Pith. sign in

REVIEW 3 major objections 5 minor 62 references

The local interstellar medium cycles gas between warm and cold phases every few million years, with almost half of its midplane neutral mass caught in a thermally unstable state.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

T0 review · deepseek-v4-flash

2026-08-01 23:35 UTC pith:YWPM73L6

load-bearing objection Genuine first 3D thermal-phase map of the local ISM, with honest caveats; headline numbers are resolution-dependent but the qualitative cycling picture holds. the 3 major comments →

arxiv 2607.15352 v1 pith:YWPM73L6 submitted 2026-07-16 astro-ph.GA

The first three-dimensional map of thermal phases in the local interstellar medium

classification astro-ph.GA
keywords interstellar mediumthermal phasesthermal instabilitycold neutral mediumwarm neutral mediumthree-dimensional dust extinctionFUV radiation fieldstar formation density PDF
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper builds P3D, a three-dimensional map of thermal phase in the gas within about 500 pc of the Sun, on a 2 pc grid. It assigns every cell to one of three categories—warm, cold, or thermally unstable—by combining a 3D dust-based density map with a newly computed 3D ultraviolet radiation field and a steady-state thermochemical model. The central claim is that the local neutral medium is not a quiet two-phase gas: about 41% of the dust-traced neutral mass in the |z|≤150 pc midplane layer is thermally unstable, meaning gas is continuously cycled between warm and cold states on timescales of roughly 3–6 million years. The map also finds that cold clouds are wrapped in coherent unstable envelopes, and that the cold phase internally is only subsonic to transonic (Mach number ≤1.5), contradicting the strongly supersonic cold-gas assumption used in many star-formation models. If correct, the paper reframes where and how interstellar gas assembles into the cold clouds that form stars.

Core claim

The paper's central discovery is the three-dimensional architecture and statistical weight of thermal phases in the local ISM. On the paper's own terms, the reconstruction shows cold neutral clouds surrounded by thermally unstable envelopes, all embedded in a pervasive warm neutral medium. Averaged over |z|≤150 pc, the dust-traced neutral mass fractions are cold ≈0.27, unstable ≈0.41, warm ≈0.32; the unstable category is the single largest component near the midplane and carries about half the mass at z=0. The instability fraction implies a dynamical cycling time t_dyn≈2.9 Myr, or up to ≈6 Myr if resolution effects are conservatively accounted for, still comparable to the thermal relaxation

What carries the argument

The central object is P3D, a 3D map of thermal phase built on a 2 pc Cartesian grid spanning a 1 kpc-diameter volume centered on the Sun. Its inputs are a parsec-scale 3D dust extinction map (converted to gas density) and a new 3D far-ultraviolet radiation field that treats individual O and B stars as point sources with radiative transfer through the dust. The classifying mechanism is the S-shaped steady-state pressure curve of a neutral-ISM thermochemical model: for a local FUV intensity, the curve has two turning points, n_W,max and n_C,min, marking the ends of the stable warm and cold branches. A cell is classified warm below n_W,max, cold above n_C,min, and unstable in between. The unsta

Load-bearing premise

The whole classification assumes that the 2-pc-sampled, dust-derived density field represents the gas density that actually controls heating and cooling in each cell; unresolved sub-parsec cold structures would bias the unstable fraction upward and the cold-phase density width downward.

What would settle it

A sub-parsec-resolution 3D map of the same local volume—from deeper dust extinction or 21 cm absorption tomography—could count cold mass below the current 2 pc resolution; if the midplane cold fraction rises sharply and the unstable fraction falls well below ~0.2, the 41% figure is largely a resolution artifact. Conversely, velocity-resolved maps of individual cold clouds showing internal Mach numbers above ~2 would falsify the transonic cold-phase claim.

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

If this is right

  • Near the midplane, 41% of the dust-traced neutral mass is thermally unstable, so the local ISM must be cycled by dynamics on ~2.9–6 Myr timescales, comparable to thermal relaxation.
  • Cold clouds are not isolated; they are wrapped in coherent unstable envelopes, giving multiphase simulations a concrete 3D benchmark for the solar neighbourhood.
  • The cold-phase density PDF is narrow (σ_s=0.485, internal Mach number 0.5–1.5); star-formation models that assume a single, strongly supersonic cold-gas density field should be reassessed.
  • With increasing height above the midplane the gas becomes warmer, and the warm phase dominates beyond |z|≈150 pc, so 21 cm projected fractions are warm-weighted relative to this local slab.
  • Apparent high Mach numbers from 21 cm CNM linewidths may be dominated by bulk motions of cold structures along the sightline rather than internal turbulence.

Where Pith is reading between the lines

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

  • If the unstable envelopes are real and long-lived, they may be the main sites where H2 and CO become detectable before gas fully cools into dense clouds—an implicit chemical consequence the paper does not develop.
  • The paper's method could be applied to other nearby regions or to time-dependent thermochemical models; a natural extension is to compare P3D's phase boundaries directly with velocity-resolved HI absorption maps to test whether unstable envelopes have the predicted kinematics.
  • Because the unstable fraction and cold-phase width both depend on resolution, the strongest test is a sub-parsec 3D dust map; until then, the 41% figure should be read as an upper limit on the true unstable fraction and the 0.485 width as a lower limit on the true cold-phase density variance.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper presents P3D, a three-dimensional reconstruction of the thermal phase of neutral gas in the local ISM within ~1 kpc of the Sun, sampled on a 2 pc Cartesian grid. It combines the Edenhofer et al. (2024) 3D dust extinction map with a new 3D FUV radiation field (Porter et al., in preparation) and the Bialy & Sternberg (2019) thermochemical model. Each cell is classified by its position in the (n_H, I_UV) plane relative to the thermal-stability boundaries n_W,max and n_C,min, yielding cold, warm, and unstable (intermediate-density) categories. The main quantitative claims are: (i) within |z| <= 150 pc, the unstable category carries ~41% of the dust-traced neutral mass, implying a dynamical cycling time of ~2.9-6 Myr; (ii) the volume-weighted cold-phase density PDF has logarithmic width sigma_s = 0.485, corresponding to internal sonic Mach numbers M_s = 0.5-1.5, i.e. sub- to transonic; and (iii) the spatial architecture consists of cold clouds surrounded by coherent unstable envelopes embedded in warm gas. The paper also compares the cold and unstable fractions with 21 cm-based phase studies and discusses implications for star-formation models that assume a strongly supersonic cold-gas density field.

Significance. If correct, P3D would provide a genuinely novel volumetric view of the multiphase ISM, complementing long-established line-of-sight 21 cm studies. The explicit internal robustness checks across thermochemical models and dust maps (typically delta f ~0.04-0.07) are a strength, as is the planned public release of the data cube. The cold-phase density PDF measurement is particularly noteworthy because it directly challenges the conventional assumption of strongly supersonic cold-gas turbulence. However, the two headline quantitative results — the 41% unstable fraction and the sigma_s = 0.485 cold-PDF width — both rest on the assumption that the dust-based density field faithfully represents gas density at the scales controlling thermal balance. The paper itself acknowledges that unresolved sub-parsec substructure could bias f_U high and sigma_s low, but it does not quantify this bias. Because the transonic-Mach-number claim is explicitly advertised as a central finding and would motivate revising star-formation models, this unresolved systematic is load-bearing. The work is significant and promising, but the advertised precision of the central claims is not yet fully secured.

major comments (3)
  1. [Methods, 'Volumetric density PDF and Mach number' and 'Unresolved sub-parsec structure'] The cold-phase PDF width sigma_s = 0.485 (Fig. 3b) and the derived M_s <= 1.5 are measured from a dust-based density field that is spatially smoothed by the distance-dependent resolution of the E24 map. The paper explicitly states that the result 'should not be interpreted as excluding additional substructure below the effective resolution.' That caveat directly undermines the headline claim, because unresolved cold substructure would broaden the true density PDF relative to the reconstructed one and lower the inferred Mach number. No quantitative estimate or upper bound is provided. To secure the transonic conclusion, I would need a forward-model test: take a multiphase simulation with known sub-parsec structure, degrade it to the 2 pc and distance-dependent resolution of the E24 map, and recompute sigma_s; alternatively, provide an explicit upper limit on sigma_s from the unresolved ma
  2. [Methods, 'Resolution and the unstable mass fraction'] The 41% unstable mass fraction (Results, Fig. 2a, Table 1) is similarly sensitive to resolution: smoothing cold substructure into the intermediate-density interval biases f_U high. The three arguments given (f_U(z) stability to 600-800 pc, coherent envelopes, and simulation-based statements about cold-cloud sizes) do not quantify the bias. The stability of f_U with analysis-volume cuts is not a direct test of sub-parsec smoothing, because the dust map resolution is itself distance-dependent and the cuts do not isolate unresolved structure. The coherent-envelope argument is suggestive but not quantitative. Thus the central value f_U = 0.41 is not measured at the stated precision. I acknowledge the cycling-time conclusion is more robust: even f_U = 0.24 gives t_dyn ~6 Myr, still of order t_cool. However, the abstract and Results present 41% as a headline, so this systematic needs a quantit
  3. [Methods, 'Three-dimensional FUV radiation field'] The FUV radiation field is a critical input to the phase classification, yet the model used is described as 'in prep' (P25) and was developed by a co-author. The manuscript does not provide enough detail to independently reproduce or test this model: the matching of Hipparcos and Gaia samples, the treatment of ambiguous spectral types, and the transition between the E24 and F98 dust models are described only briefly. A referee cannot verify that the FUV field is not a dominant source of systematic error. I am not claiming circularity — the phase fractions are outputs of a classification applied to independent dust and FUV maps — but the unpublished status and co-authorship of the FUV model make the central results difficult to evaluate. At minimum, the paper should include a validation against observed FUV constraints or provide a detailed public description of the model before the centr
minor comments (5)
  1. [Abstract] The phrase 'FUV intensity in 24 units' in the full text appears to be garbled; likely 'in Habing units' or similar. Please correct.
  2. [References] Reference [16] is missing its title and has an incomplete author list (arXiv93(2019)); several other arXiv identifiers are nonstandard. Please format consistently.
  3. [Figure 2a caption] The caption says 'Averaged over |z| ≤150 pc, the three categories carry comparable mass,' but then gives ⟨f_C⟩≈0.27, ⟨f_U⟩≈0.41, ⟨f_W⟩≈0.32. These are not 'comparable' in the colloquial sense; consider rephrasing to 'each carries a substantial fraction' or similar.
  4. [Data and code availability] The data and code are promised to be available 'upon publication' or 'upon reasonable request.' For a paper with quantitative central claims, consider depositing the data cube and analysis code in a public repository before acceptance to facilitate independent verification.
  5. [Methods, 'Phase classification'] The unstable category is defined geometrically as all gas between n_W,max and n_C,min, which includes both the classical thermally unstable branch and off-equilibrium gas. This is clearly explained, but the repeated shorthand 'thermally unstable' in the abstract and Results may mislead readers; consider consistently using 'thermally unstable category' or 'intermediate-density gas' when precision matters.

Circularity Check

0 steps flagged

No circularity: the phase map and its statistics are forward-model outputs from independent 3D dust and FUV maps, not fitted to the reported phase fractions.

full rationale

P3D's derivation chain is: Edenhofer et al. dust density map and a new 3D FUV field (P25) are passed through the BS19 thermochemical model to assign each cell a thermal-stability category; the reported quantities (f_U ≈ 0.41, sigma_s = 0.485, cycling time) are then computed from the classified cells via Eqs. (5)–(8). None of these target values is a fitted parameter—they are outputs of a classification applied to independent data products. The BS19 model, although co-authored by Bialy, is a published, parameter-free forward model with stated assumptions that do not include the target phase fractions; the paper additionally cross-checks with Wolfire-type models and the Leike dust map. The P25 FUV model is self-cited and in preparation, but it is described as a radiative-transfer forward model, not as an ansatz fitted to the phase results. The paper's own caveats about unresolved sub-parsec substructure—that it could bias f_U high or narrow the cold-phase PDF, and that sigma_s 'should not be interpreted as excluding additional substructure below the effective resolution'—are resolution limitations, explicitly quantified for f_U (0.24–0.41) and shown not to change the order-of-magnitude cycling time. These are honest limitations, not circular reductions of the prediction to the input. No uniqueness theorem or loaded self-citation is invoked to force the conclusion, and the central claim is not definitionally equivalent to any fitted input.

Axiom & Free-Parameter Ledger

2 free parameters · 5 axioms · 0 invented entities

The central reconstruction rests on three prior components—E24 dust map, the unpublished P25 FUV model, and the BS19 thermochemical model—plus assumptions that each cell is in thermal steady state and unresolved structure does not change conclusions. The dust-to-gas scale factor and adopted cooling time are parameters the quantitative claims inherit.

free parameters (2)
  • dust-to-gas scale factor = 2.04 (constant 1727 cm^-3 per mag pc^-1)
    Eq. (2) converts E24 differential extinction to gas density; adopted from Draine et al. cross-section plus scaling. A change shifts all densities and phase classifications.
  • adopted cooling time t_cool = ≈2 Myr (range ~1.4–3.3 Myr from literature)
    Used with Eq. (8) to convert f_U into the cycling time t_dyn≈2.9–6 Myr; the claimed timescale scales linearly with this adopted value.
axioms (5)
  • domain assumption BS19 thermochemical model correctly computes steady-state temperature, phase boundaries, and cooling times for the neutral ISM at Z'=1, zeta=1e-16 s^-1.
    Invoked in Methods 'Thermochemical model' and 'Phase classification'; model is authored by co-author Bialy and is not re-validated in this paper.
  • domain assumption Edenhofer et al. (2024) 3D dust map faithfully recovers gas density at ~parsec resolution; line-of-sight averaging and regularization do not dominate the phase classification.
    Used throughout as the density input; authors argue smoothing biases f_U only within bounds but cannot fully quantify it.
  • ad hoc to paper The 3D FUV field (P25, in prep) correctly resolves O/B star contributions and dust attenuation.
    Constructed in parallel by co-author Porter for this project; central to I_UV and hence to phase boundaries, but not publicly described or validated.
  • domain assumption Each dust-map cell is in or classifiable by thermal steady state; off-equilibrium gas is assigned by density interval rather than modeled self-consistently.
    Methods 'Phase classification': the unstable category includes off-equilibrium gas, but off-equilibrium gas at stable densities is assumed negligible.
  • domain assumption The density-variance relation Eq. (1) applies to the cold-phase PDF in a multiphase medium when M_s is the internal velocity dispersion.
    Methods 'Volumetric density PDF and Mach number'; supported only by multiphase simulations (Kobayashi et al.), with residual 10–20% uncertainty noted.

pith-pipeline@v1.3.0-alltime-deepseek · 13616 in / 12117 out tokens · 126478 ms · 2026-08-01T23:35:45.261355+00:00 · methodology

0 comments
read the original abstract

The thermal state of interstellar gas controls whether it remains warm and diffuse or cools into dense clouds. Twenty-one centimetre observations have established a statistical picture of neutral-gas phases, but their three-dimensional architecture has not previously been mapped. Here we present $\mathcal{P}_{\rm 3D}$, a three-dimensional reconstruction of thermal phase in the local interstellar medium (ISM), covering a 1 kpc-diameter region centred on the Sun and sampled on a 2 pc Cartesian grid. $\mathcal{P}_{\rm 3D}$ combines high-resolution 3D maps of dust extinction and interstellar far-ultraviolet radiation with a neutral-ISM thermochemical model. The reconstruction reveals cold clouds surrounded by thermally unstable envelopes and embedded in a pervasive warm phase. Within $|z| \leq 150$ pc of the Galactic midplane, $\sim$41 % of the dust-traced neutral mass is thermally unstable, implying phase cycling on a timescale of $\sim$2.9-6 Myr. Yet the cold phase has a narrow density distribution, consistent with internal Mach numbers that are transonic at most ($\mathcal{M}_s \lesssim 1.5$). Together, these findings favour a dynamically cycling multiphase interstellar medium and motivate reassessing star-formation models that assume a single-phase, strongly supersonic cold-gas density field.

discussion (0)

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

Reference graph

Works this paper leans on

62 extracted references · 13 linked inside Pith

  1. [1]

    B., Goldsmith, D

    Field, G. B., Goldsmith, D. W. & Habing, H. J.The Astrophysical Journal155, L149 (1969)

  2. [2]

    & Bakes, E.The Astrophysical Journal443, 152–168 (1995)

    Wolfire, M., Hollenbach, D., McKee, C., Tielens, A. & Bakes, E.The Astrophysical Journal443, 152–168 (1995)

  3. [3]

    G., McKee, C

    Wolfire, M. G., McKee, C. F., Hollenbach, D. & Tielens, A. G. G. M.The Astrophysical Journal587, 278–311 (2003)

  4. [4]

    & Sternberg, A.The Astrophysical Journal 881, 160 (2019)

    Bialy, S. & Sternberg, A.The Astrophysical Journal 881, 160 (2019). ArXiv: 1902.06764

  5. [5]

    B.The Astrophysical Journal142, 531 (1965)

    Field, G. B.The Astrophysical Journal142, 531 (1965)

  6. [6]

    & Troland, T

    Heiles, C. & Troland, T. H.The Astrophysical Journal 586, 1067–1093 (2003)

  7. [7]

    Murray, C. E.et al. The Astrophysical Journal Supple- ment Series238, 14 (2018). ArXiv: 1806.06065

  8. [8]

    & Hennebelle, P.Astronomy & Astrophysics 433, 1–13 (2005)

    Audit, E. & Hennebelle, P.Astronomy & Astrophysics 433, 1–13 (2005)

  9. [9]

    & Schmidt, W.Astronomy & Astrophysics 567, A16 (2014)

    Saury, E., Miville-Desch ˆenes, M.-A., Hennebelle, P., Audit, E. & Schmidt, W.Astronomy & Astrophysics 567, A16 (2014)

  10. [10]

    & Miville-Desch ˆenes, M.-A.The Astro- physical Journal908, 186 (2021)

    Marchal, A. & Miville-Desch ˆenes, M.-A.The Astro- physical Journal908, 186 (2021)

  11. [11]

    & Kim, J.The Astrophysical Journal, Volume 765, Issue 1, article id

    Gazol, A. & Kim, J.The Astrophysical Journal, Volume 765, Issue 1, article id. 49, 8 pp. (2013).765(2013). ArXiv: 1301.4280

  12. [12]

    M., Terzian, Y

    Dickey, J. M., Terzian, Y. & Salpeter, E. E.The Astro- physical Journal Supplement Series36, 77 (1978)

  13. [13]

    M., McClure-Griffiths, N

    Dickey, J. M., McClure-Griffiths, N. M., Gaensler, B. M. & Green, A. J.The Astrophysical Journal585, 801–822 (2003). ArXiv: astro-ph/0211298

  14. [14]

    E., Peek, J

    Murray, C. E., Peek, J. E. G. & Kim, C.-G.The Astro- physical Journal899, 15 (2020)

  15. [15]

    Astronomy & Astrophysics626, A101 (2019)

    Marchal, A.et al. Astronomy & Astrophysics626, A101 (2019)

  16. [16]

    M., Schlafly, E., Zucker, C., Speagle, J

    Green, G. M., Schlafly, E., Zucker, C., Speagle, J. S. & Finkbeiner, D.arXiv93(2019). ArXiv: 1905.02734

  17. [17]

    Astronomy & Astrophysics625, A135 (2019)

    Lallement, R.et al. Astronomy & Astrophysics625, A135 (2019)

  18. [18]

    & Enßlin, T

    Leike, R., Glatzle, M. & Enßlin, T. A.Astronomy & Astrophysics138, 1–13 (2020). ArXiv: 2004.06732

  19. [19]

    Astronomy & Astrophysics685, A82 (2024)

    Edenhofer, G.et al. Astronomy & Astrophysics685, A82 (2024)

  20. [20]

    Nature578, 237–239 (2020)

    Alves, J.et al. Nature578, 237–239 (2020). ADS Bibcode: 2020Natur.578..237A

  21. [21]

    The Astrophysical Journal Letters919, L5 (2021)

    Bialy, S.et al. The Astrophysical Journal Letters919, L5 (2021). ArXiv: 2109.09763

  22. [22]

    Nature601, 334 (2022)

    Zucker, C.et al. Nature601, 334 (2022). ArXiv: 2201.05124

  23. [23]

    Nature Astronomy(2025)

    Burkhart, B.et al. Nature Astronomy(2025)

  24. [24]

    T.The Astrophysical Journal Supplement Series36, 595 (1978)

    Draine, B. T.The Astrophysical Journal Supplement Series36, 595 (1978)

  25. [25]

    M.Reviews of Modern Physics73, 1031– 1066 (2001)

    Ferri `ere, K. M.Reviews of Modern Physics73, 1031– 1066 (2001)

  26. [26]

    T.Physics of the interstellar and intergalactic medium(2011)

    Draine, B. T.Physics of the interstellar and intergalactic medium(2011). Publication Title: Physics of the Inter- stellar and Intergalactic Medium (Princeton University Press)

  27. [27]

    ArXiv: 1910.12953

    Bialy, S.Nature Communications Physics3, 32 (2020). ArXiv: 1910.12953

  28. [28]

    & P´erault, M.Astronomy & Astrophysics 351, 309–322 (1999)

    Hennebelle, P. & P´erault, M.Astronomy & Astrophysics 351, 309–322 (1999)

  29. [29]

    J., Vazquez-Semadeni, E

    Sanchez-Salcedo, F. J., Vazquez-Semadeni, E. & Gazol, A.The Astrophysical Journal577, 768–788 (2002)

  30. [30]

    & Villagran, M

    Gazol, A. & Villagran, M. A.Monthly Notices of the Royal Astronomical Society462, 2033–2045 (2016)

  31. [31]

    Krumholz, M. R. & McKee, C. F.The Astrophysical Journal630, 250–268 (2005)

  32. [32]

    McKee, C. F. & Ostriker, E. C.Annual Review of As- tronomy and Astrophysics45, 565–687 (2007). ArXiv: 0707.3514

  33. [33]

    & Nordlund, ˚A.The Astrophysical Journal 730, 40 (2011)

    Padoan, P. & Nordlund, ˚A.The Astrophysical Journal 730, 40 (2011)

  34. [34]

    & Klessen, R

    Federrath, C. & Klessen, R. S.The Astrophysical Jour- nal761, 156 (2012)

  35. [35]

    Burkhart, B.The Astrophysical Journal863, 118 (2018)

  36. [36]

    & Jones, B

    Padoan, P., Nordlund, A. & Jones, B. J. T.Monthly Notices of the Royal Astronomical Society288, 145–152 (1997)

  37. [37]

    & Mac Low, M

    Federrath, C., Roman-Duval, J., Klessen, R., Schmidt, W. & Mac Low, M. M.Astronomy and Astrophysics, Volume 512, id.A81, 28 pp.512, A81 (2010). ArXiv: 0905.1060

  38. [38]

    M., Stanimirovi´c, S

    McClure-Griffiths, N. M., Stanimirovi´c, S. & Rybarczyk, D. R.Annual Review of Astronomy and Astrophysics61, 19–63 (2023)

  39. [39]

    Kalberla, P. M. & Kerp, J.Annual Review of Astronomy and Astrophysics47, 27–61 (2009). 6

  40. [40]

    Soler, J. D.et al. Astronomy & Astrophysics662, A96 (2022)

  41. [41]

    Kobayashi, M. I. N., Inoue, T., Tomida, K., Iwasaki, K. & Nakatsugawa, H.The Astrophysical Journal930, 76 (2022)

  42. [42]

    Zhang, X., Green, G. M. & Rix, H.-W.Monthly No- tices of the Royal Astronomical Society524, 1855–1884 (2023)

  43. [43]

    Skrutskie, M. F.et al. The Astronomical Journal131, 1163–1183 (2006)

  44. [44]

    Wright, E. L.et al. The Astronomical Journal140, 1868–1881 (2010)

  45. [45]

    F., Meisner, A

    Schlafly, E. F., Meisner, A. M. & Green, G. M.The As- trophysical Journal Supplement Series240, 30 (2019)

  46. [46]

    The Astrophysical Journal Supplement Series259, 51 (2022)

    Wang, C.et al. The Astrophysical Journal Supplement Series259, 51 (2022)

  47. [47]

    Astronomy and Astrophysics662, A66 (2022)

    Xiang, M.et al. Astronomy and Astrophysics662, A66 (2022). ADS Bibcode: 2022A&A...662A..66X

  48. [48]

    Draine, B. T.et al. The Astrophysical Journal663, 866– 894 (2007)

  49. [49]

    J., Zucker, C., Goodman, A

    O’Neill, T. J., Zucker, C., Goodman, A. A. & Edenhofer, G.The Astrophysical Journal973, 136 (2024)

  50. [50]

    A., J ´ohannesson, G

    Porter, T. A., J ´ohannesson, G. & Moskalenko, I. V.The Astrophysical Journal846, 67 (2017)

  51. [51]

    A., Strigari, L

    Porter, T. A., Strigari, L. E. & Moskalenko, I. V. (Pro- ceedings of Science, Geneva, Switzerland, 2025)

  52. [52]

    Van Leeuwen, F.Astronomy & Astrophysics474, 653– 664 (2007)

  53. [53]

    Astronomy & Astrophysics674, A39 (2023)

    Gaia Collaborationet al. Astronomy & Astrophysics674, A39 (2023)

  54. [54]

    Robitaille, T. P.et al. Astronomy and Astrophysics545, A39 (2012). ADS Bibcode: 2012A&A...545A..39R

  55. [55]

    & Kuriliene, G.Astrophysics and Space Science80, 353–368 (1981)

    Straizys, V. & Kuriliene, G.Astrophysics and Space Science80, 353–368 (1981). ADS Bibcode: 1981Ap&SS..80..353S

  56. [56]

    N.Allen’s astrophysical quantities(2000)

    Cox, A. N.Allen’s astrophysical quantities(2000). Publication Title: Allen’s Astrophysical Quantities ADS Bibcode: 2000asqu.book.....C

  57. [57]

    Godard, B., Des For ˆets, G. P. & Bialy, S.Astronomy & Astrophysics688, A169 (2024)

  58. [58]

    Kobayashi, M. I. N.et al. The Astrophysical Journal 954, 38 (2023)

  59. [59]

    Astronomy & Astrophysics643, A36 (2020)

    Bellomi, E.et al. Astronomy & Astrophysics643, A36 (2020). ArXiv: 2009.05466

  60. [60]

    Dickey, J. M.et al. The Astrophysical Journal693, 1250–1260 (2009)

  61. [61]

    Young OB

    Kalberla, P. M. & Haud, U.Astronomy and Astrophysics 619, A58 (2018). ArXiv: 1806.04085. Acknowledgments We thank Benjamin Godard for a thorough reading of the manuscript and for detailed comments that significantly improved the paper. S.B. acknowledges support from the ISF grant number 2071540, the GIF grant number I-1568- 303.7/2024, the NSF-BSF grant n...

  62. [62]

    Temperature en- ters only as an output of the thermochemical model

    We classify each volume cell into one of three categories: Warm:𝑛<𝑛 𝑊,max(𝐼 UV),(5) Unstable:𝑛 𝑊,max(𝐼 UV)<𝑛<𝑛 𝐶,min(𝐼 UV),(6) Cold:𝑛>𝑛 𝐶,min(𝐼 UV).(7) This definition is geometric in the(𝑛,𝐼 UV)plane and does not require an observed kinetic temperature. Temperature en- ters only as an output of the thermochemical model. For cells in the warm or cold cate...