Pith. sign in

REVIEW 3 major objections 6 minor 48 references

VLBI studies of FLASH H I 21-cm absorbers -- I

T0 review · 3 major / 6 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 Parsec-scale radio cores contain enough flux to account for all of the detected H I 21-cm absorption in seven of twelve FLASH absorbers, and covering-factor corrections could raise the population's velocity-integrated optical depths by up…

desk verdict Useful first VLBI continuum look at FLASH HI absorbers, but the factor-of-three VOD correction rests on an unquantified 1.4-to-0.7 GHz comparison that deserves scrutiny. read the letter →

arxiv 2507.19957 v1 pith:3B2PVMYB submitted 2025-07-26 astro-ph.GA

J.N.H.S. Aditya , Elaine M. Sadler , Raffaella Morganti , L. Y. Petrov , An Tao , Emily F. Kerrison , Elizabeth K. Mahony , Hyein Yoon
show 4 more authors
Renzhi Su Matthew Whiting Vanessa A. Moss Simon Weng
This is my paper · ORCID
classification astro-ph.GA
keywords galaxies:activeISMmethods:observationalradiolines:galaxiescontinuum:generalsurveysHI21-cmabsorptionverylongbaselineinterferometry
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 uses VLBA 1.4 GHz images of twelve neutral-hydrogen (H I) 21-cm absorbers discovered by FLASH, an untargeted ASKAP survey at redshifts 0.4-0.7, to ask whether the absorption is produced against the compact radio core or against more extended emission. It finds that in seven of the twelve sources the parsec-scale core alone has enough flux density to produce the entire absorption seen in the ASKAP spectra, and that for three compact sources with projected sizes of 305-409 pc most of the milliarcsecond-scale emission is probably occulted by gas. Using the measured core and total VLBA flux densities as lower limits on the gas covering factor, the paper derives corrected velocity-integrated optical depths (VODs) that are a median factor of 2.8 higher than the uncorrected ASKAP values, so the VOD distribution at 0.4

What carries the argument

The central device is the gas covering factor f introduced by Briggs & Wolfe (1983), related to the measured and true optical depth by $\tau_{\rm true} = -\ln\bigl(1 + (e^{-\tau_{\rm app}}-1)/f\bigr)$. Because f is unknown, the paper turns the problem around: for each source it assigns a minimum occulted flux density $S_{\rm min}$ equal to either the VLBA core flux or the total VLBA flux and asks whether $S_{\rm min}/S_{\rm RACS}$ exceeds the observed peak absorption fraction ${\rm PF} = |S_{\rm HI}|/S_{\rm ASKAP}$; whichever component first satisfies this inequality is taken as a lower limit on f. The VLBA continuum images at 1.4 GHz provide the parsec-scale morphological classification (core-jet, two-sided jet, compact double, complex, unresolved), the core identification, and the core and total flux densities, while RACS-mid supplies the matched L-band total flux density $S_{\rm RACS}$. The same inequality then yields a finite upper limit on the true optical depth per channel, and integrating over the line gives the corrected velocity-integrated optical depth.

What would settle it

Direct spectroscopic VLBI observations of the H I 21-cm line at the redshifted frequency (~700-1000 MHz) toward these twelve sources would settle the claim: if the line is detected in absorption against an unresolved component whose flux density is smaller than the adopted core, or is not detected against the core, then the core-based covering factors and the factor-of-three correction are wrong. A cheaper first test is comparing the 1.4 GHz and 5 GHz VLBA core fractions of the same targets, since a systematic difference would show that the assumed core-to-total ratio is frequency dependent.

Watch

Extended reading notes

Core claim

Eleven of the twelve targets are resolved at VLBA resolution, with six classified as core-jet, four as two-sided jet, one as complex, and one unresolved. The paper tests, source by source, whether the core flux density or the total VLBA flux density is sufficient to produce the detected H I absorption by comparing the peak absorption fraction PF with Score/SRACS and Stot/SRACS. In seven targets the core alone is sufficient, supporting the picture that narrow-line absorbers (FW20 < 250 km/s) have most of their gas concentrated around the radio core on scales of a few hundred to a thousand parsecs. In the compact sources 0903+010, 0920+161, and 1002-195, the whole VLBA source is likely to be largely covered, and for 0903+010 at least 73% of the peak absorption must arise against VLBA-detected emission, giving a lower limit on the VOD of about 104 km/s. The highest upper limit, about 169 km/s, is found for 0023+010. After applying the lower limits on covering factors, the median ratio of corrected to uncorrected VOD for the ten usable targets is 2.8, so the paper concludes that the VOD distribution at 0.4<z<1.0 could increase by up to a factor of three, with the caveat that the sample is only twelve sources.

Load-bearing premise

The method assumes that the compact core makes up the same fraction of the total radio emission at 1.4 GHz, where the VLBA images were taken, as it does at the redshifted H I 21-cm frequency near 700-1000 MHz, where the absorption was measured; a frequency-dependent difference between core and extended emission would bias every covering factor and corrected optical depth.

Editorial extensions

If this is right

  • Corrected velocity-integrated optical depths for the ten usable targets are a median factor of 2.8 above the ASKAP-measured values, so the H I 21-cm VOD distribution at 0.4<z<1.0 could shift upward by roughly a factor of three.
  • Sources with projected sizes below about 400 pc (0903+010, 0920+161, and 1002-195) are the ones where the entire VLBA source is likely covered, so compact young or peaked-spectrum sources are the best targets for high column-density H I.
  • Two targets reach VOD limits above 100 km/s: 0023+010 has an upper limit of about 169 km/s and 0903+010 a lower limit of about 104 km/s, implying H I column densities of about 1.9e22 and 2.8e22 cm^-2 at a spin temperature of 100 K.
  • Because the cores are sufficient to cause the full absorption in seven of twelve narrow-line absorbers, these sources are the natural targets for spectroscopic VLBI follow-up at the redshifted H I frequency to measure true optical depths directly.
  • Literature VODs measured without VLBI resolution likely underestimate absorption strengths, so studies of how H I 21-cm absorption strength evolves with redshift may need to be revisited once covering-factor corrections are applied.

Reading between the lines

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

  • Beyond the paper: if the factor-of-three correction holds in the full FLASH sample, published redshift trends of declining H I 21-cm absorption strength, which have not been corrected for covering factors, may need to be re-examined, and the inferred evolution of gas availability with cosmic time could change.
  • Beyond the paper: the continuum-ratio method offers a low-cost route to statistically correct existing single-dish absorption surveys, since only a compact-array continuum image per source is needed to place limits on covering factors without requiring VLBI line observations.
  • Beyond the paper: the planned 5 GHz VLBA observations can directly test the frequency stability of the core fraction; if peaked-spectrum sources show a different core fraction at 5 GHz than at 1.4 GHz, a spectral-index correction could be folded into the covering-factor estimates.
  • Beyond the paper: for the compact sources with sizes below a few hundred parsecs, a single roughly 100 pc H I cloud could cover the entire radio source, meaning their absorption may be near saturation and their H I column densities, not just their optical depths, could be underestimated.
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

3 major / 6 minor

Summary. The paper presents VLBA 1.4 GHz continuum observations of twelve H I 21-cm absorbers detected in the FLASH pilot surveys at redshifts 0.4 < z < 0.7. The authors classify the parsec-scale source morphologies, identify the radio core in each source, and develop a methodology to test whether the core alone, or the total emission detected in the VLBA image, has sufficient flux density to cause the entire H I absorption measured by ASKAP. Using the ratio of VLBA to RACS flux densities as a lower limit on the gas covering factor, they derive upper limits on the velocity-integrated optical depth (VOD). They report that for seven of twelve sources the core is sufficient to explain the absorption, that three compact sources likely have most of their VLBA-scale emission occulted, and that the distribution of H I VODs at 0.4 < z < 1.0 could increase by up to a factor of three after covering-factor correction. The paper also highlights two high-VOD candidates, 0023+010 and 0903+010.

Significance. If the central claims hold, this work provides a practical method for correcting H I 21-cm absorption measurements for covering factors at intermediate redshifts, where spectroscopic VLBI at the redshifted line frequency is not currently feasible. The methodology is clearly described, the per-source tables are internally consistent, and the arithmetic in Appendix 1 is correct. The paper makes good use of public astrogeo images and is transparent about its assumptions and small sample size. The conclusion that the VOD distribution may be underestimated by a median factor of roughly 2.8 is important for evolutionary studies of cold gas, and the identification of two absorbers with VOD limits above 100 km/s is of significant interest. However, the central quantitative results depend on comparing flux ratios at 1.4 GHz with absorption fractions at ~0.7 GHz, and this frequency mismatch is not quantified. Because the sample is dominated by peaked-spectrum sources, this is a load-bearing issue that affects the 'seven of twelve' claim and the factor-three correction.

major comments (3)
  1. [Appendix 1, Eq. (9) and §3.2.2] The test Smin/SRACS > PF compares the 1.4 GHz VLBA-to-RACS flux ratio with the ~0.7 GHz ASKAP absorption peak fraction. This is valid only if the core-to-total flux ratio is the same at both frequencies. Writing R(ν) for the core fraction and S ∝ ν^α for each component, R(ν21)/R(1.4 GHz) = (ν21/1.4 GHz)^(α_c − α_t). For a self-absorbed core (α_c > 0) with optically thin extended emission (α_t < 0), this ratio can be about 0.4–0.5, so the tabulated Score/SRACS values can overestimate the covering factor at the line frequency. In this sample, nine of twelve sources have peaked SEDs (§4.1), and borderline cases such as 2236-251 (Score/SRACS = 0.33 vs PF = 0.32) and 1136+004 (0.27 vs 0.18) could change classification if the bias is at this level. The paper acknowledges in Appendix 1 that a direct comparison cannot be made, but it does not quantify the resulting systematic error; this is load-bearing for both the 'seven of twelve' claim and the median factor-2.8 VOD correction.
  2. [§3.3 and §4.1] The source 2007-245 is included among the seven sources for which 'the core has sufficient flux density' (§4.1), but in §3.3 the authors deliberately adopt Smin = Stot instead of Score because the absorber may be intervening. The methodology in §3.2.2 defines core sufficiency by the condition PF < Score/SRACS, which is satisfied for this source, yet the adopted covering-factor limit is based on Stot. Similarly, 0023+010 is excluded from the seven in §4.1 even though §3.3 states that its core is sufficient based on Score/Stot. The counting rule and the treatment of these two exceptions should be stated explicitly so that the reader can verify the 'seven of twelve' claim.
  3. [§2 and Table 3] The VLBA observations were made in February–April 2024, while the RACS-mid and FLASH/ASKAP measurements are from different epochs. For 0023+010 the authors note that Stot exceeds SRACS and interpret this as variability, but no variability check is presented for the remaining sources. Because the inequality tests in §3.2.2 are sensitive to the flux ratios (e.g., 2236-251, where Score/SRACS = 0.33 and PF = 0.32 are nearly equal), variability at the ten to twenty percent level, which is common in compact AGN, could change the classification of individual sources and hence the upper-limit VODs and the median correction factor.
minor comments (6)
  1. [Table 1] Table 1 lists source '2236-351' while all other tables and the text use '2236-251'; the coordinates match 2236-251, so this appears to be a typo.
  2. [Figure 3 / Table 2] The Figure 3 caption gives zHI = 0.5159 for 0023+010, while Table 2 lists z = 0.6745; one of these is incorrect and should be corrected.
  3. [§3.3, 0023+010] The text states Smin/Stot = 0.36, but Table 3 reports Smin/SRACS = 0.36; since the authors use Stot as the total flux for this source, the table entry should be labeled consistently with the text.
  4. [§4.3] The text includes 'NHI > 1022 cm–1', but the unit should be cm^-2 for a column density; please correct this typographical error.
  5. [Throughout] The text contains formatting artifacts such as 'Y oon et al.' and 'V ol.' in the references; these should be fixed to 'Yoon et al.' and 'Vol.'.
  6. [§3.4] Since PF is defined as the peak absorption fraction, it would be helpful to restate explicitly that the inequality test in §3.2.2 applies to the line peak, while the covering-factor correction is applied per channel when computing the VOD upper limit.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the VLBA continuum fluxes and the ASKAP absorption spectra are independent data sets, and the covering-factor test is a conditional comparison rather than a fitted prediction.

full rationale

The paper's central derivation compares two independently measured quantities: the parsec-scale VLBA continuum flux density ratios (Score/SRACS, Stot/SRACS) and the ASKAP absorption peak fraction (PF = |SHI|/SASKAP). Neither quantity is fitted from the other, and the VOD upper limits are obtained by applying the VLBI-based covering-factor lower limits to the ASKAP optical depths through the standard relation tau_true = -ln(1 + (e^-tau_app - 1)/f). This is a correction of published values, not a prediction derived from the same data it is meant to test. The self-citations to Yoon et al. 2024, Su et al. 2022, and related FLASH papers supply the input spectra and source classifications, but those are external observational data rather than a load-bearing argument that reduces the conclusion to a prior claim by the same authors. The acknowledged frequency mismatch between the 1.4 GHz VLBA/RACS ratios and the roughly 700 MHz HI line (Appendix 1, Eq. 9, and the note that 'a direct comparison cannot be made') is a real systematic-uncertainty concern about spectral-index differences between compact and extended emission, but it is not circularity: it concerns the validity of an assumption, not a definitional equivalence or a fitted parameter renamed as a prediction. The paper is self-contained in the sense that its classifications and corrected VOD limits follow from stated measurements and stated assumptions, with the main risk being bias from the frequency mismatch rather than circular reasoning.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No free parameters are fitted; the covering factors are measured ratios of public-archive flux densities. The central assumptions are the frequency-ratio equivalence and the fiducial spin temperature, neither of which is independently calibrated in this paper.

assumptions (3)
  • ad hoc to paper The VLBA-to-RACS flux ratio at 1.4 GHz equals the covering fraction at the redshifted HI frequency near 700-1000 MHz.
    The method compares Smin/SRACS (measured at 1.4 GHz) with PF (measured at about 700 MHz from ASKAP). This is valid only if the compact-to-total flux ratio is frequency-independent; the paper notes a direct comparison cannot be made but does not quantify the bias.
  • domain assumption Spin temperature Ts = 100 K for the HI gas.
    Used in Section 4.3 to convert VOD to HI column density via Eq. (2). The paper notes Ts could be up to an order of magnitude higher, so this is a fiducial value, not fitted.
  • domain assumption FW20 < 250 km/s defines narrow-line absorbers whose gas is likely concentrated near the radio core.
    Adopted from Maccagni et al. (2017) and used to justify using the core flux as a proxy for the absorbing region; all sample sources satisfy this criterion.

how reviews work

0 comments
Cite this review

Pith. "Pith review of VLBI studies of FLASH H I 21-cm absorbers -- I." pith.science (2026). https://pith.science/paper/3B2PVMYB

@misc{pith2026250719957,
  author       = {Pith},
  title        = {Pith review of: VLBI studies of FLASH H I 21-cm absorbers -- I},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3B2PVMYB}},
  note         = {Machine review of arXiv:2507.19957}
}
abstract

We have conducted VLBA 1.4 GHz (L-band) continuum observations towards twelve sources with HI 21-cm absorption detections at redshift $0.4<z<0.7$ in the pilot surveys of FLASH, an ongoing survey with the ASKAP radio telescope. 11 of the 12 targets are resolved in the VLBA observations. Using the parsec scale radio images, we have classified the source morphology and identified the radio core. Six of the twelve targets have core-jet morphology, four have two-sided jet morphology, one has a complex morphology, and one is unresolved. We describe a methodology to test whether the emission from the core or the total emission detected in the VLBA image has sufficient flux density to cause the entire HI 21-cm absorption, and we estimate limits on the gas covering factor and velocity-integrated optical depth (VOD). We find that for seven of the twelve sources, the core has sufficient flux density to cause all the HI 21-cm absorption detected in the ASKAP spectrum. For three other targets, with projected sizes in the range $\rm 305-409 \ pc$, a large fraction of the entire emission in the VLBA map could be occulted by the gas. For 0903+010 (NVSS J090331+010846), we estimate that at least $\approx 73\%$ of the peak absorption detected in the ASKAP spectrum could arise against the emission detected in the VLBA image. For the target 0023+010 (NVSS J002331+010114), we estimate an upper limit on the VOD of $\rm 169 \ km \ s^{-1}$, the highest in our sample. For 0903+010 (NVSS J090331+010846) we estimate a lower limit of $\rm 104 \ km \ s^{-1}$ on the VOD. We find that the distribution of HI 21-cm VODs at $0.4<z<1.0$ could increase by up to a factor of three after correction for the covering factors using our VLBA measurements.

Figures

Figures reproduced from arXiv: 2507.19957 by the authors.

Figure 1
Figure 1. Histogram showing the ratios of integrated VLBA and RACS flux densities. The vertical red line shows the median ratio. (see e.g. de Vries et al., 2009): 1. Unresolved or barely resolved (U). 2. Compact Double (CD). The source has two components with similar flux density. 3. Core-Jet (CJ). A source with two or more components with significantly different flux densities, where the compo￾nent with the highest peak flux… view at source ↗
Figure 2
Figure 2. 0011-023: On left, the VLBA L-band continuum image. Contours are at the levels 0.6 mJy × (1, 2, 4, 8..), where 0.6 mJy is the three times the r.m.s noise (σ) on the continuum image, as listed in [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. 0023+010: Details of the top left and top right images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figures from the paper (12 more)
Figure 4
Figure 4. Figure 4: 0141-231: Details of the top left and top right images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]
Figure 5
Figure 5. Figure 5: 0518-245: Details of the top left and top right images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: 0903+010: Details of the images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]
Figure 7
Figure 7. Figure 7: 0920+161: Details of the images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: 1002-195: Details of the images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]
Figure 9
Figure 9. Figure 9: 1136+004: Details of the images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p017_9.png]
Figure 10
Figure 10. Figure 10: 1701-294: Details of the images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p018_10.png]
Figure 11
Figure 11. Figure 11: 2007-245: Details of the top left and top right images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p019_11.png]
Figure 12
Figure 12. Figure 12: 2233-015: Details of the images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p020_12.png]
Figure 13
Figure 13. Figure 13: 2236-251: Details of the images are same as that of [PITH_FULL_IMAGE:figures/full_fig_p020_13.png]
Figure 14
Figure 14. Figure 14: Left panel: the estimated limits on the covering factor plotted as a function of the spatial extent of the radio source at the H I redshift. The three sources marked in Red represent 0903+010, 0920+161 and 1002-195, for which we estimate that a large fraction of the e…
Figure 15
Figure 15. Figure 15: Optical depth distribution of H I 21-cm absorbers at 0.4 < z < 1.0. Red bars represent the estimated limits to the H I 21-cm velocity integrated optical depth, of the FLASH detections. The black triangles are AGN-associated H I 21-cm absorbers in the literature, while…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 41 canonical work pages

  1. [1]

    Aditya, J. N. H. S. 2019, MNRAS, 482, 5597

    work page 2019

  2. [2]

    Aditya, J. N. H. S., & Kanekar, N. 2018, MNRAS, 481, 1578

    work page 2018

  3. [3]

    Aditya, J. N. H. S., Kanekar, N., & Kurapati, S. 2016, MNRAS, 455, 4000

    work page 2016

  4. [4]

    Aditya, J. N. H. S., Kanekar, N., Prochaska, J. X., et al. 2017, MNRAS, 465, 5011

    work page 2017

  5. [5]

    Aditya, J. N. H. S., Y oon, H., Allison, J. R., et al. 2024, MNRAS, 527, 8511

    work page 2024

  6. [6]

    R., Sadler, E

    Allison, J. R., Sadler, E. M., Amaral, A. D., et al. 2022, PASA, 39, e010

    2022

  7. [7]

    D., Rodríguez, C., Pihlström, Y., et al

    Araya, E. D., Rodríguez, C., Pihlström, Y., et al. 2010, AJ, 139, 17

    work page 2010

  8. [8]

    H., & W olfe, A

    Briggs, F. H., & W olfe, A. M. 1983, ApJ, 268, 76

    work page 1983

Show all 48 references
  1. [9]

    L., & Roberts, M

    Brown, R. L., & Roberts, M. S. 1973, ApJ, 184, L7

    1973

  2. [10]

    L., Rupen, M

    Carilli, C. L., Rupen, M. P., & Yanny, B. 1993, ApJ, 412, L59

    1993

  3. [11]

    N., de Bruyn, A

    Chengalur, J. N., de Bruyn, A. G., & Narasimha, D. 1999, A&A, 343, L79

    1999

  4. [12]

    Curran, S. J. 2012, ApJ, 748, L18 —. 2024, PASA, 41, e007

    2012

  5. [13]

    J., Allison, J

    Curran, S. J., Allison, J. R., Glowacki, M., Whiting, M. T., & Sadler, E. M. 2013, MNRAS, 431, 3408

    2013

  6. [14]

    J., Hunstead, R

    Curran, S. J., Hunstead, R. W., Johnston, H. M., et al. 2019, MNRAS, 484, 1182 —. 2017, MNRAS, 470, 4600

    2019

  7. [15]

    J., Whiting, M

    Curran, S. J., Whiting, M. T., Murphy, M. T., et al. 2006, MNRAS, 371, 431

    2006

  8. [16]

    J., Whiting, M

    Curran, S. J., Whiting, M. T., Wiklind, T., et al. 2008, MNRAS, 391, 765

    2008

  9. [17]

    2004, ApJ, 612, 58 de Vries, N., Snellen, I

    Darling, J. 2004, ApJ, 612, 58 de Vries, N., Snellen, I. A. G., Schilizzi, R. T., Mack, K.-H., & Kaiser, C. R. 2009, A&A, 498, 641

    2004

  10. [18]

    J., Lang, D., et al

    Dey, A., Schlegel, D. J., Lang, D., et al. 2019, AJ, 157, 168

    2019

  11. [19]

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

    2011

  12. [20]

    W., Thomson, A

    Duchesne, S. W., Thomson, A. J. M., Pritchard, J., et al. 2023, PASA, 40, e034

    2023

  13. [21]

    F., Baum, S

    Gallimore, J. F., Baum, S. A., O’Dea, C. P., Pedlar, A., & Brinks, E. 1999, ApJ, 524, 684

    1999

  14. [22]

    K., & Stocke, J

    Grasha, K., Darling, J., Bolatto, A., Leroy, A. K., & Stocke, J. T. 2019, ApJS, 245, 3

    2019

  15. [23]

    Greisen, E. W. 2003, in Astrophysics and Space Science Library, V ol. 285, Information Handling in Astronomy - Historical Vistas, ed. A. Heck, 109

    2003

  16. [24]

    J., Saikia, D

    Gupta, N., Salter, C. J., Saikia, D. J., Ghosh, T., & Jeyakumar, S. 2006, MN- RAS, 373, 972

    2006

  17. [25]

    Kanekar, N., & Chengalur, J. N. 2003, A&A, 399, 857

    2003

  18. [26]

    M., Momjian, E., Briggs, F

    Kanekar, N., Lane, W. M., Momjian, E., Briggs, F. H., & Chengalur, J. N. 2009, MNRAS, 394, L61

    2009

  19. [27]

    X., Smette, A., et al

    Kanekar, N., Prochaska, J. X., Smette, A., et al. 2014, MNRAS, 438, 2131

    2014

  20. [28]

    F., Allison, J

    Kerrison, E. F., Allison, J. R., Moss, V. A., Sadler, E. M., & Rees, G. A. 2024, MNRAS, 533, 4248

    2024

  21. [29]

    M., & Briggs, F

    Lane, W. M., & Briggs, F. H. 2001, ApJ, 561, L27

    2001

  22. [30]

    M., Morganti, R., Oosterloo, T

    Maccagni, F. M., Morganti, R., Oosterloo, T. A., Geréb, K., & Maddox, N. 2017, A&A, 604, A43

    2017

  23. [31]

    2014, ARA&A, 52, 415

    Madau, P., & Dickinson, M. 2014, ARA&A, 52, 415

    2014

  24. [32]

    R., Hollenbach, D

    Maloney, P. R., Hollenbach, D. J., & Tielens, A. G. G. M. 1996, ApJ, 466, 561

    1996

  25. [33]

    2013, Sci- ence, 341, 1082

    Morganti, R., Fogasy, J., Paragi, Z., Oosterloo, T., & Orienti, M. 2013, Sci- ence, 341, 1082

    2013

  26. [34]

    2018, A&A Rev., 26, 4

    Morganti, R., & Oosterloo, T. 2018, A&A Rev., 26, 4

    2018

  27. [35]

    N., & Oosterloo, T

    Morganti, R., Tadhunter, C. N., & Oosterloo, T. A. 2005, A&A, 444, L9

    2005

  28. [36]

    2023, A&A, 678, A42

    Morganti, R., Murthy, S., Oosterloo, T., et al. 2023, A&A, 678, A42

    2023

  29. [37]

    A., Allison, J

    Moss, V. A., Allison, J. R., Sadler, E. M., et al. 2017, MNRAS, 471, 2952

    2017

  30. [38]

    2022, arXiv e-prints, arXiv:2201.06625

    Murthy, S., Morganti, R., Kanekar, N., & Oosterloo, T. 2022, arXiv e-prints, arXiv:2201.06625

    2022 arXiv

  31. [39]

    Murthy, S., Morganti, R., Oosterloo, T., & Maccagni, F. M. 2021, A&A, 654, A94

    2021

  32. [40]

    2024, A&A, 688, A84

    Murthy, S., Morganti, R., Oosterloo, T., Schulz, R., & Paragi, Z. 2024, A&A, 688, A84

    2024

  33. [41]

    P., Radcliffe, J

    Njeri, A., Deane, R. P., Radcliffe, J. F., et al. 2024, MNRAS, 528, 6141 O’Dea, C. P., & Saikia, D. J. 2021, A&A Rev., 29, 3

    2024

  34. [42]

    2006, A&A, 457, 531

    Orienti, M., Morganti, R., & Dallacasa, D. 2006, A&A, 457, 531

    2006

  35. [43]

    Y., & Kovalev, Y

    Petrov, L. Y., & Kovalev, Y. Y. 2025, ApJS, 276, 51

    2025

  36. [44]

    J., Saikia, D

    Salter, C. J., Saikia, D. J., Minchin, R., Ghosh, T., & Chandola, Y. 2010, ApJ, 715, L117

    2010

  37. [45]

    2021, A&A, 647, A63

    Schulz, R., Morganti, R., Nyland, K., et al. 2021, A&A, 647, A63

    2021

  38. [46]

    M., Allison, J

    Su, R., Sadler, E. M., Allison, J. R., et al. 2022, MNRAS, 516, 2947 van Gorkom, J. H., Knapp, G. R., Raimond, E., Faber, S. M., & Gallagher, J. S. 1986, AJ, 91, 791

    2022

  39. [47]

    C., Pihlström, Y

    Vermeulen, R. C., Pihlström, Y. M., Tschager, W., et al. 2003, A&A, 404, 861

    2003

  40. [48]

    T., Darling, J., et al

    Yan, T., Stocke, J. T., Darling, J., et al. 2016, AJ, 151, 74 Y oon, H., Sadler, E. M., Mahony, E. K., et al. 2024, arXiv e-prints, arXiv:2408.06626

    2016 arXiv

Pith tools

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