How Are Red and Blue Quasars Different? The Radio Properties

Round 1

Reviewer 1 Report

SEE ATTACHED FILE

Comments for author File: Comments.pdf

Author Response

Abstract – line 8: We show this enhancement is driven….. But no enhancement has been discussed therefore reader cannot understand what does it mean ‘this enhancement’. It is discussed in the text but should be introduced also in the abstract.
Thank you for pointing this out – we have now edited this sentence to make the abstract self-consistent.

Introduction, page 2 line 45. Authors should discuss why a difference in the SMBH properties in red and blue QSS is excluded.
Updated – we exclude this since in our upcoming study using X-shooter spectra of red and blue QSOs, we find no significant differences in the black hole properties. We include a line to comment of this future paper.

Line 77 it is used the same Greek letter (alpha) for Right Ascension and spectral index.
Updated.

I will prefer RA range in h.min instead of degree; I assume J2000 coordinates? Right?
Updated.

Figure 1 Survey area (deg) → square degree or deg^2
Updated.

Sect. 4.2 line 226: Is it statistically significant? What is the spatial scale of cQSO emission? Is it different?
In our previous studies, we have shown that red QSOs have a preference for compact (unresolved) radio emission (<43 kpc at z=1.5). Our e-MERLIN study observed rQSOs and cQSOs that were known to have compact radio emission at the resolution of FIRST (5’’; 43 kpc at z=1.5) in order to probe this compact population. If there were no differences on the galaxy-scale radio emission in rQSOs and cQSOs we would not expect a difference in the distribution of projected sizes. However, we find a statistically significant excess in the incidence of 2-10 kpc scale radio emission in rQSOs compared to cQSOs.
We have edited the text to make these points more clear.

Conclusions, line 273: any test? Is the peak statistically significant? Is the similar decreases statistically significant? I expect a low number of object in the low and upper range.
The shaded region shown in Fig. 5 encompasses the error in radio-detection enhancement. Therefore, both the peak and the decrease are statistically significant. The S82+C3GHz sample has large error bars due to the small sample size, but in both the FIRST and LoTSS samples, there are a large number of sources, even at the low and high radio-loudness values. We have edited the text and figure caption to make this more clear.

References: too long list of authors; I suggest: more than 5 authors, list 5 and et al.
Checking the reference style guide for MDPI journals they suggest: ‘For documents co-authored by a large number of persons (more than 10 authors), you can either cite all authors, or cite the first ten authors, then add a semicolon and add ‘et al.’ at the end’. We have updated the references following this guidance to 10 names, fewer if the author list becomes alphabetical and more if the number of authors is 11-12.

Reviewer 2 Report

This work summarizes previous results of the authors on the existence of a higher radio detection rate of red quasars, at around the RL/RQ boundary, when compared with a sample of quasars selected around the mean distribution of (g-i) color index. In addition, they perform a joint analysis for four quasar samples observed in four different radio surveys and present some new results of a subsample observed with e-Merlin.

 

I found the paper interesting and I think it is able to be published in Galaxies after some changes, as there are a few points that need clarification in the paper.

 

This referee understands that this paper contains a summary of the previous work, plus some additional data and supplementary analysis, but as a general comment I suggest:

 

- Highlight more clearly which are the new contributions of this paper with respect to the ones previously published. Given that several of the conclusions are not new with respect to the previous cited papers, the net contribution of this study should be better pointed out and the novel results of this work should be more clearly stated.

 

Other specific comments:

 

- The abstract must be self-consistent. A fundamental sentence has been lost in it. The abstract mentions an enhancement (sentence "We show this enhancement..." ), but it does not say what kind of enhancement, or with respect to what,…

 

- In the data shown in Figure 2 there appears to have a difference in the 1.4 GHz luminosity distributions between rQSOs and cQSOs in the FIRST sample, with cQSO (FIRST) showing a different distribution of L1.4GHz (higher luminosities) than rQSO, while are more similar in the LoTSS sample. Given that, as noted in the paper, it is essential that both rQSO and cQSO samples are as similar as possible, I suggest, as it would be illustrative, to include on the sides of the two bottom figures, the ones of L (6 microns) vs. L (1.4 GHz), histograms of the luminosity distributions in L(1.4 GHz) and L (6microns) for both rQSO and cQSOs for the two most numerous samples (FIRST and LoTSS), and that have the greatest statistical significance for the study. Also it will be very useful to include a more detailed comment about its possible influence in the results.

Are there common objects included in both FIRST and LoTSS samples, in cQSO and rQSOS, that could shed more light on the origin of radio enhancement?

Are there differences for bins in redshift?

 

- Fig. 3. It is difficult to distinguish the different samples in this figure. Representing with different colors (+ different line types) each sample would help to see them more clearly. Also there will be a correspondence in the colors of the upper and lower plots, and concordance with next figures in the paper. Also an expansion of the y-axis scale in the botton plot will help to appreciate the behaviour of the radio enhancement of rQSO with L(1.4GHz).

 

- Section 4.- The paper concludes very categorically that the enhancement is not due to SF, but a further explanation of the data and arguments on which this conclusion is based, and probably a larger discussion, would be required in the opinion of this referee to stablish that conclusion. Can the authors state that there is not a population of red quasars in which radio enhancement could be due to a significant contribution of SF? As the authors themselves mention, a more in-depth analysis of the spectroscopic properties is necessary. There are studies in the literature showing the possible importance of SF on the radio emission in quasars or at least in a fraction of quasars (see e.g. Ganci et al. 2019; Macfarlane et al. 2021, ...)

 

- One of the arguments, in the discussion, against the idea that outflows are connected to the radio emission of the rQSO is based on the fact that there are red nonBALs showing the enhancement, however the authors must take into account in their discussion that not only BALs show outflows. There is evidence, published in many papers (for example the ones related to CIV1540A), that highly accreting non-BAL quasars show also strong nuclear winds/outflows.

 

- I agree with the conclusion that red quasars are an interesting population in the context of quasar evolution, and that their results about the enhancement in the radio detection of red quasars is a remarkable result, but also that a deep study of the BLR properties, accretion rates,.. of the presented samples is necessary to help to understand the origin of the radio enhancement.

Author Response

We thank the referee for their constructive comments. We address the comments below, changes in the manuscript are marked up using the “Track Changes” function in LaTeX.
Our main changes are to the discussion section, which we have expanded to more carefully and clearly compare the potential mechanisms driving the enhanced radio emission. We have also condensed the conclusions from 3 bullet points to 2, highlighting the novel work in this paper compared to previous studies.

This referee understands that this paper contains a summary of the previous work, plus some additional data and supplementary analysis, but as a general comment I suggest:

- Highlight more clearly which are the new contributions of this paper with respect to the ones previously published. Given that several of the conclusions are not new with respect to the previous cited papers, the net contribution of this study should be better pointed out and the novel results of this work should be more clearly stated.
We agree that the new analyses undertaken in this work were not presented clearly enough. We have condensed the conclusions from 3 bullet points to 2, in order to highlight the novel work presented in this study. In this work, we use a consistent parent sample to explore and compare the radio properties in red and blue QSOs. We extend Klindt et al. 2019, who use SDSS DR7, to a sample size 10 times larger with DR14. With this, we confirm the results to a much higher significance, and also find a decrease in the radio-detection enhancement towards the radio-quiet end, which was not seen in the previous work and is consistent with the deeper radio data.

Other specific comments:

- The abstract must be self-consistent. A fundamental sentence has been lost in it. The abstract mentions an enhancement (sentence "We show this enhancement..." ), but it does not say what kind of enhancement, or with respect to what,…
Thank you for pointing this out – we have now edited this sentence to make the abstract self-consistent.

- In the data shown in Figure 2 there appears to have a difference in the 1.4 GHz luminosity distributions between rQSOs and cQSOs in the FIRST sample, with cQSO (FIRST) showing a different distribution of L1.4GHz (higher luminosities) than rQSO, while are more similar in the LoTSS sample. Given that, as noted in the paper, it is essential that both rQSO and cQSO samples are as similar as possible, I suggest, as it would be illustrative, to include on the sides of the two bottom figures, the ones of L (6 microns) vs. L (1.4 GHz), histograms of the luminosity distributions in L(1.4 GHz) and L (6microns) for both rQSO and cQSOs for the two most numerous samples (FIRST and LoTSS), and that have the greatest statistical significance for the study. Also it will be very useful to include a more detailed comment about its possible influence in the results.
We agree that the way the data was presented in the original plot was not clear. We have updated and split Fig. 3 to make this point more clear. Fig. 2 now has the colour distribution for the overall parent sample and Fig. 3 has the radio-6micron luminosity distributions, with 1.4 GHz histograms for the FIRST and LoTSS samples.
We do not match in radio luminosity between rQSOs and cQSOs as this is something we want to explore and compare between the two samples. The reason that rQSOs tend to have lower radio luminosities for a given 6micron luminosity is reflected in Fig. 5: they tend to be more radio-quiet compared to their blue counterparts (although not at the extreme radio-quiet end). We have also edited the text to make this point more clear.

Are there common objects included in both FIRST and LoTSS samples, in cQSO and rQSOS, that could shed more light on the origin of radio enhancement?
There are common objects included in both the FIRST and LoTSS samples. However, a comparison of radio spectral indexes was already carried out in Rosario+2020, where we do not find a difference. We have included a sentence in the paper that points the reader towards this study for more details and also use this to expand on what we might expect to see with our new uGMRT data.

Are there differences for bins in redshift?
We do find differences in the detection rate and enhancement between low and high redshift QSOs (Fig. 8 in Klindt et al. 2019), but an analysis of the population indicates this is driven by host-galaxy contamination affecting the colour selection in the lower-luminosity QSOs at low redshift. Host-galaxy contamination is negligible at redshifts > 1. We will compare host-galaxy/dust reddened quasars in a future study using deeper optical data.

- Fig. 3. It is difficult to distinguish the different samples in this figure. Representing with different colors (+ different line types) each sample would help to see them more clearly. Also there will be a correspondence in the colors of the upper and lower plots, and concordance with next figures in the paper. Also an expansion of the y-axis scale in the botton plot will help to appreciate the behaviour of the radio enhancement of rQSO with L(1.4GHz).
Agreed – we have updated the colours to make this more clear.

- Section 4.- The paper concludes very categorically that the enhancement is not due to SF, but a further explanation of the data and arguments on which this conclusion is based, and probably a larger discussion, would be required in the opinion of this referee to stablish that conclusion. Can the authors state that there is not a population of red quasars in which radio enhancement could be due to a significant contribution of SF? As the authors themselves mention, a more in-depth analysis of the spectroscopic properties is necessary. There are studies in the literature showing the possible importance of SF on the radio emission in quasars or at least in a fraction of quasars (see e.g. Ganci et al. 2019; Macfarlane et al. 2021, ...)
There may be a population of red QSOs where this could be the case, but we do not believe these are the “special” red QSOs with enhanced radio emission. In our various supporting papers of this work (Fawcett+2020; Rosario+2020; Calistro Rivera+2021) we show that the radio enhancement is unlikely to be due to a significant contribution by SF. In an upcoming ALMA study we also demonstrate that the direct FIR emission in red QSOs is not different from normal blue QSOs. We have edited the text to make it clear that we rule out SF as the driver of the radio enhancement in red QSOs.

- One of the arguments, in the discussion, against the idea that outflows are connected to the radio emission of the rQSO is based on the fact that there are red nonBALs showing the enhancement, however the authors must take into account in their discussion that not only BALs show outflows. There is evidence, published in many papers (for example the ones related to CIV1540A), that highly accreting non-BAL quasars show also strong nuclear winds/outflows.
We agree with this statement – the sentence in the paper was poorly worded and has been updated.

- I agree with the conclusion that red quasars are an interesting population in the context of quasar evolution, and that their results about the enhancement in the radio detection of red quasars is a remarkable result, but also that a deep study of the BLR properties, accretion rates,.. of the presented samples is necessary to help to understand the origin of the radio enhancement.
We are currently working on a paper using X-shooter data to explore the accretion properties of red and blue quasars, to see if this could be driving the enhancement in the radio properties. We do not find any significant differences in the accretion properties between red and blue QSOs.
We have added a line to comment on this future paper.

Round 2

Reviewer 2 Report

Congratulations to the authors for their review process The new version of the paper has been improved and I think it is now suitable for publication.