Pith. sign in

REVIEW 4 major objections 5 minor 42 references

Element nucleosynthetic origins from abundance spatial distributions beyond the Milky Way

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read In NGC 5253, oxygen and sulfur abundances show injection scales of ~46–62 pc, matching supernova blast waves, while nitrogen is injected on a ~7 pc scale, and oxygen-sulfur correlation far exceeds either's correlation with nitrogen.

desk verdict First independent extragalactic N and S abundance correlation maps, with a qualitative signal that is likely real but quantitative parameters that are over-fitted. read the letter →

arxiv 2506.21365 v1 pith:2GUGH6YU submitted 2025-06-26 astro-ph.GA

Zefeng Li , Mark R. Krumholz , Anna F. McLeod , A. Mark Swinbank , Emily Wisnioski , J. Trevor Mendel , Francesco Belfiore , Giovanni Cresci
show 2 more authors
Giacomo Venturi Jialai Kang
This is my paper · ORCID
classification astro-ph.GA
keywords galaxies:abundancesISMnucleosynthesisNGC5253integralfieldspectroscopyspatialautocorrelationcross-correlationinjectionwidth
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

This paper tries to establish that the spatial distribution of an element's abundance in a galaxy encodes its nucleosynthetic origin, and that the signal can be read from integral-field spectra beyond the Milky Way. Using MUSE data on the dwarf galaxy NGC 5253 at 3.5 pc resolution, it produces direct-method maps of oxygen, nitrogen, and sulfur and measures their spatial auto- and cross-correlations. Fitted injection widths place oxygen and sulfur on scales of ~46–62 pc, matching the maximum radii of supernova blast waves, while nitrogen's injection width is a factor of ~8 smaller, consistent with AGB star winds. The oxygen-sulfur zero-lag cross-correlation is far higher than either's correlation with nitrogen, showing that elements synthesized in the same sites stay spatially coupled in the gas. If correct, the method opens a new observational avenue for testing nucleosynthesis and implies stellar abundance patterns are structured into a small number of correlated groups.

What carries the argument

The stochastically-forced diffusion model of Krumholz & Ting (2018) for the abundance fluctuation field, expressed as a two-point autocorrelation function $\xi_{\rm model}(r)$ (Eq. 15), with parameters $w_{\rm inj}$ (the injection width, the characteristic radius of the region where a nucleosynthetic event deposits fresh metals), $l$ (the turbulent mixing length), and $f$ (the factor by which observational uncertainties inflate the variance). The model treats the metal field as a random injection process subsequently diffused by ISM turbulence, and it is this functional form that converts the measured autocorrelation shapes into physical injection scales. The cross-correlation analysis separately quantifies the shared spatial structure between pairs of elements, with zero-lag estimates corrected by the same variance-inflation factors $f_{XX}$, $f_{YY}$.

What would settle it

Re-derive the oxygen, nitrogen, and sulfur abundance maps for NGC 5253 using a directly measured [O III] 4363 auroral temperature (or independent temperature diagnostics) and recompute the autocorrelation fits; if the inferred injection widths shift by more than the quoted uncertainties, the $w_{\rm inj}$ values are not robust measures of nucleosynthetic injection scales.

Watch

Extended reading notes

Core claim

The central discovery is that the two-point autocorrelation functions of O, N, and S abundance maps in NGC 5253 are quantitatively different and match theoretical injection scales. Fitting the stochastically-forced diffusion model of Krumholz & Ting (2018) yields injection widths of $w_{\rm inj} = 61.5 \pm 0.3$ pc for oxygen, $7.3 \pm 0.2$ pc for nitrogen, and $45.7 \pm 0.7$ pc for sulfur. The O and S widths agree with the characteristic maximum radii of supernova blast waves (~59–67 pc), while the nitrogen width is a factor of ~8 smaller, attributed to AGB star winds with a subdominant ~20% core-collapse supernova contribution. After correcting for measurement noise, the zero-lag cross-correlations are $\Xi_{\rm NO}(0) = 0.691 \pm 0.007$, $\Xi_{\rm OS}(0) = 1.949 \pm 0.006$, and $\Xi_{\rm NS}(0) = 0.456 \pm 0.014$, so oxygen and sulfur are far better correlated with each other than with nitrogen. The paper argues these statistics open a new observational avenue for nucleosynthesis and imply stellar abundance patterns are structured into a small number of correlated groups.

Load-bearing premise

The entire conclusion rests on the assumption that a single injection scale and a single mixing length describe how each element is spread in NGC 5253; if the real fluctuation pattern is shaped by multiple injection scales, outflows, or ionization variations that masquerade as abundance changes, the fitted widths would not measure nucleosynthetic origins.

Editorial extensions

If this is right

  • The two-point autocorrelation of oxygen and sulfur abundance maps in galaxies directly encodes supernova remnant sizes, making it possible to measure blast-wave radii even in galaxies without recent supernovae.
  • The ~8-fold smaller nitrogen injection width, detected at >10σ significance, provides a quantitative ISM signature that nitrogen enrichment is dominated by AGB stars with a subdominant core-collapse supernova component.
  • Element pairs sharing a nucleosynthetic origin (O and S) have substantially higher zero-lag cross-correlation than pairs with different origins, confirming that production-site grouping leaves an imprint on gas-phase abundance correlations.
  • Because gas-phase abundances are highly structured by nucleosynthetic group, stellar chemical abundances are likely decomposable into a small number of correlated components, which sets limits on the information recoverable by chemical tagging.

Reading between the lines

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

  • If the injection width is set by the mechanical energy of the source, the measured O/N width ratio could be used to calibrate yield models for AGB winds versus supernovae in other nearby galaxies.
  • Extending the same analysis to elements such as argon or neon, where auroral lines are also detectable, could test whether the correlation grouping follows the predicted CC-SN versus Type-Ia classification.
  • A multi-galaxy sample spanning different star formation histories could determine whether the nitrogen injection width varies with stellar mass or metallicity, which would connect the ISM correlation signal to the AGB contribution.
  • The result implies that chemical tagging in external galaxies must account for the gas-phase correlation structure; spatial abundance maps could serve as a prior on stellar abundance covariance in unresolved stellar populations.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper presents a new observational analysis of the dwarf galaxy NGC 5253 using MUSE integral-field data. The authors derive maps of oxygen, nitrogen, and sulfur abundances using direct electron-temperature methods, with nitrogen measured independently via the [N II] auroral line. They compute two-point spatial autocorrelation and cross-correlation functions of the abundance residual maps and fit a stochastic-diffusion model (Krumholz & Ting 2018, Eq. 15) to infer the injection widths of each element. They report winj = 61.5 ± 0.3 pc for O, 7.3 ± 0.2 pc for N, and 45.7 ± 0.7 pc for S, interpreting the large O and S values as the scale of supernova blast waves and the much smaller N value as evidence for a dominant AGB injection channel. The zero-lag cross-correlation between O and S is found to be higher than between N and O or N and S, consistent with shared core-collapse supernova origins. The paper argues that these spatial statistics open a new avenue to test nucleosynthetic models and constrain chemical-tagging methods.

Significance. If the quantitative results hold, the paper would provide the first extragalactic measurements of injection scales for oxygen, nitrogen, and sulfur, and would demonstrate a promising link between abundance spatial statistics and nucleosynthetic origin. The qualitative ordering is visible directly in the raw autocorrelation and cross-correlation functions (Figures 2 and 3), and the robustness of the nitrogen result to an alternative temperature diagnostic is tested in Appendix A, which is a genuine strength. The comparison to simulation predictions by Zhang et al. (2024) adds context and a falsifiable element. However, the quantitative claims rely on a forward model whose validity for the observed, inclined, three-dimensional geometry is not demonstrated, and on error bars that explicitly exclude the dominant systematic uncertainties. The significance is therefore moderate: the result is plausible but the quantitative conclusions are not yet supported with the current level of validation.

major comments (4)
  1. [§3.2, Eq. (15)] The inference of injection widths from the stochastic-diffusion model assumes a single characteristic injection scale and a single mixing length in an isotropic two-dimensional field, but NGC 5253 is observed at 64° inclination (b/a = 0.43) and the maps are line-of-sight projections through a clumpy, rotating, three-dimensional ISM. The paper does not validate that Eq. (15) recovers input injection scales when applied to projected, multi-scale abundance fields with realistic noise. Because the entire physical interpretation rests on the mapping between the measured autocorrelation shape and winj, I request a synthetic test: inject a field with known winj, project it at the observed inclination, add realistic noise, run the same fitting pipeline, and demonstrate that the recovered winj is unbiased. Without such a test, the quoted values (61.5 pc, 7.3 pc, 45.7 pc) cannot be claimed as measurements of nucleosynthetic injection scales.
  2. [§3.2, paragraph on observational uncertainty] The paper explicitly states that the bootstrap uncertainties 'only take measurement errors into account, and are sub-dominant compared to the much larger systematic uncertainties arising from the assumptions embedded in the line diagnostics used to convert fluxes to metallicity measurements.' Despite this, the headline claim is that the nitrogen injection width is a factor of ~8 smaller, 'detected at very high statistical significance (≫10σ).' This significance is computed from bootstrap-only errors and does not include any contribution from systematic uncertainties in the temperature calibration, reddening, or abundance conversion. As written, the ≫10σ statement is not a claim about the physical quantity winj but about the noise in the flux measurements only. The authors should either propagate the systematic uncertainties or explicitly rephrase the significance claim to refer solely to the measurement-noise component.
  3. [§3.3 and Appendix C] The corrected zero-lag cross-correlation for O-S, Ξ_OS(0) = 1.949 ± 0.006, exceeds the mathematical upper bound of 1 for a true correlation coefficient. Appendix C attributes this overshoot to neglected correlations in the measurement uncertainties, specifically the shared electron-temperature scale. This implies that Eq. (17) is not an adequate model for the noise correction in this dataset. The main-text conclusion that 'the recovered O-S zero-lag cross-correlation Ξ_OS(0) is significantly larger than Ξ_NO(0) and Ξ_NS(0)' relies on this corrected quantity, even though the raw cross-correlation functions in Figure 3 do show the same ordering. I recommend either (a) developing a noise-deconvolution or explicit covariance model that produces physical (≤1) zero-lag correlations, or (b) basing the quantitative comparison on the uncorrected cross-correlation functions and presenting the corrected values only as an auxiliary diagnostic with clear caveats.
  4. [§3.2, Figure 2 and fit] The paper provides no goodness-of-fit statistic, residual analysis, or comparison of Eq. (15) against alternative models (e.g., a two-component injection scale, or a model with an independent outer scale). The fitted winj values are the central quantitative products, and without any assessment of whether the model actually describes the measured autocorrelation functions, the reader cannot evaluate whether the inference is robust or whether the parameters are constrained by a few features of the curve. I request that the authors include residual plots and, if possible, a model comparison (e.g., AIC or a likelihood-ratio test) against at least one alternative functional form.
minor comments (5)
  1. [§3.3, paragraph after Figure 4] The word 'potentiallt' in the sentence 'so was missing a potentiallt significant source of nitrogen' should be 'potentially'.
  2. [Appendix A, Figure 6 and text] The uncertainty on Ξ_N'S(0) is quoted as 1.195 ± 0.006 in the text of Appendix A but as ±0.06 in the caption of Figure 6; these should be made consistent.
  3. [Eq. (15)] The placement of the factor '× 1/f' after the integral is confusing; it is not clear whether the factor multiplies the entire integral or only part of it, and it should be moved before the integral or otherwise clarified.
  4. [Figure 2, lower panel] The corrected autocorrelation functions shown in the lower panel have no propagated uncertainties, which makes it hard to judge the fit quality; a propagated error envelope would be helpful.
  5. [Figure 2, caption] The colored 'stripes' indicating the injection widths are not clearly described in the caption; please state what the shaded regions represent and how they were derived.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the self-cited diffusion model is an independent forward model, and the injection-width interpretation is anchored by external benchmarks.

full rationale

The derivation chain is self-contained with respect to the circularity patterns checked. The abundance maps are produced from independent emission-line diagnostics (Eqs. 8-12), and the auto/cross-correlation functions (Eqs. 13-16) are direct data products whose shapes are not imposed by the nucleosynthetic-origin hypothesis. The injection widths winj are free parameters fit by MCMC to the measured autocorrelation shapes; they are not outputs of a theory that already assumes O and S from core-collapse supernovae and N from AGB stars. The Krumholz & Ting (2018) model (Eq. 15) is a parameter-free forward model of stochastically forced diffusion; it does not encode the nucleosynthetic origin of any element, and the fitted winj values are compared against external blast-wave benchmarks (Kolborg et al. 2022; Draine 2011). The cross-correlation correction (Eq. 17) uses noise-variance factors fit to the same maps, but the measured zero-lag ordering is not imposed by that correction: the uncorrected ξOS exceeds ξNO and ξNS before correction, and the paper explicitly flags the super-unity corrected value Ξ_OS(0)=1.95>1 as a systematic artifact of shared temperature diagnostics (Appendix C), which is a correctness caveat rather than a circular reduction. The self-citations (Krumholz & Ting 2018; Li et al. 2021, 2023; Zhang et al. 2024) are method and prediction citations with stated assumptions and external comparators; none is a uniqueness theorem or an ansatz that smuggles the conclusion into the input. The score 2 reflects repeated same-group citations in method and interpretation, but these are not load-bearing circular reductions.

Assumptions & free parameters 9 free parameters · 5 assumptions · 0 invented entities

The central quantitative results (injection widths and zero-lag cross-correlations) are derived from fitting a parametric model to the measured autocorrelation functions. The model has three free parameters per element, nine in total. No new physical entities are introduced. The interpretation relies on literature nucleosynthetic fractions and on the validity of the diffusion model.

free parameters (9)
  • winj (oxygen) = 61.5 pc
    Injection width parameter fitted to the oxygen autocorrelation function via Eq. 15; it underpins the claim that oxygen is injected on supernova-blast-wave scales.
  • l (oxygen) = 4.8 pc
    Correlation length parameter for oxygen fitted at 4.8 pc; it is poorly constrained by the data.
  • f (oxygen) = 2.347
    Variance-inflation factor for oxygen, absorbing measurement noise; fitted value 2.347.
  • winj (nitrogen) = 7.3 pc
    Injection width fitted to the nitrogen autocorrelation function; at 7.3 pc it is the key discriminator of the paper.
  • l (nitrogen) = 7.0 pc
    Correlation length parameter for nitrogen, fitted at 7.0 pc.
  • f (nitrogen) = 2.209
    Variance-inflation factor for nitrogen, fitted value 2.209.
  • winj (sulfur) = 45.7 pc
    Injection width fitted to the sulfur autocorrelation function; fitted value 45.7 pc.
  • l (sulfur) = 54.9 pc
    Correlation length parameter for sulfur, fitted at 54.9 pc.
  • f (sulfur) = 2.135
    Variance-inflation factor for sulfur, fitted value 2.135.
assumptions (5)
  • domain assumption The stochastic diffusion model (Krumholz & Ting 2018) with a single injection scale and single mixing length describes the abundance fluctuation field of each element.
    Employed at Section 3.2, Eq. 15, to interpret autocorrelation shapes as injection widths. If false, fitted winj does not measure a physical injection scale.
  • domain assumption The direct-method temperature calibrations of Pérez-Montero (2017) correctly convert line ratios into electron temperatures and abundances in NGC 5253.
    Used throughout Section 3.1, Eqs. 1-12. Systematic uncertainties from these calibrations are not quantified.
  • domain assumption The measured ion species (O+, O++, S+, S++, N+) trace the total element abundances, with negligible unmeasured ionization states.
    Assumed in Section 3.1. Spatial variations in ionization parameter could imprint non-nucleosynthetic structure on the abundance maps.
  • domain assumption Literature nucleosynthetic origin fractions (N: 80% AGB, 20% CC; O: 100% CC; S: 80% CC, 20% Ia) are correct.
    Adopted from ChemPy/Rybizki et al. 2017 and used in Figure 2 and Section 3.2 to connect winj to origin sites. These fractions are not derived in this paper.
  • domain assumption Standard observational assumptions (Case B recombination ratio 2.86, R_V = 4.05 extinction curve, electron density 100 cm^-3) apply.
    Part of Section 2. Incorrect choices would shift abundances but are standard for the field.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Element nucleosynthetic origins from abundance spatial distributions beyond the Milky Way." pith.science (2026). https://pith.science/paper/2GUGH6YU

@misc{pith2026250621365,
  author       = {Pith},
  title        = {Pith review of: Element nucleosynthetic origins from abundance spatial distributions beyond the Milky Way},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2GUGH6YU}},
  note         = {Machine review of arXiv:2506.21365}
}
read the original abstract

An element's astrophysical origin should be reflected in the spatial distribution of its abundance, yielding measurably different spatial distributions for elements with different nucleosynthetic sites. However, most extragalactic multi-element analyses of gas-phase abundances to date have been limited to small numbers of sightlines, making statistical characterization of differences in spatial distributions of elements impossible. Here we use integrated field spectroscopic data covering the full face of the nearby dwarf galaxy NGC 5253 sampled at 3.5-pc resolution to produce maps of the abundances of oxygen, nitrogen, and sulfur using independent direct methods. We find strong evidence for differences in the elements' spatial statistics that mirror their predicted nucleosynthetic origins: the spatial distributions of oxygen and sulfur, both predominantly produced in core-collapse supernovae, indicate that initial injection occurs on larger scales than for nitrogen, which is predominantly produced by asymptotic giant branch stars. All elements are well-correlated but oxygen and sulfur are much better correlated with each other than with nitrogen, consistent with recent results for stellar abundances in the Milky Way. These findings both open a new avenue to test nucleosynthetic models, and make predictions for the structure of stellar chemical abundance distributions.

Figures

Figures reproduced from arXiv: 2506.21365 by the authors.

Figure 1
Figure 1. [Siii] electron temperature map (lower left) and nitrogen (upper left), oxygen (upper right), and sulfur (lower right) abundance maps for NGC 5253. The fields-of-view are centered on the galactic center and are the same in each panel. The coverage of the nitrogen abundance map is smaller than those of oxygen and sulfur maps because of the faintness of the auroral line [Nii]λ5755 and the resulting low coverage of ele… view at source ↗
Figure 2
Figure 2. Two-point autocorrelation functions (thick solid lines) for oxygen (blue), nitrogen (red), and sulfur (yellow) and our best-fit parametric models (thin dashed curves). In both panels correlation = 0 means uncorrelated and = 1 means perfectly correlated. The upper panel shows the measured autocorrelation, while the lower panel shows the true autocorrelation corrected by our median estimate for the factor by which the… view at source ↗
Figure 3
Figure 3. Cross-correlation functions for N-O (purple), O-S (green), and N-S (orange), uncorrected for observational un￾certainties. Correlation increases from lower (uncorrelated) to upper (nearly perfectly correlated). At scales smaller than 60 pc, the O-S cross-correlation functions are most correlated, followed by those of N-O and N-S, thanks to the same origin site (core-collapse supernovae) that oxygen and sulfur share.… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Posterior PDFs of true zero-lag cross-correlations ΞXY(0) for each pair of elements, corrected for observational uncertainties. Correction increases from left to right. The vertical dashed lines indicate the zero-lag cross-correlations predicted by simulations (C. Zhan…
Figure 5
Figure 5. Figure 5: Alternative versions of the upper left panel of [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Alternative versions of [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: The distributions of our three fit parameters (injection width winj, correlation length l, and the factor by which the autocorrelation is reduced by measurement uncertainties f), describing the autocorrelation functions of nitrogen (upper left), oxygen (upper right), a…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

42 extracted references · 7 canonical work pages

  1. [1]

    2021, MNRAS, 505, 2361, doi: 10.1093/mnras/stab1467 Arellano-C´ ordova, K

    Amayo, A., Delgado-Inglada, G., & Stasi´ nska, G. 2021, MNRAS, 505, 2361, doi: 10.1093/mnras/stab1467 Arellano-C´ ordova, K. Z., Rodr ´ ıguez, M., Mayya, Y. D., & Rosa-Gonz´ alez, D. 2016, MNRAS, 455, 2627, doi: 10.1093/mnras/stv2461 Arellano-C´ ordova, K. Z., Berg, D. A., Mingozzi, M., et al. 2024, ApJ, 968, 98, doi: 10.3847/1538-4357/ad34cf

    doi:10.1093/mnras/stab1467 2021

  2. [2]

    2010, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol

    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

    doi:10.1117/12.856027 2010

  3. [3]

    A., Skillman, E

    Berg, D. A., Skillman, E. D., Croxall, K. V., et al. 2015, ApJ, 806, 16, doi: 10.1088/0004-637X/806/1/16

    doi:10.1088/0004-637x/806/1/16 2015

  4. [4]

    2025, arXiv e-prints, arXiv:2502.00649, doi: 10.48550/arXiv.2502.00649

    Bresolin, F., Fern´ andez-Arenas, D., Rousseau-Nepton, L., et al. 2025, arXiv e-prints, arXiv:2502.00649, doi: 10.48550/arXiv.2502.00649

    arXiv doi:10.48550/arxiv.2502.00649 2025

  5. [5]

    C., et al

    Calzetti, D., Armus, L., Bohlin, R. C., et al. 2000, ApJ, 533, 682, doi: 10.1086/308692

    doi:10.1086/308692 2000

  6. [6]

    2021, A&A, 654, A151, doi: 10.1051/0004-6361/202141779 Dom ´ ınguez-Guzm´ an, G., Rodr ´ ıguez, M., Garc ´ ıa-Rojas, J.,

    Soubiran, C. 2021, A&A, 654, A151, doi: 10.1051/0004-6361/202141779 Dom ´ ınguez-Guzm´ an, G., Rodr ´ ıguez, M., Garc ´ ıa-Rojas, J.,

    doi:10.1051/0004-6361/202141779 2021

  7. [7]

    2022, MNRAS, 517, 4497, doi: 10.1093/mnras/stac2974

    Esteban, C., & Toribio San Cipriano, L. 2022, MNRAS, 517, 4497, doi: 10.1093/mnras/stac2974

    doi:10.1093/mnras/stac2974 2022

  8. [8]

    Draine, B. T. 2011, Physics of the Interstellar and Intergalactic Medium

    2011

Show all 42 references
  1. [9]

    2024, ApJL, 962, L25, doi: 10.3847/2041-8213/ad23e3

    Gao, Y., Gu, Q., Zhou, P., Shi, Y., & Li, X. 2024, ApJL, 962, L25, doi: 10.3847/2041-8213/ad23e3

    2024 doi

  2. [10]

    J., Hogg, D

    Griffith, E. J., Hogg, D. W., Hasselquist, S., et al. 2024, arXiv e-prints, arXiv:2410.22121, doi: 10.48550/arXiv.2410.22121

    2024 doi arXiv

  3. [11]

    Thuan, T. X. 2006, A&A, 448, 955, doi: 10.1051/0004-6361:20053763

    2006 doi

  4. [12]

    Johnson, J. A. 2019, Science, 363, 474, doi: 10.1126/science.aau9540

    2019 doi

  5. [13]

    J., Nicholls, D

    Kewley, L. J., Nicholls, D. C., & Sutherland, R. S. 2019, ARA&A, 57, 511, doi: 10.1146/annurev-astro-081817-051832

    2019 doi

  6. [14]

    I., & Lugaro, M

    Kobayashi, C., Karakas, A. I., & Lugaro, M. 2020, ApJ, 900, 179, doi: 10.3847/1538-4357/abae65

    2020 doi

  7. [15]

    A., Skillman, E

    Kobulnicky, H. A., Skillman, E. D., Roy, J.-R., Walsh, J. R., & Rosa, M. R. 1997, ApJ, 477, 679, doi: 10.1086/303742

    1997 doi

  8. [16]

    N., Martizzi, D., Ramirez-Ruiz, E., et al

    Kolborg, A. N., Martizzi, D., Ramirez-Ruiz, E., et al. 2022, ApJL, 936, L26, doi: 10.3847/2041-8213/ac8c98

    2022 doi

  9. [17]

    A., et al

    Kreckel, K., Ho, I.-T., Blanc, G. A., et al. 2020, MNRAS, arXiv:2009.02342. https://arxiv.org/abs/2009.02342

    2020 arXiv

  10. [18]

    R., & Ting, Y.-S

    Krumholz, M. R., & Ting, Y.-S. 2018, MNRAS, 475, 2236, doi: 10.1093/mnras/stx3286

    2018 doi

  11. [19]

    L., Irwin, M

    Kumari, N., James, B. L., Irwin, M. J., Amor ´ ın, R., & P´ erez-Montero, E. 2018, MNRAS, 476, 3793, doi: 10.1093/mnras/sty402

    2018 doi

  12. [20]

    Lauberts, A., & Valentijn, E. A. 1989, The surface photometry catalogue of the ESO-Uppsala galaxies

    1989

  13. [21]

    R., & S´ anchez, S

    Li, S.-L., Li, Z., Wisnioski, E., Krumholz, M. R., & S´ anchez, S. F. 2025, MNRAS, 536, 430, doi: 10.1093/mnras/stae2603

    2025 doi

  14. [22]

    R., Wisnioski, E., et al

    Li, Z., Krumholz, M. R., Wisnioski, E., et al. 2021, MNRAS, 504, 5496, doi: 10.1093/mnras/stab1263

    2021 doi

  15. [23]

    T., et al

    Li, Z., Wisnioski, E., Mendel, J. T., et al. 2023, MNRAS, 518, 286, doi: 10.1093/mnras/stac3028

    2023 doi

  16. [24]

    C., Hammer, F., Yin, S

    Liang, Y. C., Hammer, F., Yin, S. Y., et al. 2007, A&A, 473, 411, doi: 10.1051/0004-6361:20077436 L´ opez-S´ anchez,´A. R., Mesa-Delgado, A., L´ opez-Mart ´ ın, L., & Esteban, C. 2011, MNRAS, 411, 2076, doi: 10.1111/j.1365-2966.2010.17847.x

    2007

  17. [25]

    Luridiana, V., Morisset, C., & Shaw, R. A. 2015, A&A, 573, A42, doi: 10.1051/0004-6361/201323152

    2015 doi

  18. [26]

    2023, A&A, 670, A92, doi: 10.1051/0004-6361/202244895

    Marasco, A., Belfiore, F., Cresci, G., et al. 2023, A&A, 670, A92, doi: 10.1051/0004-6361/202244895

    2023 doi

  19. [27]

    2021, MNRAS, 508, 489, doi: 10.1093/mnras/stab2554

    Metha, B., Trenti, M., & Chu, T. 2021, MNRAS, 508, 489, doi: 10.1093/mnras/stab2554

    2021 doi

  20. [28]

    2022, MNRAS, 514, 4465, doi: 10.1093/mnras/stac1484 11

    Metha, B., Trenti, M., Chu, T., & Battisti, A. 2022, MNRAS, 514, 4465, doi: 10.1093/mnras/stac1484 11

    2022 doi

  21. [29]

    R., & V ´ ılchez, J

    Monreal-Ibero, A., Walsh, J. R., & V ´ ılchez, J. M. 2012, A&A, 544, A60, doi: 10.1051/0004-6361/201219543

    2012 doi

  22. [30]

    K., Wheeler, A

    Ness, M. K., Wheeler, A. J., McKinnon, K., et al. 2022, ApJ, 926, 144, doi: 10.3847/1538-4357/ac4754

    2022 doi

  23. [31]

    Osterbrock, D. E. 1989, Astrophysics of gaseous nebulae and active galactic nuclei P´ erez-Montero, E. 2017, PASP, 129, 043001, doi: 10.1088/1538-3873/aa5abb P´ erez-Montero, E., & Contini, T. 2009, MNRAS, 398, 949, doi: 10.1111/j.1365-2966.2009.15145.x

    1989

  24. [32]

    L., Horiuchi, S., Anderson, L

    Pineda, J. L., Horiuchi, S., Anderson, L. D., et al. 2024, ApJ, 973, 89, doi: 10.3847/1538-4357/ad615a

    2024 doi

  25. [33]

    2017, A&A, 605, A59, doi: 10.1051/0004-6361/201730522

    Rybizki, J., Just, A., & Rix, H.-W. 2017, A&A, 605, A59, doi: 10.1051/0004-6361/201730522

    2017 doi

  26. [34]

    2023, MNRAS, 526, 6214, doi: 10.1093/mnras/stad3108

    Mathur, S. 2023, MNRAS, 526, 6214, doi: 10.1093/mnras/stad3108

    2023 doi

  27. [35]

    Ting, Y.-S., & Weinberg, D. H. 2022, ApJ, 927, 209, doi: 10.3847/1538-4357/ac5023

    2022 doi

  28. [36]

    B., Courtois, H

    Tully, R. B., Courtois, H. M., & Sorce, J. G. 2016, AJ, 152, 50, doi: 10.3847/0004-6256/152/2/50

    2016 doi

  29. [37]

    A., et al

    Vangioni, E., Dvorkin, I., Olive, K. A., et al. 2018, MNRAS, 477, 56, doi: 10.1093/mnras/sty559

    2018 doi

  30. [38]

    M., & Esteban, C

    Vilchez, J. M., & Esteban, C. 1996, MNRAS, 280, 720, doi: 10.1093/mnras/280.3.720

    1996 doi

  31. [39]

    2018, A&A, 610, L16, doi: 10.1051/0004-6361/201732395

    Vincenzo, F., & Kobayashi, C. 2018, A&A, 610, L16, doi: 10.1051/0004-6361/201732395

    2018 doi

  32. [40]

    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

    2020 doi

  33. [41]

    G., Kreckel, K., Belfiore, F., et al

    Williams, T. G., Kreckel, K., Belfiore, F., et al. 2022, MNRAS, 509, 1303, doi: 10.1093/mnras/stab3082

    2022 doi

  34. [42]

    Zhang, C., Li, Z., Hu, Z., & Krumholz, M. R. 2024, arXiv e-prints, arXiv:2411.01518, doi: 10.48550/arXiv.2411.01518 12 12 + log(N'/H) 10"=177 pc 6.0 6.5 7.0 12 + log(N'/H) 0 50 100 150 200 250 300 350 r (pc) 0.0 0.5 1.0XX(r) X = oxygen X = nitrogen (alternative) X = sulfur Oxy...

    2024 doi arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.