Pith. sign in

REVIEW 3 major objections 4 minor 123 references

Modeling UV Radiation Feedback from Massive Stars: III. Escape of Radiation from Star-forming Giant Molecular Clouds

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Turbulence, not mean dust, controls how much UV light escapes star-forming clouds.

desk verdict Solid simulation paper with a new optical-depth PDF framework; the headline accuracy claims are in-sample and absolute escape fractions rest on a flagged subgrid assumption, so treat the quantitative range as indicative rather than final. read the letter →

arxiv 1908.07549 v1 pith:5J6GS2VM submitted 2019-08-20 astro-ph.GA

Jeong-Gyu Kim , Woong-Tae Kim , Eve C. Ostriker This is my paper · ORCID
classification astro-ph.GA
keywords escapefractionUVradiationfeedbackgiantmolecularcloudshydrodynamicsopticaldepthdistributionturbulenceHIIregionsdustabsorption
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 argues that the fraction of ultraviolet radiation escaping a giant molecular cloud is controlled not by the cloud's mean optical depth but by the spread of optical depths around that mean, shaped by turbulence. Using radiation-hydrodynamic simulations of 14 model clouds spanning two orders of magnitude in mass and surface density, it finds cumulative escape fractions of 5–58% for ionizing radiation and 7–72% for non-ionizing radiation before the first supernovae. The central mechanism is that turbulence and feedback carve low-column-density holes, so that escape along the clearest sightlines far exceeds the naive estimate $e^{-\langle \tau\rangle}$. Because both ionizing and non-ionizing photons escape through the same fully ionized, low-density channels, their escape fractions are nearly equal at late times, with dust rather than hydrogen absorbing most of the radiation that does not escape. The paper also proposes two observationally usable estimators, based on the mean dust optical depth or its projected distribution, that recover the simulated non-ionizing escape fraction to within about 20%.

What carries the argument

The central object is the optical-depth probability distribution function, $P(\ln\tau)$, measured either as a solid-angle-weighted distribution from the source position or as an area-weighted distribution projected on the sky. The escape fraction for non-ionizing radiation is the solid-angle average $\langle e^{-\tau}\rangle_\Omega$, which by Jensen's inequality is always at least $e^{-\langle\tau\rangle}$; a lognormal fit with measured mean and width reproduces the true escape fractions within 7% (non-ionizing) and 20% (ionizing). For observers, the paper introduces a reduction factor $F = -\ln(f_{\rm esc})/\langle \tau\rangle$ and two estimators: $f_{\rm esc,n} = \exp(-\eta_1 \langle \tau_{\rm ext}\rangle_A)$ with $\eta_1 = 0.56/(1 + 1.25\langle \tau_{\rm ext}\rangle_A^{0.48})$ for marginally resolved clouds, and $f_{\rm esc,n} = \langle \exp(-0.3\, \tau_{\rm ext})\rangle_A$ for resolved clouds.

What would settle it

Measure the projected dust optical depth distribution around a young embedded cluster in a resolved giant molecular cloud (for example, via near-infrared extinction) and compare the predicted $f_{\rm esc,n} = \langle \exp(-0.3\, \tau_{\rm ext})\rangle_A$ with the escape fraction inferred by comparing the cluster's Hα or free-free luminosity to the ionizing photon rate expected from its stellar content; a systematic discrepancy larger than about 20% would falsify the calibration.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the escape fraction of both ionizing and non-ionizing radiation from a star-forming GMC is largely set by the solid-angle distribution of dust optical depth as seen from the stellar luminosity center: escape increases with smaller mean $\langle \tau\rangle$ and larger dispersion $\sigma$, and exceeds the one-zone estimate $e^{-\langle \tau\rangle}$ by up to three orders of magnitude because turbulence produces a wide lognormal-like tail of low-opacity sightlines. The simulations show that once H II regions break out, photons escape through the same fully ionized low-density channels, so $f_{\rm esc,n} \approx f_{\rm esc,i}$ at late times even though the two frequency bins have very different physical opacities; dust becomes the dominant absorber of ionizing photons. As a practical corollary, the escape fraction can be recovered from externally projected dust optical depth using two simple calibrations, accurate to about 20% in the models.

Load-bearing premise

The simulations assume that every photon emitted by a star particle leaves the unresolved control volume around it without any absorption ($f_{\rm esc,*} = 1$); if the accreting gas right around the stars absorbs a significant part of the ultraviolet radiation, all quoted cloud-scale escape fractions and the calibration of the estimators are systematically too high.

Editorial extensions

If this is right

  • Star formation rates in individual clouds inferred from dust-corrected Hα or ultraviolet emission will be underestimated unless the escaped photon fraction is added back; the paper's estimators provide a correction from the same dust maps used for extinction.
  • The similarity of $f_{\rm esc,i}$ and $f_{\rm esc,n}$ at late times implies that diffuse ionized gas on galactic scales can be powered by Lyman-continuum leakage from GMCs without requiring a separate population of exceptionally leaky H II regions.
  • The cumulative escape fraction before the first supernova ranges from a few percent to roughly 50%, with compact clouds of high surface density, which are destroyed within 3 Myr, contributing the highest fractions; massive $10^6\,M_\odot$ clouds leak very little before supernovae.
  • If dust is destroyed in ionized gas, the ionizing cumulative escape fraction for the fiducial cloud rises by about 0.2, bounding the role of grain destruction in boosting photon escape.

Reading between the lines

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

  • A testable consequence not drawn in the paper: the same two estimators should work on observed column-density PDFs of Galactic clouds, provided the dust opacity per hydrogen is known; the scatter in the paper's Figure 12 predicts that the direct area-averaged estimator ($\eta = 0.3$) will outperform the mean-based one for clouds with resolved column-density maps.
  • The reduction-factor framework suggests a dimensionless diagnostic: for any star-forming cloud, measuring $\tau_{\rm ext}$ along many sightlines and plotting $F$ versus $\langle \tau\rangle$ should collapse onto the simulation band if turbulence and feedback are the main sources of opacity spread.
  • Extending beyond the paper's pre-supernova horizon, the steep drop in ionizing photon production after 3 Myr implies that supernovae may clear remaining gas but not add many escaping Lyman-continuum photons, shifting the epoch of maximum leakage earlier in the cloud lifetime.
  • The subgrid escape assumption $f_{\rm esc,*} = 1$ is the paper's stated caveat; if accreting flows near sink particles absorb photons, the quoted cumulative escape fractions are upper limits, and the calibration constants $\eta_1$ and $\eta_2$ would need recalibration toward smaller escape.
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

3 major / 4 minor

Summary. This paper presents radiation-hydrodynamic simulations of star cluster formation in turbulent giant molecular clouds, covering two orders of magnitude in cloud mass and surface density, and analyzes the escape fractions of ionizing (LyC) and non-ionizing (FUV) radiation. The authors find that escape fractions increase with time as H II regions break out, that cumulative escape fractions before the first supernova range from 5% to 58% for ionizing and 7% to 72% for non-ionizing radiation, and that the escape fraction is largely determined by the mean and width of the optical depth distribution rather than the naive exp(-<tau>) estimate. They propose two observationally oriented estimators for fesc,n based on the area-averaged dust optical depth or its resolved distribution, claiming accuracy within about 20%.

Significance. If the results hold, this is a valuable systematic numerical study: it spans a wider parameter space than most prior work, treats both ionizing and non-ionizing radiation consistently, and provides a physically transparent reduction of the escape problem to optical-depth PDF moments. The two proposed estimators are potentially useful for converting observed dust column maps into escape-fraction corrections for star formation indicators. The paper is also honest in its treatment of caveats: Appendices A and B explicitly address subgrid escape, dust destruction, and resolution effects, and the authors state in Section 5.2.5 that the cloud-scale escape fraction may be overestimated because of the fesc,* = 1 assumption. These features make the work a substantive contribution to the theory of UV feedback and photon escape from star-forming clouds.

major comments (3)
  1. [Section 2.2 and Section 5.2.5] The assumption fesc,* = 1, namely that all photons emitted by a sink particle emerge from the 3^3-cell control volume without subgrid absorption, is load-bearing for the two central quantitative claims: the cumulative escape fraction range (Table 1, Columns 9 and 12) and the calibration of the estimators in Equations (12) and (13). The authors explicitly concede in Section 5.2.5 that the cloud-scale escape fraction may be an overestimate due to this assumption. However, the mitigation offered in Appendix A does not directly address the problem: the experiments in Section A.3 vary fesc,* only in radiation-pressure-only models and report changes in the net star formation efficiency (Figure 13), not changes in fesc,n or fesc,i in the photoionization-dominated models that set the headline results. The statement in Section 5.2.1 that effects on fesc are 'relatively modest' is therefore not supported by the presented tests. This is a specific, fixable gap: the authors should either run the fesc,* variation for photoionization-dominated clouds and report the resulting escape fractions, or visibly weaken the quantitative claims about the absolute escape fraction range.
  2. [Section 4.2.1, Equations (12) and (13)] The stated accuracy of the two estimators ('within ~20%') is measured on the same simulation snapshots used to fit the parameters a = 1.25, b = 0.48, and eta2 = 0.30. This is an in-sample fit, so the 20% figure is not a predictive accuracy; it is a measure of the fit's residuals. For the claim that these methods can estimate observed escape fractions, the manuscript should provide some out-of-sample assessment (e.g., cross-validation across models or across time, or a held-out subset of the snapshot data). As written, a reader cannot distinguish the estimator's intrinsic scatter from the flexibility of the adopted functional form.
  3. [Section 2.2 and Section 4.1] The use of a single luminosity center to characterize the optical-depth PDF (Section 4.1) is systematically inaccurate in the early embedded phase when sources are clustered in a few widely separated regions; the authors acknowledge this in the text. This is not a fatal flaw because the quantitative escape fractions themselves are directly measured from the ray tracing, but it should be kept in mind when interpreting the PDF-based estimators: the claimed agreement between the luminosity-center reconstruction and the true escape fraction is demonstrated only for the later, more centrally concentrated phases. A quantitative statement of the early-phase discrepancy (how much larger the luminosity-center estimate is) would strengthen the paper.
minor comments (4)
  1. [Section 3.1] There is a typo: 'Galctic H II regions' should be 'Galactic H II regions'.
  2. [Appendix B, Figure 16 caption] The caption of Figure 16 lists 'Dest-i' and 'Dest-i/n' but the text in Section 5.2.4 refers to 'complete destruction' cases; consider defining the abbreviations more explicitly at first use.
  3. [Figure 15 caption] The caption appears to have a typo: it lists sigma_c = 1.0, 2.0, 2.0, 2.5, while the text discusses sigma_c = 1.0, 1.5, 2.0, 2.5.
  4. [Section 5.2.5] The phrase 'could further increase' appears in the discussion of stellar winds and outflows; consider 'could also increase' to avoid the implication that the previously discussed subgrid absorption already increases porosity.

Circularity Check

1 steps flagged · score 6.0 of 10

The claimed ~20% estimator accuracy is in-sample calibration: the constants in Equations (12) and (13) are least-squares fitted to the same simulation snapshots on which the 'prediction' is then evaluated.

  1. fitted input called prediction [Section 4.2.1, Equations (12) and (13), Figures 10-12]
    "We perform a least square fit to find parameters a = 1.25, b = 0.48 that minimize the sum of squared errors ((fesc,n - fest,1esc,n)^2) compared to our simulation results. ... This estimator predicts fesc,n within ~20%. ... We also adopt the constant value of eta2 for all snapshots and find that eta2 = 0.30 minimizes the sum of the square of the differences (fest,2esc,n - fesc,n)^2. Figure 12(b) compares fest,2esc,n with the actual escape fraction, again showing that this method predicts fesc,n within 20%."

    The parameters a, b, and eta2 that define estimators (12) and (13) are obtained by least-squares minimization against the full set of simulation snapshots from all models, and the claimed 'predicts within ~20%' accuracy is then demonstrated by plotting those same estimators against those same snapshots in Figure 12. Thus the quoted 20% is the in-sample residual of a fit, not an out-of-sample prediction. By construction, the fitted curve minimizes the very squared-error statistic that is used to support the accuracy claim, so the predictive statement reduces to the calibration procedure unless an independent test set, cross-validation, or external data were used, which the paper does not provide.

full rationale

The central physical derivation is not circular: the escape fractions fesc,i and fesc,n are directly measured from the adaptive ray-tracing simulations, the optical-depth PDFs are computed independently from the density fields, and the inequality in Equation (9) is a general mathematical statement. The finding that fesc increases with smaller mean optical depth and larger dispersion is a genuine analysis of simulation data, not an input assumption. The lognormal approximation is also tested against the raw PDFs rather than assumed to be identical to them. The main circularity is confined to the estimator section: the paper fits the free constants of Equations (12) and (13) to all simulation snapshots and then calls the resulting agreement a prediction within 20%. This is in-sample by construction. The paper's self-citations to Paper I and Paper II are methodological provenance for the code and simulation suite, not load-bearing citations that smuggle in the escape-fraction result. The limitation flagged in Section 5.2.5 that the subgrid assumption fesc,* = 1 may make the cloud-scale escape fractions overestimates is an acknowledged modeling caveat, not a circular step; similarly, Appendix A tests subgrid escape effects on SFE rather than on fesc, which is a support gap but not circularity.

Assumptions & free parameters 3 free parameters · 7 assumptions · 0 invented entities

The central claim rests on modeling assumptions about the thermodynamics of ionized gas, the constant dust cross section, and the unimpeded escape of radiation from sink control volumes. The estimators themselves introduce two fitted parameters (a, b) and one constant (eta2) that are calibrated on the same simulations. No new physical entities are postulated.

free parameters (3)
  • eta1 normalization parameter a = 1.25
    Fitted by least squares to all simulation snapshots in Equation (12) to minimize squared error in fesc,n.
  • eta1 exponent parameter b = 0.48
    Fitted by least squares along with a in Equation (12).
  • eta2 correction factor = 0.30
    Constant chosen to minimize squared error between fesc,n and <exp(-eta2 tau_ext_n)>_A across all snapshots, Section 4.2.1.
assumptions (7)
  • domain assumption The gas temperature is fixed at 20 K when fully neutral and 8000 K when fully ionized, with smooth interpolation based on neutral fraction; cooling time is assumed short.
    Section 2.1. This ignores helium ionization and non-equilibrium cooling, affecting H II region pressure and dynamics and therefore the escape fraction evolution.
  • domain assumption Photons emitted by sink particles escape their 3^3-cell control volume without absorption (fesc,* = 1).
    Section 2.2. Could overestimate cloud-scale escape fractions if accreting gas absorbs radiation inside the unresolved control volume.
  • domain assumption Constant dust absorption cross section per hydrogen, sigma_d = 1.17e-21 cm^2, applies to both ionizing and non-ionizing radiation; dust scattering is neglected.
    Section 2.1. Appendix B tests dust destruction, but the baseline analysis uses a constant cross section and no scattering.
  • standard math The solid-angle optical-depth distribution P(ln tau) can be approximated as lognormal to relate the mean and width to the escape fraction.
    Section 4.1, Equation (10). This heuristic is tested against raw PDFs, but the lognormal shape is assumed, not derived.
  • domain assumption The initial cloud is a uniform-density sphere with virial parameter alpha_vir,0 = 2 and a decaying turbulent velocity field, with no magnetic fields or external pressure.
    Section 2.2-2.3. These initial conditions shape the density structure and dynamics, and may not cover all GMC environments.
  • domain assumption Stellar UV luminosity and ionizing photon rate scale with cluster mass via median IMF fits, without time evolution during the 3 Myr period.
    Section 2.1. Neglects stellar evolution and IMF stochasticity beyond the median relation.
  • standard math Jensen's inequality for the exponential of optical depth: <exp(-tau)> >= exp(-<tau>).
    Equation (9) in Section 4.1. Used to establish that a broad optical-depth PDF boosts the escape fraction above the mean-based estimate.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Modeling UV Radiation Feedback from Massive Stars: III. Escape of Radiation from Star-forming Giant Molecular Clouds." pith.science (2026). https://pith.science/paper/5J6GS2VM

@misc{pith2026190807549,
  author       = {Pith},
  title        = {Pith review of: Modeling UV Radiation Feedback from Massive Stars: III. Escape of Radiation from Star-forming Giant Molecular Clouds},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5J6GS2VM}},
  note         = {Machine review of arXiv:1908.07549}
}
abstract

Using a suite of radiation hydrodynamic simulations of star cluster formation in turbulent clouds, we study the escape fraction of ionizing (Lyman continuum) and non-ionizing (FUV) radiation for a wide range of cloud masses and sizes. The escape fraction increases as H II regions evolve and reaches unity within a few dynamical times. The cumulative escape fraction before the onset of the first supernova explosion is in the range 0.05-0.58; this is lower for higher initial cloud surface density, and higher for less massive and more compact clouds due to rapid destruction. Once H II regions break out of their local environment, both ionizing and non-ionizing photons escape from clouds through fully ionized, low-density sightlines. Consequently, dust becomes the dominant absorber of ionizing radiation at late times and the escape fraction of non-ionizing radiation is only slightly larger than that of ionizing radiation. The escape fraction is determined primarily by the mean $\langle \tau\rangle$ and width $\sigma$ of the optical-depth distribution in the large-scale cloud, increasing for smaller $\langle \tau\rangle$ and/or larger $\sigma$. The escape fraction exceeds (sometimes by three orders of magnitude) the naive estimate $e^{-\langle \tau\rangle}$ due to non-zero $\sigma$ induced by turbulence. We present two simple methods to estimate, within $\sim20\%$, the escape fraction of non-ionizing radiation using the observed dust optical depth in clouds projected on the plane of sky. We discuss implications of our results for observations, including inference of star formation rates in individual molecular clouds, and accounting for diffuse ionized gas on galactic scales.

Figures

Figures reproduced from arXiv: 1908.07549 by the authors.

Figure 1
Figure 1. Snapshots of the fiducial model M1E5R20 (M0 = 105 M and R0 = 20 pc) at 0.5, 1.5, 3.0, and 5.0 Myr (left to right) after the first star formation. (top row) Gas surface density projected along the y-direction. In each panel, the projected positions of star particles are indicated by small circles, with age indicated by color. The star particle center of mass is indicated with a star symbol. The plus signs and the dot… view at source ↗
Figure 2
Figure 2. Same as [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Evolution of the instantaneous escape fraction fesc,i (top), the hydrogen absorption fraction fgas,i (middle), and the dust absorption fraction fdust,i (bottom) for ionizing radiation. Time is measured from the creation of the first star particle (t 0 = t − t∗,0), in units of tff,0 (left) or Myr (right). All models are shown, with the thickness and color of each line indicating the initial cloud mass M0 and surface … view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Dependence of (a) the instantaneous escape fraction fesc,i, (b) the hydrogen absorption fraction fgas,i, and (c) the dust absorption fraction fdust,i of ionizing radiation on the product of the total photoionization rate Qgas,i and the rms number density of the ionized…
Figure 5
Figure 5. Figure 5: Evolution of the instantaneous escape fraction of ionizing (fesc,i, orange) and non-ionizing (fesc,n, blue) radiation for selected models whose mass and radius are specified in each panel. The dotted and dashed lines in black draw the escape fraction of ionizing radiat…
Figure 6
Figure 6. Figure 6: Cumulative escape fractions of ionizing (orange) and non-ionizing (blue) radiation up to time t 0 = 3 Myr after the first star formation, plotted against the initial cloud surface density Σ0. Although the cumulative escape fraction tends to decrease with in￾creasing Σ0…
Figure 7
Figure 7. Figure 7: (a) Escape fraction of non-ionizing radiation as a func￾tion of the solid-angle averaged optical depth hτ c iΩ seen from the source. With solid lines we show expectations based on log￾normal distributions of the optical depth, with standard deviation σ c = 0, 0.5, · · …
Figure 8
Figure 8. Figure 8: (Top) Solid-angle-weighted PDFs of the optical depth for non-ionizing (τ c d, blue) and ionizing (τ c i , orange) radiation measured from the stellar center of luminosity for the fiducial model at 0.5, 1.5, 3, and 5 Myr after the first epoch of star formation. (Bottom)…
Figure 9
Figure 9. Figure 9: Comparison of the true escape fraction as seen from the cluster center with an estimated escape fraction for (a) non-ionizing and (b) ionizing radiation, for all simulation snapshots. The true escape fraction he −τ c iΩ on the ordinate is calculated from Equa￾tion (9) …
Figure 10
Figure 10. Figure 10: Instantaneous escape fraction fesc,n of non-ionizing ra￾diation against the area-averaged dust optical depth hτ ext n iA within the half-mass radius for all models. The color of each dot indicates the initial surface density of the cloud. The horizontal bars repre￾sen…
Figure 12
Figure 12. Figure 12: Actual escape fraction fesc,n of non-ionizing radiation vs. estimated escape fraction (a) based on the area-averaged optical depth f est,1 esc,n = exp(−η1hτ ext n iA) with η1 = 0.56/(1 + 1.25hτ ext n i 0.48 A ) and (b) the area-averaged escape fraction f est,2 esc,n =…
Figure 14
Figure 14. Figure 14: Cloud-scale reduction factor F for non-ionizing radia￾tion as measured from the stellar center of luminosity as a function of hτ c niΩ. The circles are from our simulations, with colors corre￾sponding to σ c . The lines draw the reduction factor expected for lognormal…
Figure 15
Figure 15. Figure 15: 2D histograms of the grid-scale escape fraction fesc,∗ and the mass accretion rate M˙ ∗ onto sink particles for the fiducial cloud model, under the assumption that the optical-depth PDF at subgrid scales follows a lognormal distribution with the mean hln τsiΩ = ln(hτs…
Figure 16
Figure 16. Figure 16: Effect of dust destruction in ionized gas for the fiducial cloud with M0 = 105 M and R0 = 20 pc. Time evolution of (a) total gas mass in the simulation domain (black), stellar mass (green), ejected gas mass (red), and photoevaporated gas mass (yellow); (b) instantaneo…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

123 extracted references · 67 canonical work pages

  1. [1]

    bڴi ؃ > 磺mmmq뭷ƬY

    thebibliography [1] 20pt to REFERENCES 6pt =0pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key on reference command Each re...

    2019

  2. [2]

    D.\ 2002, , 330, L53

    Abel, T., & Wandelt, B. D.\ 2002, , 330, L53

    2002

  3. [3]

    F.\ 1992, , 395, 140

    Bertoldi, F., & McKee, C. F.\ 1992, , 395, 140

    1992

  4. [4]

    A., & Povich, M

    Binder, B. A., & Povich, M. S.\ 2018, , 864, 136

    2018

  5. [5]

    M., Leitherer, C., & Overzier, R

    Borthakur, S., Heckman, T. M., Leitherer, C., & Overzier, R. A.\ 2014, Science, 346, 216

    2014

  6. [6]

    J., Illingworth, G

    Bouwens, R. J., Illingworth, G. D., Oesch, P. A., et al.\ 2011, , 737, 90

    2011

  7. [7]

    Bromm, V., & Yoshida, N.\ 2011, , 49, 373

    2011

  8. [8]

    Burkhart, B., & Lazarian, A.\ 2012, , 755, L19

    2012

Show all 123 references
  1. [9]

    Chastenet, J., Sandstrom, K., Chiang, I., et al.\ 2019, arXiv:1904.02705

    2019 arXiv

  2. [10]

    Chastenet, J., Sandstrom, K., Chiang, I.-D., et al.\ 2019, , 876, 62

    2019

  3. [11]

    E.\ 2015, , 68, 1

    Dale, J. E.\ 2015, , 68, 1

    2015

  4. [12]

    E., Ercolano, B., & Bonnell, I

    Dale, J. E., Ercolano, B., & Bonnell, I. A.\ 2013, , 430, 234

    2013

  5. [13]

    E., Ercolano, B., & Bonnell, I

    Dale, J. E., Ercolano, B., & Bonnell, I. A.\ 2012, , 424, 377

    2012

  6. [14]

    S.\ 1994, , 428, 647

    Domgorgen, H., & Mathis, J. S.\ 1994, , 428, 647

    1994

  7. [15]

    A., Groves, B

    Dopita, M. A., Groves, B. A., Sutherland, R. S., & Kewley, L. J.\ 2003, , 583, 727

    2003

  8. [16]

    I., Crowther, P

    Doran, E. I., Crowther, P. A., de Koter, A., et al.\ 2013, , 558, A134

    2013

  9. [17]

    B., Shull, J

    Dove, J. B., Shull, J. M., & Ferrara, A.\ 2000, , 531, 846

    2000

  10. [18]

    T.\ 2003, , 41, 241

    Draine, B. T.\ 2003, , 41, 241

    2003

  11. [19]

    T.\ 2011, , 732, 100

    Draine, B. T.\ 2011, , 732, 100

    2011

  12. [20]

    T.\ 2011, Physics of the Interstellar and Intergalactic Medium by Bruce T

    Draine, B. T.\ 2011, Physics of the Interstellar and Intergalactic Medium by Bruce T. Draine. Princeton University Press, 2011. ISBN: 978-0-691-12214-4,

    2011

  13. [21]

    Ferguson, A. M. N., Wyse, R. F. G., Gallagher, J. S., III, & Hunter, D. A.\ 1996, , 111, 2265

    1996

  14. [22]

    L., D'Aloisio, A., Paardekooper, J.-P., et al.\ 2019, arXiv:1902.02792

    Finkelstein, S. L., D'Aloisio, A., Paardekooper, J.-P., et al.\ 2019, arXiv:1902.02792

    2019 arXiv

  15. [23]

    L., D'Aloisio, A., Paardekooper, J.-P., et al.\ 2019, , 879, 36

    Finkelstein, S. L., D'Aloisio, A., Paardekooper, J.-P., et al.\ 2019, , 879, 36

    2019

  16. [24]

    L., Papovich, C., Ryan, R

    Finkelstein, S. L., Papovich, C., Ryan, R. E., et al.\ 2012, , 758, 93

    2012

  17. [25]

    Geen, S., Rosdahl, J., Blaizot, J., Devriendt, J., & Slyz, A.\ 2015, , 448, 3248

    2015

  18. [26]

    D., & Hennebelle, P.\ 2017, , 471, 4844

    Geen, S., Soler, J. D., & Hennebelle, P.\ 2017, , 471, 4844

    2017

  19. [27]

    Glatzle, M., Ciardi, B., & Graziani, L.\ 2019, , 482, 321

    2019

  20. [28]

    M., Hivon, E., Banday, A

    G \'o rski, K. M., Hivon, E., Banday, A. J., et al.\ 2005, , 622, 759

    2005

  21. [29]

    C.\ 2013, , 204, 8

    Gong, H., & Ostriker, E. C.\ 2013, , 204, 8

    2013

  22. [30]

    A., Pineda, J

    Goodman, A. A., Pineda, J. E., & Schnee, S. L.\ 2009, , 692, 91

    2009

  23. [31]

    Y., Hopkins, P

    Grudi \'c , M. Y., Hopkins, P. F., Faucher-Gigu \`e re, C.-A., et al.\ 2018, , 475, 3511

    2018

  24. [32]

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

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

    2009

  25. [33]

    M., Borthakur, S., Overzier, R., et al.\ 2011, , 730, 5

    Heckman, T. M., Borthakur, S., Overzier, R., et al.\ 2011, , 730, 5

    2011

  26. [34]

    G., Kurtz, S

    Hoare, M. G., Kurtz, S. E., Lizano, S., Keto, E., & Hofner, P.\ 2007, Protostars and Planets V, 181

    2007

  27. [35]

    J., & Tielens, A

    Hollenbach, D. J., & Tielens, A. G. G. M.\ 1999, Reviews of Modern Physics, 71, 173

    1999

  28. [36]

    G., Walterbos, R

    Hoopes, C. G., Walterbos, R. A. M., & Greenwalt, B. E.\ 1996, , 112, 1429

    1996

  29. [37]

    F., & Grudi \'c , M

    Hopkins, P. F., & Grudi \'c , M. Y.\ 2019, , 483, 4187

    2019

  30. [38]

    D.\ 2007, CSE, 9, 90

    Hunter, J. D.\ 2007, CSE, 9, 90

    2007

  31. [39]

    Howard, C., Pudritz, R., & Klessen, R.\ 2017, , 834, 40

    2017

  32. [40]

    S., Pudritz, R

    Howard, C. S., Pudritz, R. E., Harris, W. E., & Klessen, R. S.\ 2018, , 475, 3121

    2018

  33. [41]

    Iffrig, O., & Hennebelle, P.\ 2015, , 576, A95

    2015

  34. [42]

    K.\ 2002, , 570, L97

    Inoue, A. K.\ 2002, , 570, L97

    2002

  35. [43]

    K., Hirashita, H., & Kamaya, H.\ 2001, , 555, 613

    Inoue, A. K., Hirashita, H., & Kamaya, H.\ 2001, , 555, 613

    2001

  36. [44]

    I., Schaerer, D., Thuan, T

    Izotov, Y. I., Schaerer, D., Thuan, T. X., et al.\ 2016, , 461, 3683

    2016

  37. [45]

    I., Schaerer, D., Worseck, G., et al.\ 2018, , 474, 4514

    Izotov, Y. I., Schaerer, D., Worseck, G., et al.\ 2018, , 474, 4514

    2018

  38. [46]

    Kainulainen, J., Beuther, H., Henning, T., & Plume, R.\ 2009, , 508, L35

    2009

  39. [47]

    Kakiichi, K., & Gronke, M.\ 2019, arXiv e-prints, arXiv:1905.02480

    2019 arXiv

  40. [48]

    C., & Evans, N

    Kennicutt, R. C., & Evans, N. J.\ 2012, , 50, 531

    2012

  41. [49]

    C., & Kim, W.-T.\ 2013, , 776, 1

    Kim, C.-G., Ostriker, E. C., & Kim, W.-T.\ 2013, , 776, 1

    2013

  42. [50]

    C., & Raileanu, R.\ 2017, , 834, 25

    Kim, C.-G., Ostriker, E. C., & Raileanu, R.\ 2017, , 834, 25

    2017

  43. [51]

    C.\ 2017, , 846, 133

    Kim, C.-G., & Ostriker, E. C.\ 2017, , 846, 133

    2017

  44. [52]

    C.\ 2016, , 819, 137

    Kim, J.-G., Kim, W.-T., & Ostriker, E. C.\ 2016, , 819, 137

    2016

  45. [53]

    C.\ 2018, , 859, 68 (Paper II)

    Kim, J.-G., Kim, W.-T., & Ostriker, E. C.\ 2018, , 859, 68 (Paper II)

    2018

  46. [54]

    C., & Skinner, M

    Kim, J.-G., Kim, W.-T., Ostriker, E. C., & Skinner, M. A.\ 2017, , 851, 93 (Paper I)

    2017

  47. [55]

    Kim, K.-T., & Koo, B.-C.\ 2003, , 596, 362

    2003

  48. [56]

    Kim, K.-T., & Koo, B.-C.\ 2001, , 549, 979

    2001

  49. [57]

    Kimm, T., & Cen, R.\ 2014, , 788, 121

    2014

  50. [58]

    Kimm, T., Katz, H., Haehnelt, M., et al.\ 2017, , 466, 4826

    2017

  51. [59]

    Kimm, T., Blaizot, J., Garel, T., et al.\ 2019, arXiv:1901.05990

    2019 arXiv

  52. [60]

    Kimm, T., Blaizot, J., Garel, T., et al.\ 2019, , 486, 2215

    2019

  53. [61]

    A., Schinnerer, E., et al.\ 2016, , 827, 103

    Kreckel, K., Blanc, G. A., Schinnerer, E., et al.\ 2016, , 827, 103

    2016

  54. [62]

    R., Stone, J

    Krumholz, M. R., Stone, J. M., & Gardiner, T. A.\ 2007, , 671, 518

    2007

  55. [63]

    R., Bate, M

    Krumholz, M. R., Bate, M. R., Arce, H. G., et al.\ 2014, Protostars and Planets VI, 243

    2014

  56. [64]

    R.\ 2018, , 480, 3468

    Krumholz, M. R.\ 2018, , 480, 3468

    2018

  57. [65]

    R., McKee, C

    Krumholz, M. R., McKee, C. F., & Bland-Hawthorn, J.\ 2018, arXiv e-prints , arXiv:1812.01615

    2018 arXiv

  58. [66]

    Lacerda, E. A. D., Cid Fernandes, R., Couto, G. S., et al.\ 2018, , 474, 3727

    2018

  59. [67]

    C.\ 2009, , 704, 1640

    Laursen, P., Sommer-Larsen, J., & Andersen, A. C.\ 2009, , 704, 1640

    2009

  60. [68]

    Lefloch, B., & Lazareff, B.\ 1994, , 289, 559

    1994

  61. [69]

    Leitet, E., Bergvall, N., Hayes, M., Linn \'e , S., & Zackrisson, E.\ 2013, , 553, A106

    2013

  62. [70]

    C., & Oey, M

    Leitherer, C., Hernandez, S., Lee, J. C., & Oey, M. S.\ 2016, , 823, 64

    2016

  63. [71]

    Loeb, A., & Barkana, R.\ 2001, , 39, 19

    2001

  64. [72]

    J.\ 2014, , 566, A45

    Lombardi, M., Bouy, H., Alves, J., & Lada, C. J.\ 2014, , 566, A45

    2014

  65. [73]

    A., Krumholz, M

    Lopez, L. A., Krumholz, M. R., Bolatto, A. D., et al.\ 2014, , 795, 121

    2014

  66. [74]

    F., et al.\ 2015, , 453, 960

    Ma, X., Kasen, D., Hopkins, P. F., et al.\ 2015, , 453, 960

    2015

  67. [75]

    S.\ 2000, , 544, 347

    Mathis, J. S.\ 2000, , 544, 347

    2000

  68. [76]

    C., et al.\ 2019, , 51, 535

    McCandliss, S., Calzetti, D., Ferguson, H. C., et al.\ 2019, , 51, 535

    2019

  69. [77]

    F., Dale, J

    McLeod, A. F., Dale, J. E., Evans, C. J., et al.\ 2019, , 486, 5263

    2019

  70. [78]

    F., & Ostriker, E

    McKee, C. F., & Ostriker, E. C.\ 2007, , 45, 565

    2007

  71. [79]

    9th Python in Science Conf., Data Structures for Statistical Computing in Python, ed

    McKinney , W.\ 2010, in Proc. 9th Python in Science Conf., Data Structures for Statistical Computing in Python, ed. S. van der Walt & J. Millman (Austin, TX: SciPy), 51

    2010

  72. [80]

    S., Meurer, G

    Oey, M. S., Meurer, G. R., Yelda, S., et al.\ 2007, , 661, 801

    2007

  73. [81]

    C., McKee, C

    Ostriker, E. C., McKee, C. F., & Leroy, A. K.\ 2010, , 721, 975

    2010

  74. [82]

    Paardekooper, J.-P., Khochfar, S., & Dalla Vecchia, C.\ 2015, , 451, 2544

    2015

  75. [83]

    J., & McKee, C

    Parravano, A., Hollenbach, D. J., & McKee, C. F.\ 2003, , 584, 797

    2003

  76. [84]

    W., Oey, M

    Pellegrini, E. W., Oey, M. S., Winkler, P. F., et al.\ 2012, , 755, 40

    2012

  77. [85]

    E.\ 2007, CSE, 9, 21

    P\'erez, F., & Granger, B. E.\ 2007, CSE, 9, 21

    2007

  78. [86]

    B.\ 1972, , 177, L69

    Petrosian, V., Silk, J., & Field, G. B.\ 1972, , 177, L69

    1972

  79. [87]

    J., Groves, B., et al.\ 2019, , 1197

    Poetrodjojo, H., D'Agostino, J. J., Groves, B., et al.\ 2019, , 1197

    2019

  80. [88]

    W., Glover, S

    Rahner, D., Pellegrini, E. W., Glover, S. C. O., et al.\ 2017, , 470, 4453

    2017

  81. [89]

    C., & Skinner, M

    Raskutti, S., Ostriker, E. C., & Skinner, M. A.\ 2016, , 829, 130

    2016

  82. [90]

    C., & Skinner, M

    Raskutti, S., Ostriker, E. C., & Skinner, M. A.\ 2017, , 850, 112

    2017

  83. [91]

    J.\ 1984, , 282, 191

    Reynolds, R. J.\ 1984, , 282, 191

    1984

  84. [92]

    R., Berg, D., Bordoloi, R., et al.\ 2019, arXiv e-prints , arXiv:1905.05566

    Rigby, J. R., Berg, D., Bordoloi, R., et al.\ 2019, arXiv e-prints , arXiv:1905.05566

    2019 arXiv

  85. [93]

    Rigby, J., Berg, D., Bordoloi, R., et al.\ 2019, , 51, 245

    2019

  86. [94]

    E., Dahle, H., Chisholm, J., et al.\ 2019, arXiv:1904.08186

    Rivera-Thorsen, T. E., Dahle, H., Chisholm, J., et al.\ 2019, arXiv:1904.08186

    2019 arXiv

  87. [95]

    E., Ellis, R

    Robertson, B. E., Ellis, R. S., Dunlop, J. S., McLure, R. J., & Stark, D. P.\ 2010, , 468, 49

    2010

  88. [96]

    E., Ellis, R

    Robertson, B. E., Ellis, R. S., Furlanetto, S. R., & Dunlop, J. S.\ 2015, , 802, L19

    2015

  89. [97]

    M.\ 2013, , 431, 1337

    Rogers, H., & Pittard, J. M.\ 2013, , 431, 1337

    2013

  90. [98]

    Rosdahl, J., Schaye, J., Teyssier, R., et al.\ 2015, , 451, 34

    2015

  91. [99]

    D., et al.\ 2016, , 830, 118

    Salgado, F., Bern \'e , O., Adams, J. D., et al.\ 2016, , 830, 118

    2016

  92. [100]

    R., Howk, J

    Sembach, K. R., Howk, J. C., Ryans, R. S. I., & Keenan, F. P.\ 2000, , 528, 310

    2000

  93. [101]

    E., Steidel, C

    Shapley, A. E., Steidel, C. C., Strom, A. L., et al.\ 2016, , 826, L24

    2016

  94. [102]

    A., & Ostriker, E

    Skinner, M. A., & Ostriker, E. C.\ 2015, , 809, 187

    2015

  95. [103]

    Smith, N.\ 2006, , 367, 763

    2006

  96. [104]

    J.\ 2007, , 379, 1279

    Smith, N., & Brooks, K. J.\ 2007, , 379, 1279

    2007

  97. [105]

    M., & Gardiner, T.\ 2009, , 14, 139

    Stone, J. M., & Gardiner, T.\ 2009, , 14, 139

    2009

  98. [106]

    M., Gardiner, T

    Stone, J. M., Gardiner, T. A., Teuben, P., Hawley, J. F., & Simon, J. B.\ 2008, , 178, 137

    2008

  99. [107]

    C., Beltr \'a n, M

    Tan, J. C., Beltr \'a n, M. T., Caselli, P., et al.\ 2014, Protostars and Planets VI, 149

    2014

  100. [108]

    Tielens, A. G. G. M.\ 2008, , 46, 289

    2008

  101. [109]

    Trebitsch, M., Blaizot, J., Rosdahl, J., Devriendt, J., & Slyz, A.\ 2017, , 470, 224

    2017

  102. [110]

    T.-H., & Milosavljevi \'c , M.\ 2018, , 478, 4142

    Tsang, B. T.-H., & Milosavljevi \'c , M.\ 2018, , 478, 4142

    2018

  103. [111]

    J., Smith, B

    Turk, M. J., Smith, B. D., Oishi, J. S., et al.\ 2011, , 192, 9

    2011

  104. [112]

    C., & Varoquaux, G.\ 2011, CSE, 13, 22

    van der Walt, S., Colbert, S. C., & Varoquaux, G.\ 2011, CSE, 13, 22

    2011

  105. [113]

    S., Oey, M

    Voges, E. S., Oey, M. S., Walterbos, R. A. M., & Wilkinson, T. M.\ 2008, , 135, 1291

    2008

  106. [114]

    M.\ 1992, , 258, 841

    Voit, G. M.\ 1992, , 258, 841

    1992

  107. [115]

    K., Whitworth, A

    Walch, S. K., Whitworth, A. P., Bisbas, T., W \"u nsch, R., & Hubber, D.\ 2012, , 427, 625

    2012

  108. [116]

    M., Monreal-Ibero, A., Verhamme, A., et al.\ 2018, , 611, A95

    Weilbacher, P. M., Monreal-Ibero, A., Verhamme, A., et al.\ 2018, , 611, A95

    2018

  109. [117]

    C., & Draine, B

    Weingartner, J. C., & Draine, B. T.\ 2001, , 548, 296

    2001

  110. [118]

    H., Demchenko, V

    Wise, J. H., Demchenko, V. G., Halicek, M. T., et al.\ 2014, , 442, 2560

    2014

  111. [119]

    H.\ 2019, arXiv e-prints, arXiv:1907.06653

    Wise, J. H.\ 2019, arXiv e-prints, arXiv:1907.06653

    2019 arXiv

  112. [120]

    G., Hollenbach, D., McKee, C

    Wolfire, M. G., Hollenbach, D., McKee, C. F., Tielens, A. G. G. M., & Bakes, E. L. O.\ 1995, , 443, 152

    1995

  113. [121]

    G., McKee, C

    Wolfire, M. G., McKee, C. F., Hollenbach, D., & Tielens, A. G. G. M.\ 2003, , 587, 278

    2003

  114. [122]

    D., et al.\ 2010, , 523, A6

    Deharveng, L., Schuller, F., Anderson, L. D., et al.\ 2010, , 523, A6

    2010

  115. [123]

    E.\ 2000, , 363, 9

    Zurita, A., Rozas, M., & Beckman, J. E.\ 2000, , 363, 9

    2000

Pith tools

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