The Long-Lasting Territories of Forest Apex Predators Sustain Diverse Bird Communities throughout the Year

Round 1

Reviewer 1 Report

I read with interest the manuscript "The long-lasting territories of forest top raptors sustain diverse bird communities throughout the year". The authors demonstrated that long-lasting raptor nest sites retain higher bird richness and abundance. Furthermore, the authors revealed the characteristics of these occupied and unoccupied sites, which are important for recommendations for forest management practice (as stated in the Discussion and Conclusion of this manuscript). Overall, the study has important data that advance our knowledge regarding raptors as biodiversity surrogates. However, there are many concerns and ambiguities to be dealt with or corrected throughout the manuscript. For the presentation of this study, I encourage the authors' amendments so that this work will be successfully published with scientifically rigorous quality. Please see the following specific comments on each section.

Introduction:

Major comments:

The authors explained why multiple raptors should be assessed as biodiversity indicators and why wintering season should be focused on. These are of course good and well-written; however, I do not think the authors clearly explained why biodiversity indicators are necessary for practical conservation. Given the outcome of this study, the center of the authors' objective may be demonstrating the effectiveness of multiple raptors as year-round avian diversity surrogates. If so, in the Introduction, the authors should explain why biodiversity proxy is necessary and why raptors are more appropriate than other indicator candidates. For example, biodiversity surrogates should be readily measurable than directly measuring biodiversity per se. Alternatively, as reviewed by Sergio et al. (2008, Annu. Rev. Ecol. Evol. Syst. 39: 1–19.) and stated in Burgas et al. (2014, J. Appl. Ecol. 51: 786–794) and Senzaki & Yamaura (2016, Wetl. Ecol. Manag. 24: 427–441.), this aspect may be explained as the flagship role of raptors, which provides essential value for conservation. Anyway, I would suggest the authors neatly write why raptors are good candidates as indicators.

Furthermore, the authors focused on the importance of surveying wintering birds in testing the indicator role of raptors. From this point of view, important literature is lacking. To my knowledge, so far, Natsukawa (2021, Urban For. Urban Green. 60: 127066.) is the sole study that tests whether goshawks (the authors' target species) are proxies for wintering bird diversity; therefore, it is a bit confusing why the authors' missed the article. More directly speaking, the authors need to mention how this study is superior to the existing Natsukawa (2021). As stated by the authors themselves, the study surveyed bird diversity throughout the year (this is an excellent point of this manuscript), so the authors could clearly explain it. 

Minor comments:

Line 26: "a good indicators of biodiversity" to "a good indicator of biodiversity"

Line 31: I do not understand the term, "including anthropogenic". Please clarify.

Lines 57-59: "Raptor territories being indicators of biodiversity should show similar patterns over a year [20, 21], therefore studies on wintering and breeding bird communities could verify that assumption" I would change it to "...[20, 21]; therefore, studies..." 

Materials and Methods:

Major comments:

If I understand correctly, the authors censused bird diversity within c.a. 1 km2 at the territories of goshawks and Ural owls without locating their occupied nests. I am a bit doubtful if this is methodologically robust. As the authors themselves explained in their manuscript, goshawks tend to have multiple alternative nests in their territories. The problem is how long the distances between these nests are apart from each other in the surveyed areas in Poland. As stated by Reynolds et al. (2017, Wildl. Monogr. 197: 1–40), they are highly mobile even in their territories; thus, the distances are sometimes greater than 500 m. In that case, there is a possibility that the extent the authors censused would be located quite apart from long-lasting nests of goshawks. Although it is unavoidable that the location of their long-lasting nests was not available as stated in the manuscript, If the distances between multiple nests in the same territory are often great, then the census the authors conducted may not adequately reflect the environment of the historic nest sites of goshawks. The authors surveyed whether each long-lasting territory was occupied from 2019 to 2021. Thus, I am wondering why the authors did not survey bird species within 1.0 km from the newest location of the nests detected in 2019 to 2021. This concern is also true in Ural owls. The owl has a narrow range compared to goshawks, but the same problems would potentially occur. I hope my concern is unfounded and not applicable to this study. If so, I would suggest the authors clearly explain so that readers will not have the same doubts as mine.

Minor comments:

Line 125: “The localization of breeding sites were”, please change it to “The localization of breeding sites was”.

Line 186: “This work do not assume to analyze details”, please change it to “This work does not assume to analyze details”. 

Results and Discussion:

Major comments:

Whether the outcomes and conclusion of the study are reasonable or not depends on the methodological concerns mentioned above. Furthermore, the Figures presented are not so reader-friendly. The text in the Figures is too small to decipher. I would suggest the authors radically recreate all the Figures. Lastly, I personally think the authors' statements are a bit redundant and exaggerated. For instance, the authors suggested the specific conservation recommendations in Lines 467-471, "It would be enough to restrict timber logging in these core areas, left many old trees (which are important for localization of nests, but also as shelters and hunting zones as both raptors hunt from an ambush), and particularly to avoid tree cutting during mating and breeding periods (late winter and spring) (recommendations already available for Ural owl conservation [61])." Such implications are of course important but should be stated with scientifically precise (ad hoc) evidence of the habitat selection of the two target raptors. Apparently, the study did not perform such ad hoc analysis on the habitat requirements of these raptors, as stated by the authors themselves (Line 181; "This work do not assume to analyze details about habitat requirements of goshawks or Ural owls"). Indeed, the effects of overwintering habitat alterations around breeding territories adversely influence the occupancy of raptors (Morant et al. 2018, Anim. Conserv. 21: 465–473). Given this, the authors should be more prudent in conserving long-lasting raptor habitats and present additional convincing data if they want to suggest such recommendations.

Minor comments:

Line 412: “(‘protective’ associates; e.g. [55-57])”, please consider citing Greeney et al. (2015, Sci. Adv. 1, e1500310), which is more appropriate in this context.

Author Response

We are grateful to both Reviewers for their valuable comments and suggestions, which enable us to improve our manuscript and allow for the avoidance of errors and biases in our study.

Below we provided answers or justifications to all comments.  

Comments and Suggestions for Authors

I read with interest the manuscript "The long-lasting territories of forest top raptors sustain diverse bird communities throughout the year". The authors demonstrated that long-lasting raptor nest sites retain higher bird richness and abundance. Furthermore, the authors revealed the characteristics of these occupied and unoccupied sites, which are important for recommendations for forest management practice (as stated in the Discussion and Conclusion of this manuscript). Overall, the study has important data that advance our knowledge regarding raptors as biodiversity surrogates. However, there are many concerns and ambiguities to be dealt with or corrected throughout the manuscript. For the presentation of this study, I encourage the authors' amendments so that this work will be successfully published with scientifically rigorous quality. Please see the following specific comments on each section.

Introduction:

Major comments:

The authors explained why multiple raptors should be assessed as biodiversity indicators and why wintering season should be focused on. These are of course good and well-written; however, I do not think the authors clearly explained why biodiversity indicators are necessary for practical conservation. Given the outcome of this study, the center of the authors' objective may be demonstrating the effectiveness of multiple raptors as year-round avian diversity surrogates. If so, in the Introduction, the authors should explain why biodiversity proxy is necessary and why raptors are more appropriate than other indicator candidates. For example, biodiversity surrogates should be readily measurable than directly measuring biodiversity per se. Alternatively, as reviewed by Sergio et al. (2008, Annu. Rev. Ecol. Evol. Syst. 39: 1–19.) and stated in Burgas et al. (2014, J. Appl. Ecol. 51: 786–794) and Senzaki & Yamaura (2016, Wetl. Ecol. Manag. 24: 427–441.), this aspect may be explained as the flagship role of raptors, which provides essential value for conservation. Anyway, I would suggest the authors neatly write why raptors are good candidates as indicators.

Response: We added justification in this matter. All references listed above were already cited in our paper.

Furthermore, the authors focused on the importance of surveying wintering birds in testing the indicator role of raptors. From this point of view, important literature is lacking. To my knowledge, so far, Natsukawa (2021, Urban For. Urban Green. 60: 127066.) is the sole study that tests whether goshawks (the authors' target species) are proxies for wintering bird diversity; therefore, it is a bit confusing why the authors' missed the article. More directly speaking, the authors need to mention how this study is superior to the existing Natsukawa (2021). As stated by the authors themselves, the study surveyed bird diversity throughout the year (this is an excellent point of this manuscript), so the authors could clearly explain it. 

Response: We added this reference, however we do not think that there is a need for justification why our study is or is not to this research as work of Natsukawa (2021) is about winter birds in urban environment, and our work is on forest birds.

Minor comments:

Line 26: "a good indicators of biodiversity" to "a good indicator of biodiversity"

Response: Done

Line 31: I do not understand the term, "including anthropogenic". Please clarify.

Response: We rewritten this to: …both natural and caused by human…

Lines 57-59: "Raptor territories being indicators of biodiversity should show similar patterns over a year [20, 21], therefore studies on wintering and breeding bird communities could verify that assumption" I would change it to "...[20, 21]; therefore, studies..." 

Response: Done

Materials and Methods:

Major comments:

If I understand correctly, the authors censused bird diversity within c.a. 1 km2 at the territories of goshawks and Ural owls without locating their occupied nests. I am a bit doubtful if this is methodologically robust. As the authors themselves explained in their manuscript, goshawks tend to have multiple alternative nests in their territories. The problem is how long the distances between these nests are apart from each other in the surveyed areas in Poland. As stated by Reynolds et al. (2017, Wildl. Monogr. 197: 1–40), they are highly mobile even in their territories; thus, the distances are sometimes greater than 500 m. In that case, there is a possibility that the extent the authors censused would be located quite apart from long-lasting nests of goshawks. Although it is unavoidable that the location of their long-lasting nests was not available as stated in the manuscript, If the distances between multiple nests in the same territory are often great, then the census the authors conducted may not adequately reflect the environment of the historic nest sites of goshawks. The authors surveyed whether each long-lasting territory was occupied from 2019 to 2021. Thus, I am wondering why the authors did not survey bird species within 1.0 km from the newest location of the nests detected in 2019 to 2021. This concern is also true in Ural owls. The owl has a narrow range compared to goshawks, but the same problems would potentially occur. I hope my concern is unfounded and not applicable to this study. If so, I would suggest the authors clearly explain so that readers will not have the same doubts as mine.

Response: Thank you for this important remark. Indeed, this was the greatest problem when we decided to focus on this topic. We tried to get information about localization of nest from the past but this occurred to be not possible. Therefore, we had to rely only on more general information about localization of territories from 1980s till current times. That is why in the paper we write “long-lasting territories” not “long-lasting nests”. This is particularly true as in managed forests there is almost not possible that nest will be present over 10-20 years or longer as either tree with nest will be fallen during logging, or the nest will be dropped down by local people (because goshawks are still persecuted in Poland). We had to assume some distance for bird counts, and we decided to use c. 500 m simply due to limitations of work in the field – if we select larger distance so larger area, it would be better to count birds on more points to have representative data for such an area. Another reason for use of 560 m radius was that we were aware that in managed stands many forest patches could be logged so if we included larger area as a defined territory, we probably had some point counts in forests being logged recently or during our surveys. We are aware that this decision has some impact on results of our study but as all counts were made in the same methodology, we expect that we controlled systematic errors related with sampling. In our opinion this was the only solution to conduct such kind of study – otherwise we will either resign or do some other kind of research, not focusing on long-lasting territories but rather on nests known to be occupied by some time (however not longer than a few years), what would make our study not being different from previous similar research on the impact of raptors on biodiversity (without temporal aspect).

Minor comments:

Line 125: “The localization of breeding sites were”, please change it to “The localization of breeding sites was”.

Response: Done

Line 186: “This work do not assume to analyze details”, please change it to “This work does not assume to analyze details”. 

Response: Done

Results and Discussion:

Major comments:

Whether the outcomes and conclusion of the study are reasonable or not depends on the methodological concerns mentioned above. Furthermore, the Figures presented are not so reader-friendly. The text in the Figures is too small to decipher. I would suggest the authors radically recreate all the Figures. Lastly, I personally think the authors' statements are a bit redundant and exaggerated. For instance, the authors suggested the specific conservation recommendations in Lines 467-471, "It would be enough to restrict timber logging in these core areas, left many old trees (which are important for localization of nests, but also as shelters and hunting zones as both raptors hunt from an ambush), and particularly to avoid tree cutting during mating and breeding periods (late winter and spring) (recommendations already available for Ural owl conservation [61])." Such implications are of course important but should be stated with scientifically precise (ad hoc) evidence of the habitat selection of the two target raptors. Apparently, the study did not perform such ad hoc analysis on the habitat requirements of these raptors, as stated by the authors themselves (Line 181; "This work do not assume to analyze details about habitat requirements of goshawks or Ural owls"). Indeed, the effects of overwintering habitat alterations around breeding territories adversely influence the occupancy of raptors (Morant et al. 2018, Anim. Conserv. 21: 465–473). Given this, the authors should be more prudent in conserving long-lasting raptor habitats and present additional convincing data if they want to suggest such recommendations.

Response: Thank you. We re-write largely discussion, removing some unnecessary fragments and we also modified conclusions avoiding speculative parts. We supported some recommendations by results of other studies, including our own from the same area and species, as well the above citation.  

Minor comments:

Line 412: “(‘protective’ associates; e.g. [55-57])”, please consider citing Greeney et al. (2015, Sci. Adv. 1, e1500310), which is more appropriate in this context.

Response: Done

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript is novel and shows interesting information about how raptors are a good indicator of the avian diversity in the forests of Polland. However, there are several issues throughout the manuscript (especially in the methods and results). Further, the authors should strongly consider a service of English copy-editing, since many of the sentences are unclear, making the whole manuscript hard to follow.

Title: Replace top raptor with top predator

Introduction: Introduction shows well the conceptual framework in which the authors are working. However, it is not well-written, and it is hard to follow.

Aims: Aims are well stated. Maybe the manuscript would be benefitted if the authors add a couple of sentences with the rough methods used.

Methods: Authors selected the sites by choosing old raptor territories and control sites (they should be clearly named control in the manuscript, instead of unoccupied). I think the methodological design is fine, but then it is quite hard to read. Also, statistics need more clarity on what was done, for achieving which aim, and if the statistical assumptions of parametric tests were met before of using linear regressions and ANOVAs.

Results: Results are interesting, and seem correct. But again, it is quite unorganized and hard to follow. Please, rewrite to be more clear. A relevant suggestion is to avoid using the names of the variables that you named in the models (e.g., NATURA), and write the ecological significance instead 

Discussion: I think the discussion is well structured and relates the results of this research with the literature.

Specific comments:

L27: Replace species rich with diverse

L28: Replace "apex raptors" with "apex predators such as raptors"

L45: Replace "constant" with "permanent"

L47: The statement is not so clear. 

L49: Replace by "Usually it is unknown if raptor territories" 

L54: Add "The" to the beginning of the statement ("The majority")

L54: Delete top from top raptor (here and elsewhere)

L57: Replace being by being

L60: Replace were with are

L63: It is stated as if raptors would shelter the communities, and does not work as that. The habitat selection of the raptors is as a specialist, and the community is in general richer.

L65: Why the third constraint? Because of the other limitations? I suggest replacing the enumeration. This statement is not well-written and should be more clear.

L69: Replace by "Consequently, some habitats are missing [...]"

L83: Replace surrogates with indicators.

L87: Why "had to be"? It is better if you state the reason here.

L93: It would be valuable to add a map with the Study area (could be in Supplementary material)

L107: It is not clear why you did not work with all the sites (24 and 22)

L135: It is not a bird inventory, but a bird survey

L136: Merge this sentence with the next paragraph.

L139: Why did you make transects instead of point counts (generally point counts work better for forest birds)? In the breeding season, you work with point counts, so it is not completely comparable.

L155: 100 meters of radius seems a bit large for point counts (usually, 50 meters are suggested)

L164: It is not clear what 20.III (etc) means.

L181: What do you mean with "do not assume to analyze"? do not aim to analyze?

L186: Eliminate However

L188: Improve the clarity of this sentence. What was designed?

L192: Replace noted by "quantified"

L202: And which variable did you finally choose for building the models?

L204: You should first mention why you are plotting the rarefaction curves

L206: Replace augmenting by increasing

L206: Which nonparametric method are you using?

L208: Replace unoccupied by control (here and elsewhere)

L210: Justify why you used these species as a taxonomic control. Also, you checked for differences in the abundance of these three species between the treatments and the control? It is not clear.

L213: Did you check if the data achieve the ANOVA assumptions?

L217: It's the presence of the raptors or the comparison between the territories and the control? This sentence is very unclear. Also, it's not clear if you are talking about the GLM, or simply linear models.

L225:Abundance of which species? For each species?

L233: These results should be shown in a table or graph to visualize them better. Table S2 does not show this clearly

L250: The majority

L250: were recorded instead of noted (change here and elsewhere)

L250: Is this difference significant? (did you test it?) If it is tested, you should provide the t-value, p-value, and model estimate.

L252: In the treatment or the controls?

L253: Check the rule of capitals when you're mentioning the common name of the species

L256: Again, are these differences significant?

L267: This is the main result of the manuscript, so should be earlier in the text. Also, you have to show the statistics (see comment L250)

L267: Seeing Figure 2, what you did is apparently a binomial GLM instead of a linear regression! Calculating the probability of occurrence... Please, state this clearly in the methods, if I'm right.

L272: Lack of statistics again

L281: Do not mention the name of your variables (WOOD/NATURA), but their ecological significance (here and elsewhere). You should avoid the public having to come back to Methods to read what WOOD means.

L290: Showed that in winter... not wintering period

Figure 1: It is very unclear what all the abbreviations mean. Be more clear.

L452: It is not so clear to me that what is relevant is the maintenance f the territories or the maintenance of the habitat conditions. 

Author Response

We are grateful to both Reviewers for their valuable comments and suggestions, which enable us to improve our manuscript and allow for the avoidance of errors and biases in our study.

Below we provided answers or justifications to all comments.  

Comments and Suggestions for Authors

The manuscript is novel and shows interesting information about how raptors are a good indicator of the avian diversity in the forests of Polland. However, there are several issues throughout the manuscript (especially in the methods and results). Further, the authors should strongly consider a service of English copy-editing, since many of the sentences are unclear, making the whole manuscript hard to follow.

Response: We ordered proofreading in the commercial company by a native speaker (Dr Emily Kane, ordered to company http://specjalistyczne.biz/index.php?p=2_2_About-Us ). English is corrected in the current version.

The whole article was intensively re-edited, especially the introduction, results and discussion (considering the comments of other reviewers).

Title: Replace top raptor with top predator

Response: We consistently change “top raptors” into “apex predators” in the text, but we left “raptors” where we refer to diurnal birds of prey or owls (not specifying if top or subordinate taxa).

Introduction: Introduction shows well the conceptual framework in which the authors are working. However, it is not well-written, and it is hard to follow.

Response: The introduction was re-written (also considering comments of other reviewers).

Aims: Aims are well stated. Maybe the manuscript would be benefitted if the authors add a couple of sentences with the rough methods used.

Response: Thank you. We do not think that the Introduction (particularly aims) is the place for a description of methods, which are described in the following chapter.

Methods: Authors selected the sites by choosing old raptor territories and control sites (they should be clearly named control in the manuscript, instead of unoccupied). I think the methodological design is fine, but then it is quite hard to read. Also, statistics need more clarity on what was done, for achieving which aim, and if the statistical assumptions of parametric tests were met before of using linear regressions and ANOVAs.

Response: We changed accordingly (now all control plots are called “controls”).

The description of statistics was re-edited (also considering the comments of other Reviewers).

Results: Results are interesting, and seem correct. But again, it is quite unorganized and hard to follow. Please, rewrite to be more clear. A relevant suggestion is to avoid using the names of the variables that you named in the models (e.g., NATURA), and write the ecological significance instead 

Response: Please see our response in this matter to the comment of Reviewer 3.

Previously we arranged results according to the order of methods in use. Particular aims of this study are solved by various methods, therefore it is not possible to simply present results according to aims (as in this way some results need to be repeated). We propose the different organization of result presentation following sequence: presentation of general bird diversity metrics, comparison of bird diversity metrics with respect to raptor territories, description of differences in environmental variables in raptor territories, bird diversity metrics in relation to the environment and raptor territories.

We would rather left the names of variables in capital letters instead of describing their ecological significance as the latter way force much more text in the article. We explained all these variables in methods so it should be clear what they are.

Discussion: I think the discussion is well structured and relates the results of this research with the literature.

Response: Thank you. However, other Reviewers had the opposite opinions, therefore we had to edit the discussion also.

Specific comments:

L27: Replace species rich with diverse

Response: We left it as is because here we mean “species richness”, not exactly ‘species diversity”, which is resultant of richness and abundance.

L28: Replace "apex raptors" with "apex predators such as raptors"

Response: Done

L45: Replace "constant" with "permanent"

Response: Done

L47: The statement is not so clear. 

Response: Rewritten.

L49: Replace by "Usually it is unknown if raptor territories" 

Response: Done

L54: Add "The" to the beginning of the statement ("The majority")

Response: Done

L54: Delete top from top raptor (here and elsewhere)

Response: “Top” was changed to “apex”

L57: Replace being by being

Response: I am sorry but I do not understand this suggestion.

L60: Replace were with are

Response: Done

L63: It is stated as if raptors would shelter the communities, and does not work as that. The habitat selection of the raptors is as a specialist, and the community is in general richer.

Response: Rewritten.

L65: Why the third constraint? Because of the other limitations? I suggest replacing the enumeration. This statement is not well-written and should be more clear.

Response: Because most studies on raptors as surrogates of biodiversity were done on single species (of raptors), so simultaneous, cumulative or adverse impact of various raptors is poorly known.

L69: Replace by "Consequently, some habitats are missing [...]"

Response: Rewritten.

L83: Replace surrogates with indicators.

Response: Done

L87: Why "had to be"? It is better if you state the reason here.

Response: This is explained in next sentences.

L93: It would be valuable to add a map with the Study area (could be in Supplementary material)

Response: We prepared such maps in google maps and added this as supplementary File 1.

L107: It is not clear why you did not work with all the sites (24 and 22)

Response: We intended to use the same number of sites for a goshawk and Ural owl (and for control plots). Moreover, we were aware that some sites could be logging during the year of bird surveys, which could lead to erroneous results caused by the intensive presence of forestry workers and machines in the surrounding. We explained this also in the revised text.

L135: It is not a bird inventory, but a bird survey

Response: Corrected

L136: Merge this sentence with the next paragraph.

Response: Done

L139: Why did you make transects instead of point counts (generally point counts work better for forest birds)? In the breeding season, you work with point counts, so it is not completely comparable.

Response: Point counts work well in the breeding period, but during winter birds are highly dispersed in forests, which some areas of high concentration (e.g. around sources of food), and birds are mobile during the winter period. Point counts could be good in studies on some specific sites, but we wanted to know bird diversity over a territory of raptors, which are large. That is why we selected transects, which enable us to know bird species richness and abundance for a large forest area and this method shows some average bird diversity not affected by conditions in particular points.

L155: 100 meters of radius seems a bit large for point counts (usually, 50 meters are suggested)

Response: We used 100 m in other studies and also many other studies used such a distance. 100 m could be too long in dense forests, but oak and pine forests are relatively open, and it is not a problem to hear bird songs over 100 m.

L164: It is not clear what 20.III (etc) means.

Response: III = March. We rewritten these numbers into months.

L181: What do you mean with "do not assume to analyze"? do not aim to analyze?

Response: Yes. Such a analyzes of habitat requirements were done in the previous paper (cited).

L186: Eliminate However

Response: Done

L188: Improve the clarity of this sentence. What was designed?

Response: Done

L192: Replace noted by "quantified"

Response: Done

L202: And which variable did you finally choose for building the models?

Response: All because these 2 pairs of variables were not strongly correlated.

L204: You should first mention why you are plotting the rarefaction curves

Response: Done

L206: Replace augmenting by increasing

Response: Done

L206: Which nonparametric method are you using?

Response: Chao 1 estimator

L208: Replace unoccupied by control (here and elsewhere)

Response: Done

L210: Justify why you used these species as a taxonomic control. Also, you checked for differences in the abundance of these three species between the treatments and the control? It is not clear.

Response: We planned to add such a taxonomic control, but finally we resigned and this sentence was just left in Methods but should be removed.

We selected 3 bird species of moderate size, moderate abundance and omnivorous to check if their occurrence was also related to higher diversity of birds (like raptors).

L213: Did you check if the data achieve the ANOVA assumptions?

Response: Yes. All analyzed data (bird diversity metrics) have a normal population distribution, similar variance, and are independent (within each analysis).

L217: It's the presence of the raptors or the comparison between the territories and the control? This sentence is very unclear. Also, it's not clear if you are talking about the GLM, or simply linear models.

Response: Apologize, indeed here we used Binomial Linear Regression. Each was done separately for raptor species (either goshawk vs control or owl vs control) and we made 3 analyzes, separately for each bird diversity metric (richness, abundance, natura).

L225:Abundance of which species? For each species?

Response: an abundance of bird communities (winter or breeding), not an abundance of raptors. These diversity metrics are explained in the preceding chapter.

L233: These results should be shown in a table or graph to visualize them better. Table S2 does not show this clearly

Response: They are also summarized in box plots in Fig. 1. Table S2 just show all raw data.

L250: The majority

Response: Done

L250: were recorded instead of noted (change here and elsewhere)

Response: Done

L250: Is this difference significant? (did you test it?) If it is tested, you should provide the t-value, p-value, and model estimate.

Response: No, we did not test this as records of these protected bird species are too rare (not a sufficient number of data for statistics).

L252: In the treatment or the controls?

Response: Control, rewritten.

L253: Check the rule of capitals when you're mentioning the common name of the species

Response: We consistently use the small letters in all Latin species names with exception of the Ural owl as its name refers to Ural Mts. and it should be with a capital letter.

L256: Again, are these differences significant?

Response: No, we did not test this as records of these protected bird species are too rare (not a sufficient number of data for statistics).

L267: This is the main result of the manuscript, so should be earlier in the text. Also, you have to show the statistics (see comment L250)

L267: Seeing Figure 2, what you did is apparently a binomial GLM instead of a linear regression! Calculating the probability of occurrence... Please, state this clearly in the methods, if I'm right.

Response: Apologize, indeed here we used Binomial Linear Regression.

L272: Lack of statistics again

Response: Statistics are presented in Table S3 and S4.

L281: Do not mention the name of your variables (WOOD/NATURA), but their ecological significance (here and elsewhere). You should avoid the public having to come back to Methods to read what WOOD means.

Response: We have the opposite opinion. We introduced these names for variables in methods to facilitate the presentation of results. The method is a part of the article. If we describe all variables in the results again, this will greatly enlarge the whole paper as every time we will need to describe these variables (if not use simple shortcuts)

L290: Showed that in winter... not wintering period

Response: Done

Figure 1: It is very unclear what all the abbreviations mean. Be more clear.

Response: We replace some abbrevations with full names

L452: It is not so clear to me that what is relevant is the maintenance f the territories or the maintenance of the habitat conditions. 

Response: As it is stated in this sentence, we refer to territories of raptors as this work is on long-term territories of raptors, not specifically on habitat conditions in these territories.

Author Response File: Author Response.docx

Reviewer 3 Report

This is a useful paper that provides some interesting results on survey methods to understand how the presence of an apex predator promotes diverse bird communities. I believe that this paper is suitable for the journal and I recommend publication after some adjustments and edits. 

I appreciate this was an observational piece of work, but methods should still be ethically reviewed. There is no explanation or description of the ethical review process and this needs to be rectified to validate the methods chosen.

Please thoroughly review the paper for English spelling and grammar, as there are some clunky sentences within.

Please be clear on what was counted. The number of different species? The number of individuals of different species? It can be hard to follow...

The phrase "top raptors" makes them sound like birds on top of something else. Recommend "Apex predators, such as raptors..."

Please edit the title accordingly as per my comment above. 

Line 8: Is surrogate the correct word? Conservation of apex predators, such as raptors, is often used as the proxy for wider conservation initiatives?

Line 15: can you explain what this directive is? 

Line 26: Remove particularly from the start of the sentence.

Line 38: Bubo bubo is a specific species of eagle owl. Please be specific with the common name.

Line 136 does not make sense. What is being counted in the sites? "The number of wintering and/or breeding birds (depending on time of year) were counted within each goshawk, or owl, or unoccupied site."

What is a tree repeat? Please define.

Line 144: Please give the times of dawn and dusk according to changes in the meteorological year. I.e. how did you know when was dawn and when was dusk?

Please define what good weather was.

Line 230: Please can you identify, in your data analysis section, what statistical testing was run with each programme? 

Please provide the predictors and outcomes variables for all models (the GLMs and the linear regression). Tell the reader specifically what the predictor variables were for each outcome variable. 

Line 216: Please explain in more detail what was tested with the univariate statistics and the Wald's test. 

Line 239: Please do not start a sentence with a numeric. Either write this out or change the sentence, e.g. "A total of 2643 individual..."

Line 246: Please provide further explanation of the Bird Directive and what it means for your categorisation of species in the method. 

Please place figure captions underneath the figure they link to.

Figure 2: Please provide units for bird abundance and explain what Probably of occurrence is in the figure caption.

Figure 3 is very hard to read. Can it be enlarged? The analysis presented in Figure 3 needs clearer description in the data analysis section of the Method so that others can repeat this testing. 

The results section is very hard to follow overall. You should lay out the results section by aim, and include descriptive and inferential analysis alongside each other. Currently there is a list of figures and then a list of tables, but there is no description of what the figures and tables show to the reader that explain what was tested and why. Please re-work this section of the paper to make it clearer.

Table 4 needs further context. What are the significant outputs provided? How does the reader read this table to work out what the key results are?

Line 347: Please remember the word "prove". We support or refute our hypotheses in science, we never prove.

Line 396: What do you mean by a proper understanding? Please add context.

Line 398: Please remove prove

Your conclusion is very long and contains some discursive elements. Please can you reduce the conclusion to focus on your overall findings of most importance and what they mean (their wider application). Please remove other more discursive areas of the conclusion to the Discussion where they can be properly evaluated. 

Author Response

We are grateful to both Reviewers for their valuable comments and suggestions, which enable us to improve our manuscript and allow for the avoidance of errors and biases in our study.

Below we provided answers or justifications to all comments.  

Comments and Suggestions for Authors

This is a useful paper that provides some interesting results on survey methods to understand how the presence of an apex predator promotes diverse bird communities. I believe that this paper is suitable for the journal and I recommend publication after some adjustments and edits. 

I appreciate this was an observational piece of work, but methods should still be ethically reviewed. There is no explanation or description of the ethical review process and this needs to be rectified to validate the methods chosen.

Response: We are not sure if we understand properly this comment. If this comment refers to necessary permissions for doing research, we inform that bird inventories were made in non-invasive way (birds were not captured, not disturbed), therefore no special permissions were required. We added this statement in the method section as well as in separate Ethic statement. The only required permission was for entrance to small nature reserves.

Please thoroughly review the paper for English spelling and grammar, as there are some clunky sentences within.

Response: We ordered proofreading in commercial company by a native speaker (Dr Emily Kane, ordered to company http://specjalistyczne.biz/index.php?p=2_2_About-Us ). English is corrected in current version.

Please be clear on what was counted. The number of different species? The number of individuals of different species? It can be hard to follow...

Response: In the Bird inventory we clearly stated that “Birds were noted (species and number of individuals for each species)”, and next “For analyses, the maximum number of birds of particular species observed on the transect were calculated to avoid multiple use of birds sedentary in forests”. Also further “In each point birds were noted within a radius of 100 m, by hearing territorial and con-tact voices (mostly signs of territorial male or drumming in case of woodpeckers). Also all seen birds displaying territorial, mating or breeding behavior were noted.” and “For analyses, the maximal number of territorial birds of particular species noted in two points localized within a particular site, were calculated (summarized) from data collected during three repeats.”

In our opinion these description clearly inform that we counted all birds (all individuals for particular species), but numbers used for wintering and breeding birds were differently determined. For wintering birds we used maximal number of individuals (all) from 3 counts, and for breeding birds maximal number of territorial birds (mostly signing males) from 3 counts.

The phrase "top raptors" makes them sound like birds on top of something else. Recommend "Apex predators, such as raptors..."

Please edit the title accordingly as per my comment above. 

Response: We consistently change “top raptors” into “apex predators” in the text, but we left “raptors” were we refer to diurnal birds of prey or owls (not specifying if top or subordinate taxa).

Line 8: Is surrogate the correct word? Conservation of apex predators, such as raptors, is often used as the proxy for wider conservation initiatives?

Response: In our opinion – yes, it is correct word here.

Surrogate species (=Biological indicator species) are species “used to represent other species or aspects of the environment to attain a conservation objective” (Caro, T. 2010. Conservation by proxy indicator, umbrella, keystone, flagship, and others surrogate species. Island Press, Washington, D.C., USA.)

Line 15: can you explain what this directive is? 

Response: We clarified this. It is Directive 2009/147/EC of the European Parliament and of the Council of 30 November 2009 on the conservation of wild birds.

Line 26: Remove particularly from the start of the sentence.

Response: Done

Line 38: Bubo bubo is a specific species of eagle owl. Please be specific with the common name.

Response: Done

Line 136 does not make sense. What is being counted in the sites? "The number of wintering and/or breeding birds (depending on time of year) were counted within each goshawk, or owl, or unoccupied site."

Response: You are right. We rewrite this sentence accordingly.

What is a tree repeat? Please define.

Response: This mean just that we counted birds three times. Rewritten.

Line 144: Please give the times of dawn and dusk according to changes in the meteorological year. I.e. how did you know when was dawn and when was dusk?

Response: We added approximate hours, however we are surprised by this question as dusk and dawn hours could be simply known just from field observations or from information available in internet for particular geographic position). We simply avoided parts of the day when visibility was low due to lower insolation. Also lower temperature in dusk and dawn reduced birds activity so we simply adjusted time of counts to the highest period of activity.

Please define what good weather was.

Response: This was already explained in brackets: (without snow, rain and stronger winds, but regardless of temperature)

Line 230: Please can you identify, in your data analysis section, what statistical testing was run with each programme? 

Response: Done

Please provide the predictors and outcomes variables for all models (the GLMs and the linear regression). Tell the reader specifically what the predictor variables were for each outcome variable. 

Response: Apologize, but we do not understand this question as these information are available in methods. For LR we used presence of either goshawks or owls against bird diversity metrics (species richness and abundance), each analyzed separately for winter and breeding data. For GLM we used as predictors presence of goshawks and owls, and wood type, and as explained variables – bird diversity metrics (as above).

Line 216: Please explain in more detail what was tested with the univariate statistics and the Wald's test. 

Response: We tested differences of all measured environmental variables (defined in previous chapter – 2.4.) and presence of raptor (either goshawk or owl) on 3 type of bird diversity metrics: richness, abundance and Natura (species protected under N2000). We calculated this statistic separately for goshawk and owl plots and also independently for winter and breeding periods.

This description was rewritten.

Line 239: Please do not start a sentence with a numeric. Either write this out or change the sentence, e.g. "A total of 2643 individual..."

Response: Done

Line 246: Please provide further explanation of the Bird Directive and what it means for your categorisation of species in the method. 

Response: Apologize, but is explained in Method (2.4. Bird indices). We added there also justification for use of this metric. 

Please place figure captions underneath the figure they link to.

Response: Done

Figure 2: Please provide units for bird abundance and explain what Probably of occurrence is in the figure caption.

Response: Done

Figure 3 is very hard to read. Can it be enlarged? The analysis presented in Figure 3 needs clearer description in the data analysis section of the Method so that others can repeat this testing. 

Response: We enlarged all names and numbers on this figure. To make it clearer we shortened species names. These figures are of high quality in original files, but for submission it was necessary to insert them into one doc file, what greatly reduced their size and quality.

Moreover, we extended description of CCA in Methods and caption to the figure.

The results section is very hard to follow overall. You should lay out the results section by aim, and include descriptive and inferential analysis alongside each other. Currently there is a list of figures and then a list of tables, but there is no description of what the figures and tables show to the reader that explain what was tested and why. Please re-work this section of the paper to make it clearer.

Response: We disagree with this opinion. First, figures and tables are presented after description of results in separate subchapter (3.5) as it is required by the journal to show results in this way. Second, all figures and tables are referred in the description of results (3.1-3.4) and there are explanations to tables and figures.

Previously we arranged results according to order of methods in use. Particular aims of this study are solved by various methods, therefore it is not possible to simply present results according to aims (as in this way some results need to be repeated). We propose different organization of result presentation following sequence: presentation of general bird diversity metrics, comparison of bird diversity metrics in respect to raptor territories, description of differences in environmental variables in raptor territories, bird diversity metrics in relation to environment and raptor territories.

Table 4 needs further context. What are the significant outputs provided? How does the reader read this table to work out what the key results are?

Response: We are a bit confused here as explanation of results to this table is presented in chapter 3.4. Indeed, this table is large and not easy to read as it summarized lists of best models constructed for several sets of variables, therefore it would be hard to reduce its size (it already includes only models with ∆AIC < 2, so “the best ones”). We can simply move this table to supplementary files.

Line 347: Please remember the word "prove". We support or refute our hypotheses in science, we never prove.

Response: Thank you. Rewritten.

Line 396: What do you mean by a proper understanding? Please add context.

Response: We remove word “proper” as there are no similar studies on long-lasting territories, therefore this research is the first which allows for understanding of  the role of long-lasting territories of top predators for biodiversity protection.

Line 398: Please remove prove

Response: Done

Your conclusion is very long and contains some discursive elements. Please can you reduce the conclusion to focus on your overall findings of most importance and what they mean (their wider application). Please remove other more discursive areas of the conclusion to the Discussion where they can be properly evaluated. 

Response: We intensively edited discussion and conclusions, deleting some less necessary parts and focusing discussion of conservation/management implications (therefore we would like to name this chapter as: Implications for nature conservation and forest management).

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I really appreciate the effort that the authors made with this new version of the manuscript. I only have some minor comments that can improve the clarity and replicability of the article.

The main problem that still exists is that the authors did not made case to the suggestion of avoiding the use of abbreviations, which make the text much harder to follow. Authors must know that the text should be fluent enough, to allow reading the article from the beginning to the end, without needing to comeback all the time to "Methods" to understand how the variables were called (this applies to this manuscript, or to any other manuscript). I think the editor will have to decide if this is acceptable for the journal, if is not changed in a new version.

Also, authors replied properly several of my comments in the letter, but without including them in the manuscript. For example, this wasn't added to the text:

Comment 1: Did you check if the data achieve the ANOVA assumptions?

Response1: Yes. All analyzed data (bird diversity metrics) have a normal population distribution, similar variance, and are independent (within each analysis).

Please, check that everything is added to a new version of the manuscript. The manuscript needs to be as clear as possible, and provide the information that makes it replicable by other researchers in other places.