation field (1-6 Ryd) is characterised by log U and four spectral-shape ratios; Cloudy tables map these to temperature and opacity, converging through SKIRT's existing iteration cycle. An inline solver then determines ion fractions from the converged field and temperature and evaluates line emissivities. We validate against Cloudy on 60 spherical shell models and against COLT on a Milky Way-analogue galaxy. On the 1D grid, hydrogen recombination lines agree with Cloudy to within a few per cent (Halpha median ratio 0.97) and the forbidden lines to within ~5%, except [S II] 6717 (1.23), whose offset traces a temperature overestimate near the ionisation front. In 3D, integrated luminosities agree with COLT to within 18% for the hydrogen lines and 2% for [N II], while [O III] and [S II] are elevated by ~70 and ~80%. Pixel-by-pixel correlation coefficients reach r >= 0.92, with luminosity-weighted scatter of 0.14-0.31 dex and broadly consistent BPT ratios. The module enables self-consistent synthetic observations in which ionised-gas emission lines, dust attenuation, and dust re-emission are computed in a single MCRT run, applicable to any hydrodynamical simulation.

Figures

Figures reproduced from arXiv: 2607.09961 by the authors.

Figure 1
Figure 1. Overview of the per-cell state-update procedure in DiffuseIonizedGasMix. At each SKIRT iteration, the local ion￾ising radiation field (1–6Ryd) is used to compute the ionisation parameter logU and the spectral shape ratios Ri (mean intensity ratios across five energy bins; Sect. 2.2), which together index pre-tabulated Cloudy grids to obtain the temperature and opacity. Emission-line luminosities are computed directl… view at source ↗
Figure 2
Figure 2. Five-bin structure overlaid on a BPASS SED (10 Myr, Z = 0.014; black). The step function (blue dashed) shows the mean flux in each bin. Vertical dotted lines mark the bin boundaries at 1.00, 1.80, 2.58, 3.52, 4.00, and 6.00 Ryd, which coincide with major ionisation edges as labelled. This binning is designed to accurately capture the dominant recombination and cooling channels with minimal spectral resolution. Cloud… view at source ↗
Figure 3
Figure 3. Absorption opacity from the standard STAB table as a function of wavelength for nine values of logU at fixed nH = 10 cm−3 and Ri fixed at the midpoints of their grid ranges. The three ionisation edges (H I 912 Å, He I 504 Å, He II 228 Å) are clearly resolved despite the coarse 5-bin radiation-field characterisation [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (12 more)
Figure 4
Figure 4. Figure 4: Radiation field characterisation as a function of radius for the representative benchmark model (Qion = 1049 s −1 , nH = 10 cm−3 , age 10 Myr). Top: ionisation parameter logU. Bottom: spectral-shape ratios logR2–logR5 (colours as labelled). Solid lines are Cloudy; dash…
Figure 5
Figure 5. Figure 5: Gas temperature as a function of radius for the representative benchmark model (Qion = 1049 s −1 , nH = 10 cm−3 , age 10 Myr, logU ≈ −0.6). The Cloudy reference (solid red) and the SKIRT radial-mean pro￾file (dashed blue) agree to within a few per cent across the ionis…
Figure 6
Figure 6. Figure 6: Temperature comparison across all 60 benchmark models. Ratio of mean SKIRT to mean Cloudy temperature versus ⟨T⟩Cloudy, coloured by logU. The solid line marks perfect agreement and dashed lines show ±5%. The median ratio is 1.020 with an L1 error of 2.6%. The logU prof…
Figure 7
Figure 7. Figure 7: compares SKIRT and Cloudy luminosities for all five emission lines across the 60 benchmark models; [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Median TSKIRT/TCloudy versus 1−r/rS across all 60 benchmark models (black curve; grey band: 16th–84th percentile). Coloured bars show the radial range producing 80% of each line’s luminosity (from Cloudy); dots mark the peak of dL/dr, the differential luminosity per un…
Figure 9
Figure 9. Figure 9: Emission-line luminosities from SKIRT (filled circles) and COLT (open squares) versus Cloudy across the 60 one-dimensional benchmark models, coloured by stellar age. The solid diagonal marks perfect agreement; dashed lines and the grey band indicate ±20%. Median ratios…
Figure 10
Figure 10. Figure 10: Physical conditions of the ionised gas in the Kannan et al. (2020) galaxy as inferred by SKIRT. Left: ionisation parameter logU; centre: gas temperature T. Both maps show a face-on midplane cut through the disc, projected onto a 1024×1024 pixel grid. Right: nH–T phase…
Figure 11
Figure 11. Figure 11: Integrated gas emission spectrum of the Kannan et al. (2020) Milky Way-analogue galaxy at face-on inclination (i = 0 ◦ ), as produced by SKIRT. The spectrum contains all 20 emission lines currently implemented (Sect. 2.7); the line list is extensible by supplying the …
Figure 12
Figure 12. Figure 12: Face-on (i = 0 ◦ ) emission-line maps of the Kannan et al. (2020) Milky Way-analogue galaxy. Top row: SKIRT (DiffuseIonizedGasMix); middle row: COLT reference; bottom row: log10(SKIRT/COLT) residual. From left to right: Hα λ6563, Hβ λ4861, [O III] λ5007, [N II] λ6583,…
Figure 13
Figure 13. Figure 13: Pixel-by-pixel comparison of SKIRT and COLT emission-line luminosities across all four inclinations. The dashed line marks perfect agreement. The integrated luminosity ratio ΣS/ΣC, Pearson correlation coefficient r, and luminosity-weighted L1 scatter (in dex) are quot…
Figure 14
Figure 14. Figure 14: Radial luminosity profiles (face-on, i = 0 ◦ ) for each emission line. Top sub-panels show the total luminosity per annulus (solid lines) and the cumulative luminosity L(r < R) (dash-dot/dotted lines). Blue lines are COLT; red lines are SKIRT. Bottom sub-panels show t…
Figure 15
Figure 15. Figure 15: [N II]-BPT diagram for the Kannan et al. (2020) Milky Way￾analogue galaxy (face-on, i = 0 ◦ ). Left: COLT; right: SKIRT. Hexbins are coloured by the cumulative Hα luminosity. The solid and dashed curves show the Kewley et al. (2001) and Kauffmann et al. (2003) demarca…

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. The TNG50-SKIRT Atlas: Spatially resolved synthetic galaxies from the ultraviolet to the submillimetre (DR2)

    astro-ph.GA 2026-08 accept novelty 5.0 of 10

    TNG50-SKIRT Atlas DR2 provides 1154 simulated galaxies with resolved UV-to-submm images, spectral cubes, and dust-aware synthetic observables, publicly released.

Reference graph

Works this paper leans on

70 extracted references · 1 linked inside Pith · cited by 1 Pith paper

  1. [1]

    2022, A&A, 666, A101 2 https://github.com/SKIRT/SKIRT9 Article number, page 14 of 18 Anand Utsav Kapoor ( ਅਨੰਦ ਉਤਸਵ ਕਪੂਰ ) et al.: Predicting ionised gas emission in 3D with SKIRT

    Baes, M., Camps, P ., & Matsumoto, K. 2022, A&A, 666, A101 2 https://github.com/SKIRT/SKIRT9 Article number, page 14 of 18 Anand Utsav Kapoor ( ਅਨੰਦ ਉਤਸਵ ਕਪੂਰ ) et al.: Predicting ionised gas emission in 3D with SKIRT. I

  2. [2]

    D., Lunttila, T., et al

    Baes, M., Gordon, K. D., Lunttila, T., et al. 2016, A&A, 590, A55

  3. [3]

    2011, ApJS, 196, 22

    Baes, M., Verstappen, J., De Looze, I., et al. 2011, ApJS, 196, 22

  4. [4]

    Bakes, E. L. O. & Tielens, A. G. G. M. 1994, ApJ, 427, 822

  5. [5]

    A., Phillips, M

    Baldwin, J. A., Phillips, M. M., & Terlevich, R. 1981, PASP , 93, 5

  6. [6]

    2016, MNRAS, 461, 3111

    Belfiore, F., Maiolino, R., Maraston, C., et al. 2016, MNRAS, 461, 3111

  7. [7]

    2022, A&A, 659, A26

    Belfiore, F., Santoro, F., Groves, B., et al. 2022, A&A, 659, A26

  8. [8]

    Burke, J. R. & Hollenbach, D. J. 1983, ApJ, 265, 223

Show all 70 references
  1. [9]

    & Baes, M

    Camps, P . & Baes, M. 2015, Astronomy and Computing, 9, 20

  2. [10]

    & Baes, M

    Camps, P . & Baes, M. 2020, Astronomy and Computing, 31, 100381

  3. [11]

    2013, A&A, 560, A35

    Camps, P ., Baes, M., & Saftly, W. 2013, A&A, 560, A35

  4. [12]

    U., & Grand, R

    Camps, P ., Behrens, C., Baes, M., Kapoor, A. U., & Grand, R. 2021, ApJ, 916, 39

  5. [13]

    2015, A&A, 580, A87

    Camps, P ., Misselt, K., Bianchi, S., et al. 2015, A&A, 580, A87

  6. [14]

    2003, PASP , 115, 763 Del Zanna, G., Dere, K

    Chabrier, G. 2003, PASP , 115, 763 Del Zanna, G., Dere, K. P ., Y oung, P . R., & Landi, E. 2021, ApJ, 909, 38

  7. [15]

    P ., Landi, E., Mason, H

    Dere, K. P ., Landi, E., Mason, H. E., Monsignori Fossi, B. C., & Y oung, P . R. 1997, A&AS, 125, 149

  8. [16]

    W., Victor, G

    Drake, G. W., Victor, G. A., & Dalgarno, A. 1969, Physical Review, 180, 25

  9. [17]

    J., Stanway, E

    Eldridge, J. J., Stanway, E. R., Xiao, L., et al. 2017, PASA, 34, e058

  10. [18]

    J., & Storey, P

    Ercolano, B., Barlow, M. J., & Storey, P . J. 2005, MNRAS, 362, 1038

  11. [19]

    J., Storey, P

    Ercolano, B., Barlow, M. J., Storey, P . J., & Liu, X.-W . 2003, MNRAS, 340, 1136

  12. [20]

    & Storey, P

    Ercolano, B. & Storey, P . J. 2006, MNRAS, 372, 1875

  13. [21]

    J., Chatzikos, M., Guzmán, F., et al

    Ferland, G. J., Chatzikos, M., Guzmán, F., et al. 2017, Rev. Mexicana Astron. Astrofis., 53, 385 Förster Schreiber, N. M. & Wuyts, S. 2020, ARA&A, 58, 661

  14. [22]

    L., Tsujita, A., et al

    Fujimoto, S., Faisst, A. L., Tsujita, A., et al. 2025, arXiv e-prints, arXiv:2510.16116

  15. [23]

    J., & Scott, P

    Grevesse, N., Asplund, M., Sauval, A. J., & Scott, P . 2010, Ap&SS, 328, 179

  16. [24]

    M., Dettmar, R.-J., Beckman, J

    Haffner, L. M., Dettmar, R.-J., Beckman, J. E., et al. 2009, Reviews of Modern Physics, 81, 969

  17. [25]

    & Gnedin, N

    Hui, L. & Gnedin, N. Y . 1997, MNRAS, 292, 27

  18. [26]

    2022, MNRAS, 511, 4005

    Kannan, R., Garaldi, E., Smith, A., et al. 2022, MNRAS, 511, 4005

  19. [27]

    2020, MNRAS, 499, 5732

    Kannan, R., Marinacci, F., Vogelsberger, M., et al. 2020, MNRAS, 499, 5732

  20. [28]

    2025, The Open Journal of Astro- physics, 8, 153

    Kannan, R., Puchwein, E., Smith, A., et al. 2025, The Open Journal of Astro- physics, 8, 153

  21. [29]

    2019, MNRAS, 485, 117

    Kannan, R., Vogelsberger, M., Marinacci, F., et al. 2019, MNRAS, 485, 117

  22. [30]

    U., Baes, M., van der Wel, A., et al

    Kapoor, A. U., Baes, M., van der Wel, A., et al. 2023, MNRAS, 526, 3871

  23. [31]

    U., Baes, M., van der Wel, A., et al

    Kapoor, A. U., Baes, M., van der Wel, A., et al. 2024, A&A, 692, A79

  24. [32]

    P ., Cadiou, C., et al

    Katz, H., Rey, M. P ., Cadiou, C., et al. 2025, arXiv e-prints, arXiv:2510.05201

  25. [33]

    2023, The Open Journal of Astrophysics, 6, 44

    Katz, H., Rosdahl, J., Kimm, T., et al. 2023, The Open Journal of Astrophysics, 6, 44

  26. [34]

    M., Tremonti, C., et al

    Kauffmann, G., Heckman, T. M., Tremonti, C., et al. 2003, MNRAS, 346, 1055

  27. [35]

    J., Dopita, M

    Kewley, L. J., Dopita, M. A., Sutherland, R. S., Heisler, C. A., & Trevena, J. 2001, ApJ, 556, 121

  28. [36]

    J., Nicholls, D

    Kewley, L. J., Nicholls, D. C., & Sutherland, R. S. 2019, ARA&A, 57, 511

  29. [37]

    2024, A&A, 689, A13

    Lauwers, A., Baes, M., Camps, P ., & Vander Meulen, B. 2024, A&A, 689, A13

  30. [38]

    F., Y ajima, H., Zhu, Q., & Maji, M

    Li, Y ., Gu, M. F., Y ajima, H., Zhu, Q., & Maji, M. 2020, MNRAS, 494, 1919

  31. [39]

    V ., Vogelsberger, M., Torrey, P ., & Springel, V

    Marinacci, F., Sales, L. V ., Vogelsberger, M., Torrey, P ., & Springel, V . 2019, MNRAS, 489, 4233

  32. [40]

    2023, A&A, 678, A175

    Matsumoto, K., Camps, P ., Baes, M., et al. 2023, A&A, 678, A175

  33. [41]

    H., Long, K

    Matthews, J. H., Long, K. S., Knigge, C., et al. 2025, MNRAS, 536, 879

  34. [42]

    2024, MNRAS, 535, 2889

    McCallum, L., Wood, K., Benjamin, R., Krishnarao, D., & Vandenbroucke, B. 2024, MNRAS, 535, 2889

  35. [43]

    2025, arXiv e-prints, arXiv:2510.13952

    McClymont, W., Smith, A., & Tacchella, S. 2025, arXiv e-prints, arXiv:2510.13952

  36. [44]

    2024, MNRAS, 532, 2016

    McClymont, W., Tacchella, S., Smith, A., et al. 2024, MNRAS, 532, 2016

  37. [45]

    Osterbrock, D. E. & Ferland, G. J. 2006, Astrophysics of gaseous nebulae and active galactic nuclei (University Science Books)

  38. [46]

    2017, A&A, 601, A92

    Peest, C., Camps, P ., Stalevski, M., Baes, M., & Siebenmorgen, R. 2017, A&A, 601, A92

  39. [47]

    2018, MNRAS, 473, 4077

    Pillepich, A., Springel, V ., Nelson, D., et al. 2018, MNRAS, 473, 4077

  40. [48]

    2018, MNRAS, 479, 994

    Rosdahl, J., Katz, H., Blaizot, J., et al. 2018, MNRAS, 479, 994

  41. [49]

    2014, A&A, 561, A77

    Saftly, W., Baes, M., & Camps, P . 2014, A&A, 561, A77

  42. [50]

    2013, A&A, 554, A10

    Saftly, W., Camps, P ., Baes, M., et al. 2013, A&A, 554, A10

  43. [51]

    2025, arXiv e-prints, arXiv:2508.21126

    Schaye, J., Chaikin, E., Schaller, M., et al. 2025, arXiv e-prints, arXiv:2508.21126

  44. [52]

    A., Bower, R

    Schaye, J., Crain, R. A., Bower, R. G., et al. 2015, MNRAS, 446, 521

  45. [53]

    E., Sanders, R

    Shapley, A. E., Sanders, R. L., Topping, M. W., et al. 2025, ApJ, 980, 242

  46. [54]

    2022, MNRAS, 517, 1

    Smith, A., Kannan, R., Tacchella, S., et al. 2022, MNRAS, 517, 1

  47. [55]

    Smith, A., Kannan, R., Tsang, B. T. H., Vogelsberger, M., & Pakmor, R. 2020, ApJ, 905, 27

  48. [56]

    2019, MNRAS, 484, 39

    Smith, A., Ma, X., Bromm, V ., et al. 2019, MNRAS, 484, 39

  49. [57]

    2015, MNRAS, 449, 4336

    Smith, A., Safranek-Shrader, C., Bromm, V ., & Milosavljević, M. 2015, MNRAS, 449, 4336

  50. [58]

    2010, MNRAS, 401, 791

    Springel, V . 2010, MNRAS, 401, 791

  51. [59]

    Storey, P . J. & Hummer, D. G. 1995, MNRAS, 272, 41

  52. [60]

    2018, MAPPINGS V: As- trophysical plasma modeling code, Astrophysics Source Code Library, record ascl:1807.005

    Sutherland, R., Dopita, M., Binette, L., & Groves, B. 2018, MAPPINGS V: As- trophysical plasma modeling code, Astrophysics Source Code Library, record ascl:1807.005

  53. [61]

    2022, MNRAS, 513, 2904

    Tacchella, S., Smith, A., Kannan, R., et al. 2022, MNRAS, 513, 2904

  54. [62]

    2021, A&A, 653, A34

    Vandenbroucke, B., Baes, M., Camps, P ., et al. 2021, A&A, 653, A34

  55. [63]

    & Camps, P

    Vandenbroucke, B. & Camps, P . 2020, A&A, 641, A66

  56. [64]

    & Wood, K

    Vandenbroucke, B. & Wood, K. 2018, Astronomy and Computing, 23, 40 Vander Meulen, B., Camps, P ., Savić, Ð., et al. 2024, A&A, 689, A297 Vander Meulen, B., Camps, P ., Stalevski, M., & Baes, M. 2023, A&A, 674, A123

  57. [65]

    2024, A&A, 691, A19

    Venturi, G., Carniani, S., Parlanti, E., et al. 2024, A&A, 691, A19

  58. [66]

    Verner, D. A. & Ferland, G. J. 1996, ApJS, 103, 467

  59. [67]

    A., Ferland, G

    Verner, D. A., Ferland, G. J., Korista, K. T., & Y akovlev, D. G. 1996, ApJ, 465, 487

  60. [68]

    2017, Astronomy and Computing, 20, 16

    Verstocken, S., Van De Putte, D., Camps, P ., & Baes, M. 2017, Astronomy and Computing, 20, 16

  61. [69]

    Weingartner, J. C. & Draine, B. T. 2001, ApJS, 134, 263

  62. [70]

    S., & Ercolano, B

    Wood, K., Mathis, J. S., & Ercolano, B. 2004, MNRAS, 348, 1337 Y ajima, H., Li, Y ., Zhu, Q., & Abel, T. 2012, MNRAS, 424, 884 Article number, page 15 of 18 A&A proofs: manuscript no. aanda Appendix A: Convergence of the Milky Way simulation Figure A.1 shows the convergence be...

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.