Star Formation in the Ultraviolet

Round 1

Reviewer 1 Report

This is a very nice introduction to the significant volume that the author is editing. My only minor comment is around the discussion of He(II) emission in high-redshift galaxies.

When the author discusses He(II) emission in high-redshift galaxies I suggest they reference Shapley et al. (2003, ApJ, 588, 65) as they were the first to observe this line in a stack of high redshift galaxy and indicate that this could not be matched by extant stellar evolution models.

The statement of the He(II) line from cWR requires a reference to show this, as not all spectral synthesis models were able to reproduce it. The first to show this was possible with a realistic stellar population were Eldridge & Stanway (2012, MNRAS, 419, 479). With this confirmed by most recent observational sample and theoretical models in Steidel et al. (2016, ApJ, 826, 159). This latter study may be also included in the next subsection as it does make a link between the UV emission and the observed optical nebula emission lines observed in these galaxies.

The author may also wish to explicitly state in with their discussion that not only are new observations required, new models will be required too. To explain the full observed range of complex He(II) line morphologies a combined model of ES12+GV15+nebular emission is required.

Author Response

This is a very nice introduction to the significant volume that the author is editing. My only minor comment is around the discussion of He(II) emission in high-redshift galaxies.

When the author discusses He(II) emission in high-redshift galaxies I suggest they reference Shapley et al. (2003, ApJ, 588, 65) as they were the first to observe this line in a stack of high redshift galaxy and indicate that this could not be matched by extant stellar evolution models.

> I have added a reference to Shapley+03 and the other 2 papers [mentioned next] in lines 151+152.

 

The statement of the He(II) line from cWR requires a reference to show this, as not all spectral synthesis models were able to reproduce it. The first to show this was possible with a realistic stellar population were Eldridge & Stanway (2012, MNRAS, 419, 479). With this confirmed by most recent observational sample and theoretical models in Steidel et al. (2016, ApJ, 826, 159). This latter study may be also included in the next subsection as it does make a link between the UV emission and the observed optical nebula emission lines observed in these galaxies.

> I have added these references, mentioning the Steidel+16 paper in both contexts.

The author may also wish to explicitly state in with their discussion that not only are new observations required, new models will be required too. To explain the full observed range of complex He(II) line morphologies a combined model of ES12+GV15+nebular emission is required.

> Agreed and I now mention this in the new text of lines 176+177

Reviewer 2 Report

Comments attached.

Comments for author File: Comments.pdf

Author Response

This is an enticing introduction – I look forward to reading this Special Volume! COMMENTS:

> Many thanks for the very useful comments. I have indeed decided to add an HRD and a discussion on the
> Local He II emitting galaxies.

Abstract:
Last sentence:
1. I find this sentence a bit confusing as is. Specifically, the phrase "outlining the UV emission processes formlocal star formation" and the multiple uses of distance in different contexts. A suggested clarification could be: " The purpose of this Overview is to outline the basic physical principles driving UV emission processes from local (within 100 parsecs of) star formation, ranging from huge star-forming complexes containing hundreds of massive and very-massive stars, such as 30 Doradus (the Tarantula Nebula) in the neighboring Magellanic Clouds (only 50 kpc away), to galaxies near and far, out to the epoch of Cosmic Reionization."

> I have changed these last few lines in the Abstract in this way

Section 1: Introduction: General comments:
2. I think an overview of the ULLYSES survey is needed in the introduction prior to discussing it in subsequent sections.

> I understand this comment, but I would like to keep it short. ULLYSES is already discussed in the Abstract whilst a more detailed description for the T Tauri stars is found in Schneider et al., and for massive stars in the Leitherer contribution.

3. It woud be useful to quantitatively define what is meant by a few important themes in this review: "lowmetallicity", "massive stars", "very-massive stars"

--> I am now more explicit about low-metallicity, and the HRD Figure 1 and caption should make it easier to see the difference between a canonical massive O star and a VMS.

4. To make this overview more accessible to non-experts, this section could benefit from the addition of an HR diagram figure, with the different real-estate mentioned throughout the overview labeled (e.g., "birthline", "pre-main sequence", "H-rich part", "He-rich part")

--> Agreed. Added these terms to new Figure 1

1st paragraph:
5. In the rules of logical argument, I don't think massive size conclusively follows from pristine chemistry. A second sentence supporting the connection or a citation would be useful here.

--> Done. Line 18.

Paragraph 4:
6. I find the definitions of "O-type stars" and "very massive, H-rich Wolf-Rayet (WR) stars" to be a bit confusing. E.g., in defining a floor for the O-type stars, does "above 15−20 M⊙" mean the definition is: a. main sequence masses > 17.5 ± 2.5 M⊙?
b. main sequence masses > 15−20 M⊙, depending on the source, in which case they should be cited?
c. main sequence masses of 15 < M★/M⊙ < 20? etc. Paragraph 5:

--> I have now clarified it. I have taken 18 Msun from Weidner & Vink (2010), which converts spectral types into Masses. I did not want to cite it, as it would be another self-citation and it relies a lot on the atmosphere work of others.

--> The error bar is hard to give as this would depend on systematical physical errors, which are hard to estimate.

7. The sentence "it has become clear that approximately half the massive stars are part of a close binary system" is not strictly true and would benefit from a bit more detail or a caveat. For instance, this is not my area of expertise but I think binary fractions of massive stars have only been directly measured in the Milky way and the LMC (and the SMC?) and of WR stars in M33 and M31 and red giants in some local dwarf spheroidal (Spencer+18), and these do show a large variation in binary fractions. Further, how this scales for star- forming galaxies with properties such as metallicity is not clear.

--> I have changed 'approximately' to 'roughly' to accommodate for the correct comment on the fact it is still uncertain. So far only numbers for the MW and LMC are known (SMC is not). Sana+13 gives about 50% which is similar to e.g. Kobulnicky. Thanks for the Spencer+18 reference. I did not include it as it was about red giants, and not main sequence O stars, but this is interesting.

Section 2: T Tauri and Herbig Ae/Be Stars:
4th paragraph:
8. For the sentence "This is mostly due to the fact that PMS stars are faint and spectroscopy of PMSs in low- Z environments is only slowly starting to take off", I understand that PMSs are inherently difficult to observe due to their faintness and that this limited progress, but why haven't PMSs in low-Z environments specifically been observed?

--> That is a good question! I worked on this with Kalari (cited) and another group is de Marchi (had also been cited)

Section 3: Massive and Very-Massive Stars: Second paragraph:
9. "non-LTE" needs to be defined.

Done - line 97

Third paragraph:
10. "EM" needs to be defined.

--> Done - line 109

11. The first sentence "The expanding non-LTE wind models are generally able to successfully predict the observed UV and optical spectra of massive O and WR stars." should include details of the parameter space over which observations exist or include the caveat that, at present, no UV observations of verymetal- poor O stars (< 10% solar) exist.

--> I have added a relevant discussion 100 - 104.

12. I assume "but see later" will be replaced with the actual section of the Special Volume it is referring to?

--> clarified

Section 4: WR Stars:
1st Paragraph:
13. "the WR phenomenon is a spectroscopic classification and in principle is independent of evolutionary phase", but perhaps it is also important to mention that that the "flavor" of WR star can be explained by evolutionary phase?

--> No that's the point. In principle similar WR spectra, i.e. WR flavours could be resulting from wildly different evolutionary phases.

1st paragraph, last sentence:
14. This sentence is important! But for a non-expert, the details of what part of the Castor+75 prescription breaks down (i.e., they won't know what this prescription is or what the important part is) and why this is important are missing.

--> Done - line 124

2nd paragraph:
2

15. "Yet, we know that it is these VMSs that are dominant in terms of their ionising radiation and kinetic wind input" ... the message would be more clear here if the sentence was more explicit. The VMSs are dominant over what? Or VMSs are the dominant source of ionising radiation and kinetic wind in young stellar clusters? etc. What is the main message?

--> over the canonical O star population. See Line 131

Third paragraph: 1st sentence:
16. Again, please be explicit. Do you mean the observed total HeII emission? The nebular HeII recombination emission?

--> Line 137

Section 5: HeII Emission at High Redshift:
17. General: This section needs some significant editing. The author has done tremendous work to improve our understanding massive stars and their ionizing radiation shapes. However, this section on nebular HeII emission is shaped to highlight this work but only looking at some high redshift literature, and is missing much important, relevant information from local HeII-emitting galaxies.
As it stands, this section reads very one-sided.

--> I agree it was one-sided. I have therefore added new text on page 8 (lines 169-177) to account for this fair point.

The section could progress in one of two ways:
a. The interesting problem of producing strong, narrow HeII emission could be raised without delving into the author's suggested solution from VMSs.
b. A more well-rounded view of this challenging topic could be presented by including what we know about HeII emission from local galaxies. Here I highlight some of the other factors that should be considered or included:
i. VMSs are very short lived, which means the HeII emission proposed from slow VMS winds would only be observed from very young ionizing populations. Further, the ionizing stellar population properties have been measured for some for z~0 galaxies with HeII emission, and some have ages of ~10 Myr, suggesting no VMSs would remain.

--> I understand the point, but I am not sure I believe age claims such as 10 Myrs. The reason is that if we look at massive stars in a population like 30 Dor we see a population of about 10 Myrs spread out across the region, but also we see 1 Myr old VMS. As well as older red supergiant populations of 20 Myrs.

ii. On the other hand, my understanding from talking with Ylva Gotberg is that binarystripped He stars (Section 6) are more evolved stars and for these stars to be a solution to the HeII emission problem, the ionizing stellar population would either need to be older (>~ 40 Myr?), which conflicts with the stellar pop ages measured from the UV continua or SED fitting of many of these galaxies, or the galaxy would need to have had continuous star formation or subsequent, closely-spaced bursts of star-formation, which is certainly possible. However, it is also unclear to me if the expected IMF would produce a sufficient number of these stars in these very low-mass, metal-poor galaxies to reproduce the necessary HeII ionizing flux.

--> Given my response to (i) I feel the age discussion is beyond the scope of this introduction

iii. Recent studies by Kehrig+15,+18 have investigated the spatially extended HeII emission in SBS0335-052E and IZw 18 using MUSE IFU spectroscopy, and this spatial extent would also rule out the likelihood of stellar HeII emission from VMS.

--> This is a fair point, and I now discuss the spatial mismatch on page 8. I also explicitly state that
This is a challenge to stellar emissions from VMS.

3

iv. Narrow HeII emission is commonly observed from z~0 blue compact dwarf galaxies. These galaxies are generally low-metallicity, but certainly span a range, making metal-free pop III stars unlikely.

--> I agree, but I am not arguing for Pop III (Cassate+13 and Sobral+15 were).

v. Do the HeII EWs in Grafener & Vink (2015) include nebular continuum in the calculation? Because it is included in the observed HeII EWs they are comparing too.

--> Yes this is included using Starburst99.

vi. The problem isn't just the strength of the HeII lines, but the strength relative to the suite of other emission lines observed. Berg+18 find that metal-poor BPASS population synthesis models can reproduce the suite of UV emission lines in their metal-poor z~2 lensed galaxy, but not the HeII 1640 emission feature, even when 300 M⊙ stars are included. Further, Erb+19 found that this UV spectrum is remarkably similar to that of SBS 0335-052.

--> these local He II galaxy studies are now discussed on page 8.

vii. And other interesting works by Senchyna+17,+19, Jaskot+17, Berg+19, etc.

--> These authors have now been cited

Section 7: Summary and Outlook:
Second paragraph:
18. What do you mean by "both the stellar and the nebular HeII emission of VMS are still largely neglected in current population synthesis models of massive stars"? The nebular emission comes from photoionization modeling, not just population synthesis models.

--> agreed. Now mentioned on line 177.
4

Reviewer 3 Report

The article "Star Formation in the Ultraviolet" by Jorick S. Vink is an excellent introduction to the review of low-metallicity massive stars. The writing style is engaging and easy to follow, as expected for the introduction to a review. It sets the stage for many of the topics that appear to be covered in the review with a suitable introduction and I have no problem recommending the review for publication. The massive star community will greatly benefit from the work of this review as a new window into massive star physics is observationally revealed.

My only comments are minor suggestions.

(1) The ULLYSES project is not fully introduced until the conclusion, but it is mentioned throughout. A full description of the project that describes both the young and old components as well as the extent that the targets will address the observational problems laid out throughout the review may strengthen the ensuing discussions for those not familiar with ULLYSES. This is especially important because many of the issues in this introduction appear inspired by the observational opportunity afforded by the upcoming ULLYSES sample. 

(2) The manuscript rightly details many important outstanding observational issues with WR-stars, but it does not cover the extreme observational uncertainties associated with mass-loss rates of massive stars below LMC metallicity (e.g., Tramper et al. 2014, Bouret et al. 2015, Evans et al. 2019, Garcia et al. 2019), and the discovery-space available to diagnose the wind and photospheric conditions of these stars. HST is currently devoting a large observational investment with ULLYSES and other HST Cycle 27 proposals to disentangle these properties.  

(3) A small typographical comment: The last sentence in Section 3 should be capitalized ("whether" --> "Whether"). 

Author Response

The article "Star Formation in the Ultraviolet" by Jorick S. Vink is an excellent introduction to the review of low-metallicity massive stars. The writing style is engaging and easy to follow, as expected for the introduction to a review. It sets the stage for many of the topics that appear to be covered in the review with a suitable introduction and I have no problem recommending the review for publication. The massive star community will greatly benefit from the work of this review as a new window into massive star physics is observationally revealed.

--> Thanks for the compliment.

My only comments are minor suggestions.

(1) The ULLYSES project is not fully introduced until the conclusion, but it is mentioned throughout. A full description of the project that describes both the young and old components as well as the extent that the targets will address the observational problems laid out throughout the review may strengthen the ensuing discussions for those not familiar with ULLYSES. This is especially important because many of the issues in this introduction appear inspired by the observational opportunity afforded by the upcoming ULLYSES sample.

--> I understand where the reviewer is coming from, but the ULLYSES project is extensively discussed in additional reviews of this special volume.

As I already introduce ULLYSES in the Abstract, I feel that by emphasising it again in the Introduction would be a bit of an overkill. At the end of the day, the SF Volume in the UV would be relevant even without ULLYSES.

(2) The manuscript rightly details many important outstanding observational issues with WR-stars, but it does not cover the extreme observational uncertainties associated with mass-loss rates of massive stars below LMC metallicity (e.g., Tramper et al. 2014, Bouret et al. 2015, Evans et al. 2019, Garcia et al. 2019), and the discovery-space available to diagnose the wind and photospheric conditions of these stars. HST is currently devoting a large observational investment with ULLYSES and other HST Cycle 27 proposals to disentangle these properties.

--> Good point. I added this in lines 100 - 104

(3) A small typographical comment: The last sentence in Section 3 should be capitalized ("whether" --> "Whether").

--> I could not find this typo.