Evolution and Paradigm Shift in Forest Health Research: A Review on of Global Trends and Knowledge Gaps

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

General comments: I found this to be a well-written and thought-provoking examination of the evolution of forest health research. I believe that there will be interest among the journal's readers. I have a lot of specific comments (below) that I hope will lead to a better final manuscript. Most of them are either minor editorial suggestions or touch on details that could be made clearer. In terms of an overarching comment, I think that it's worth delving into the systemic factors involved in defining the pool of forest health research that you analyzed in this manuscript. For example, a huge share of forest health research comes from the USA, where the major agents and issues of concern – i.e., those that drive research investment – tend to be driven by publicly prominent events (e.g., emerald ash borer's arrival and spread from the Great Lakes area, mountain pine beetle outbreaks in many parts of the western USA and Canada, especially in 2008-2012, megadrought events in the Southwestern USA and California over the last 20 years). In other words, forest health research has at times tended to be reactive (i.e., in response to pronounced and acute impacts noticed by policymakers and the public) rather than proactive (i.e., focused on subtle yet potentially more impactful forest health threats). This could also be seen as a tendency of research to deal with immediately obvious (certain) rather than poorly resolved (uncertain) drivers of forest health; the latter includes the many interactions that we’re only starting to tease apart.

Together, Figures 4 and 5 are compelling. In my mind, they capture the evolution of forest health science (and the forest health "conversation") over the last few decades. They could maybe use some minor adjustments (see specific comments below) but generally they’re quite illustrative. Figure 6 is a nice complement to them. Nevertheless – and I hint at this in my specific comments, too – the temporal progression of forest health research has happened within the context of broader forest health monitoring, management and response mechanisms (or processes or systems, if you prefer). For instance, in the USA, the generation of forest health professionals (at federal, state, and local levels) that emerged immediately after the period of interest in acid rain, air pollution, forest decline, etc., were typically trained as entomologists and pathologists. Undoubtedly, there was a need for this expertise – and there still is – but it has also shaped the direction of forest health research in the USA for the last couple of decades, which has always relied on relationships with practitioners to determine what’s important and to help identify critical information gaps. I suppose this is just an elaboration of my first comment about discussing systemic factors. You did a good job discussing the biases you uncovered (e.g., temperate and boreal vs. tropical forest coverage, Global North vs. Global South), so maybe all you need to do is expand on the linkages between forest health research and forest health practice, especially the degree to which they inform each other. (A related consideration is that “forest health” may not be a label that researchers readily ascribe to themselves, for whatever reason.)

Specific comments:

Line 24 – insert “for” after “need”.

Line 49 – replace “including” with “to include”.

Line 60 – delete “on” after “impacting”.

Line 79 – replace “scientist’” with “scientists’”.

Lines 89-91 – suggested rewrite: “Increasingly, measurements derived from remote sensors deployed on satellites and unmanned aerial vehicles have been applied for the detection of non-visible phenomena in forests [22-24].”

Line 103 – maybe replace “trend” with “development” or “advancement”.

Line 105 – replace “on” with “for”.

Line 108 – replace “of relevance” with “relevant”.

Line 121 – replace “contributed” with “endeavoured”.

Line 134 (Figure 1) – In one of the left-hand column boxes, it looks like it says “exporta”. In boxes in the center and right-hand columns, you need to correct the spelling of “co-authorship” as well as “co-occurrence”. With respect to the right-hand column box, I’m unfamiliar with the term “fractorial”. Did you mean something else?

Line 138 – replace “database was” with “databases were”.

Line 139 – (“grey literature”) – In my mind, grey literature is conventional, just not necessarily refereed in the same manner as formal scientific publications (i.e., journal articles and monographs). Perhaps most importantly, it is rarely rejected outright, although it still may go through many cycles of review and revision before publication, as grey literature typically must conform to some set of content and editorial standards dictated by the authors’ agency or institution.

Lines 143-144 – it is probably worth noting that the full databases for Scopus and Web of Science only extend back so far -- maybe the 1960s? There are older records, to be sure, but I doubt coverage is comprehensive. In any case, you should check on the database start dates.

Line 209 – (“The top 4 authors stand out…”) – can you quantify how many of the publications are co-authored between them? You may want to avoid inflating the individual importance of an author if a sizeable share of their work involves collaboration with other top authors. In Bibliometrix, I think you can get at this by looking at the authors’ fractionalized frequency distributions. It may not change things much, but I would at least review the fractionalized results if you haven’t already. I know that McDowell and Anderegg have co-authored several publications, for example.

Lines 210-211 – like many of the other top authors, Camarero is perhaps best known for work on drought in forests.

Line 221 – replace “last” with “recent”.

Lines 228-231 – I'm not sure what to make of these other clusters. For instance, New Zealand seems like an odd inclusion with Eastern and Northern Europe. Do you have any ideas? Is it because of cultural similarities in how these countries consider and address forest health?

Line 272 – (“already little studied”) – meaning little studied as of yet?

Line 273 (Figure 3) – “mountain pine beetle” and “dendroctonus ponderosae” are synonymous; this is probably worth mentioning. Similarly, “picea abies” and “norway spruce” are synonymous.

Line 273 (Figure 3) – regarding the left-most cluster, is "invasive species" really a niche theme? Is it being pulled this way because of the relatively recent prominence of emerald ash borer as a forest pest in the USA, Canada, and now Europe? A huge aspect of forest health monitoring (and hence forest health research) in North America, at least over the last 20 years, has been surveillance and response to non-native forest pests and pathogens, of which the emerald ash borer is perhaps the most high-profile since it’s had a significant impact in urban and suburban forests. Part of the issue, I think, is the scale at which non-native pests and pathogens operate – in isolated locations or along an advancing front – so the research is fundamentally different than it would be for something like climate change and/or drought. Biological control is at least a possibility with a non-native pest or pathogen, and pursuing that possibility requires some directed research investment. At any rate, I would caution you about your labeling in this word grouping map. Suggesting that invasive species fall very low along the “relevance” axis is likely to invite criticism from some forest health scientists and practitioners.

Line 285 – insert “the” before “late”.

Line 296 (Figure 4) – many of the labels are truncated, which detracts from the figure somewhat.

Line 313 (Figure 5) – aren't "decline", "forest decline" and "declining" essentially synonymous? If you combined their proportions, it would paint a somewhat different picture of the shift among topics (5a) over time.

Line 313 (Figure 5) – in 5a (topics), "tree mortality" overtakes "decline" around 2008, which makes sense – that's about when "climate" starts to ascend quickly as a prominent driver (5b) and when "modeling" overtakes "ecophysiology" among methods (5d). I bring this up because the subdomains aren’t independent, which is something you maybe touch upon in the Discussion but could use more emphasis. For example, the steady rise in GIS/RS methods coincides with a rise in modeling methods (5d), and both almost certainly played a part in the increase prevalence of “climate” among driver (5b) keywords. They probably have made it easier to analyze pest and diseases as forest health drivers, too (even ignoring the fact that most of the drivers interact with each other). Moreover, GIS/RS methods likely contributed to the increasing prevalence of “mortality” among the conditions (5c), since large-scale mortality is relatively easy to quantify and analyze remotely. None of this refutes what you have said in your manuscript, but I think it’s important to remind readers of key relationships between the subdomains.

Line 362 – delete “the” after “from”.

Line 383 – “peripheral” instead of “peripherical”.

Line 404 (Figure 6) – this is a good set of word clouds. I could quibble with the chronological list of events on the right, but I think you captured the main elements of interest.

Line 519 – “increase” instead of “increased”.

Line 543 – delete “the” before “molecular”.

Lines 559-561 – I have an alternate explanation for the lack of attention to EBV in forest health research: many forest health experts are tasked with and therefore most interested in maintaining forests (i.e., forest cover and forest land use) in general. In short, they're foresters / practitioners first and forest ecologists second. Furthermore, and perhaps most especially in North America, a small share of the professionals who deal with forest health are in research, so it takes some time for conceptual frameworks that arise from research to percolate throughout the larger forest health community. Going back to Pereira et al. 2013, the examples of EBVs that they presented do come up among forest health researchers (and practitioners), but there remains little agree about the most meaningful forest health indicators, whether those indicators are EBVs or not. Sometimes, embracing changes to long-standing conceptual frameworks is like trying to turn the Titanic, unfortunately.

Line 560 – replace “be very benefited” with “benefit”.

Line 564 – see comment above. People working on forest health in North America (and elsewhere) do talk about ecosystem services, but overwhelmingly the conversation has become about one thing in particular: carbon. I think this is because it can be quantified in both ecological and economic terms and, of course, is highly relevant in the face of a warming climate. It goes without saying that there are many other important ecosystem services, but carbon storage/sequestration seems to be top of mind for a lot of forest scientists and policymakers. On a related note, the concept of forests as habitat didn’t really come up in your results, either, despite the fact that wildlife habitat is an ecosystem service of forests that policymakers and the public seem to value rather highly.

Supplementary Material

Figure S1 – what is the difference between a “database” and a “register”? Please clarify. Also, replace “form” with “from” in the figure caption.

I noticed some misspellings in Table S1 (e.g., “termal stress”, “quenching coeficients”, “multiciteria decision-analysis”, “time serie”). Some of them I can understand, such as having appearances of “ips tpographus” as well as “ips typographus” (the correct spelling), but for others I’m not sure whether they had any bearing on subsequent analyses. Can you check through the table for errors?

Figure S2 – instead of a linear function, why not try to fit the data with an exponential function since you mentioned this in the main text?

Figure S3 – it's noteworthy that the most relevant authors are all focused heavily on climate drivers of forest health, including drought (and hotter drought under climate change). The highest-ranked author who works primarily on biotic agents is Fettig (a forest entomologist). Regardless, I won't disagree that these are important names in forest health research over the last few decades. Maybe it’s worth discussing the factors that might enable the primarily climate-focused authors to be more prolific, e.g., their research may be less dependent on time- and labor-intensive field and lab work. This isn’t a major concern, but I don’t think you can fully discount how much the contributions of a few individual researchers can shape the history of a research topic/discipline.

Comments on the Quality of English Language

The English is of high quality, with only some relatively minor edits required (see specific comments).

Author Response

Comments - Additional review 3

C3/ General comments:

I found this to be a well-written and thought-provoking examination of the evolution of forest health research. I believe that there will be interest among the journal's readers. I have a lot of specific comments (below) that I hope will lead to a better final manuscript. Most of them are either minor editorial suggestions or touch on details that could be made clearer. In terms of an overarching comment, I think that it's worth delving into the systemic factors involved in defining the pool of forest health research that you analyzed in this manuscript. For example, a huge share of forest health research comes from the USA, where the major agents and issues of concern – i.e., those that drive research investment – tend to be driven by publicly prominent events (e.g., emerald ash borer's arrival and spread from the Great Lakes area, mountain pine beetle outbreaks in many parts of the western USA and Canada, especially in 2008-2012, megadrought events in the Southwestern USA and California over the last 20 years). In other words, forest health research has at times tended to be reactive (i.e., in response to pronounced and acute impacts noticed by policymakers and the public) rather than proactive (i.e., focused on subtle yet potentially more impactful forest health threats). This could also be seen as a tendency of research to deal with immediately obvious (certain) rather than poorly resolved (uncertain) drivers of forest health; the latter includes the many interactions that we’re only starting to tease apart.

Together, Figures 4 and 5 are compelling. In my mind, they capture the evolution of forest health science (and the forest health "conversation") over the last few decades. They could maybe use some minor adjustments (see specific comments below) but generally they’re quite illustrative. Figure 6 is a nice complement to them. Nevertheless – and I hint at this in my specific comments, too – the temporal progression of forest health research has happened within the context of broader forest health monitoring, management and response mechanisms (or processes or systems, if you prefer). For instance, in the USA, the generation of forest health professionals (at federal, state, and local levels) that emerged immediately after the period of interest in acid rain, air pollution, forest decline, etc., were typically trained as entomologists and pathologists. Undoubtedly, there was a need for this expertise – and there still is – but it has also shaped the direction of forest health research in the USA for the last couple of decades, which has always relied on relationships with practitioners to determine what’s important and to help identify critical information gaps. I suppose this is just an elaboration of my first comment about discussing systemic factors. You did a good job discussing the biases you uncovered (e.g., temperate and boreal vs. tropical forest coverage, Global North vs. Global South), so maybe all you need to do is expand on the linkages between forest health research and forest health practice, especially the degree to which they inform each other. (A related consideration is that “forest health” may not be a label that researchers readily ascribe to themselves, for whatever reason.)

R3/ We very much appreciate your constructive view of the document. Specifically, we have included the research-practitioner link in the first part of the discussion (L404).  We greatly appreciate all the specific comments and have tried to acknowledge them in detail in order to improve the quality of the paper.

C3/ Specific comments:

Line 24 – insert “for” after “need”.

R3/ L 24, the reviewer is right, "for" was added.

C3/ Line 49 – replace “including” with “to include”.

R3/ L 45, the reviewer is right, it was changed to "to include".

C3/ Line 60 – delete “on” after “impacting”.

R3/ L56, the reviewer is right, it was deleted “on” after “impacting”.

C3/ Line 79 – replace “scientist’” with “scientists’”.

R3/ L 75, the reviewer is right, it was replaced “scientists”.

C3/ Lines 89-91 – suggested rewrite: “Increasingly, measurements derived from remote sensors deployed on satellites and unmanned aerial vehicles have been applied for the detection of non-visible phenomena in forests [22-24].”

R3/ L 85-87, indeed, the reviewer's suggested rewrite improves the text. The sentence was replaced with the indicated change.

C3/ Line 103 – maybe replace “trend” with “development” or “advancement”.

R3/ L 100, at the suggestion of the reviewer in the improvement of the text, the word was changed to "development".

C3/ Line 105 – replace “on” with “for”.

R3/ L 101, the reviewer is right, it was replaced “for”.

C3/ Line 108 – replace “of relevance” with “relevant”.

R3/ L 105, the reviewer is right, it was replaced “relevant”.

C3/ Line 121 – replace “contributed” with “endeavoured”.

R3/ This sentence was removed in the process of unifying the text of the objectives, due to the other reviewer's suggestion for improvement.

C3/ Line 134 (Figure 1) – In one of the left-hand column boxes, it looks like it says “exporta”. In boxes in the center and right-hand columns, you need to correct the spelling of “co-authorship” as well as “co-occurrence”. With respect to the right-hand column box, I’m unfamiliar with the term “fractorial”. Did you mean something else?

R3/ L 123, we fully agree with the reviewer, this figure has been replaced by adding the suggested changes and adjustments.

C3/ Line 138 – replace “database was” with “databases were”.

R3/ L 128-129, the reviewer is right, it was replaced “databases were”.

C3/ Line 139 – (“grey literature”) – In my mind, grey literature is conventional, just not necessarily refereed in the same manner as formal scientific publications (i.e., journal articles and monographs). Perhaps most importantly, it is rarely rejected outright, although it still may go through many cycles of review and revision before publication, as grey literature typically must conform to some set of content and editorial standards dictated by the authors’ agency or institution.

R3/ L 130, the reviewer is right, in his point about mainstream literature or not, we leave the term "grey literature" as the most appropriate.

C3/ Lines 143-144 – it is probably worth noting that the full databases for Scopus and Web of Science only extend back so far -- maybe the 1960s? There are older records, to be sure, but I doubt coverage is comprehensive. In any case, you should check on the database start dates.

R3/ L 135-137, we have now indicated in methods the time range of each database. With WoS from 1934 and Scopus from 1963.

C3/ Line 209 – (“The top 4 authors stand out…”) – can you quantify how many of the publications are co-authored between them? You may want to avoid inflating the individual importance of an author if a sizeable share of their work involves collaboration with other top authors. In Bibliometrix, I think you can get at this by looking at the authors’ fractionalized frequency distributions. It may not change things much, but I would at least review the fractionalized results if you haven’t already. I know that McDowell and Anderegg have co-authored several publications, for example.

R3/ L 206, this point raised by the reviewer had been reviewed by the authors. We considered it necessary to choose one of the two ways of showing this aspect by deciding on the absolute number of publications of each author. It is true, however, that it is possible to combine the two aspects, as suggested by the reviewer. We provide this data and comparison in the same figure S3 of the supplementary material.

C3/ Lines 210-211 – like many of the other top authors, Camarero is perhaps best known for work on drought in forests.

R3/ L 208 Yes, Camarero's work on drought issues is indeed remarkable. Therefore, we have included in the text: “Camarero J.J., uses growth ring analysis and remote sensing to study the interaction between forests and drought”

C3/ Line 221 – replace “last” with “recent”.

R3/ L 219, the reviewer is right, it was replaced “recent”.

C3/ Lines 228-231 – I'm not sure what to make of these other clusters. For instance, New Zealand seems like an odd inclusion with Eastern and Northern Europe. Do you have any ideas? Is it because of cultural similarities in how these countries consider and address forest health?

R3/ L 230-231, we agree that the relation is indeed strange. We think that the main possible reasons are (among others) the connections between the countries in terms of similar climatic and ecological conditions and the productive forest management. We have now included this speculation in L230.

C3/ Line 272 – (“already little studied”) – meaning little studied as of yet?

R3/ L 270, This means that it has been studied, but with little intensity. We modified the wording at the end of that sentence to improve that dimension and idea.

C3/ Line 273 (Figure 3) – “mountain pine beetle” and “dendroctonus ponderosae” are synonymous; this is probably worth mentioning. Similarly, “picea abies” and “norway spruce” are synonymous.

R3/ L 260-261, 266-268, in agreement with the reviewer, we did indeed highlight these interesting aspects in the text.  

C3/ Line 273 (Figure 3) – regarding the left-most cluster, is "invasive species" really a niche theme? Is it being pulled this way because of the relatively recent prominence of emerald ash borer as a forest pest in the USA, Canada, and now Europe? A huge aspect of forest health monitoring (and hence forest health research) in North America, at least over the last 20 years, has been surveillance and response to non-native forest pests and pathogens, of which the emerald ash borer is perhaps the most high-profile since it’s had a significant impact in urban and suburban forests. Part of the issue, I think, is the scale at which non-native pests and pathogens operate – in isolated locations or along an advancing front – so the research is fundamentally different than it would be for something like climate change and/or drought. Biological control is at least a possibility with a non-native pest or pathogen, and pursuing that possibility requires some directed research investment. At any rate, I would caution you about your labeling in this word grouping map. Suggesting that invasive species fall very low along the “relevance” axis is likely to invite criticism from some forest health scientists and practitioners.

R3/L 271. We also believe that the invasive species aspect has probably a conservative score in the analysis. There are several explanations, including the scale the reviewer mention, but also 1) that we consider a global scale where maybe invasive species in forest ecosystems are not so important than in the USA (e.g. Spain) and 2) that the research on invasive species usually do not mention the concept of forest health (or other terms used in our query).

C3/ Line 285 – insert “the” before “late”.

R3/ L 283, the reviewer is right, we incorporated "the" for grammatical improvement.

C3/ Line 296 (Figure 4) – many of the labels are truncated, which detracts from the figure somewhat.

R3/ Yes, we agree with the reviewer, it has been very difficult to find the best representation. We gave a lot of thought to this figure, and this was the best combination to show as many relevant concepts as possible. If we eliminated words, it is very simple and disconnected or decontextualised one term from another, without the possibility of observing connections. On the other hand, if we forced all the words to appear, it was impossible to read. With the facilities of the VOSviewer graph software, this version is the most balanced.

C3/ Line 313 (Figure 5) – aren't "decline", "forest decline" and "declining" essentially synonymous? If you combined their proportions, it would paint a somewhat different picture of the shift among topics (5a) over time.

R3/ L 311, we agree with the reviewer, there has indeed been a version error in the figure. This aspect has been corrected by changing the figure.

C3/ Line 313 (Figure 5) – in 5a (topics), "tree mortality" overtakes "decline" around 2008, which makes sense – that's about when "climate" starts to ascend quickly as a prominent driver (5b) and when "modeling" overtakes "ecophysiology" among methods (5d). I bring this up because the subdomains aren’t independent, which is something you maybe touch upon in the Discussion but could use more emphasis. For example, the steady rise in GIS/RS methods coincides with a rise in modeling methods (5d), and both almost certainly played a part in the increase prevalence of “climate” among driver (5b) keywords. They probably have made it easier to analyze pest and diseases as forest health drivers, too (even ignoring the fact that most of the drivers interact with each other). Moreover, GIS/RS methods likely contributed to the increasing prevalence of “mortality” among the conditions (5c), since large-scale mortality is relatively easy to quantify and analyze remotely. None of this refutes what you have said in your manuscript, but I think it’s important to remind readers of key relationships between the subdomains.

R3/We greatly appreciate this comment about interaction across subdomains. We have now included a specific text about this interaction in L550-553: “All these modeling and RS methods have particularly increased in the last two decades becoming especially useful to analyze complex drivers such as climate, pest and diseases, which also showed a similar steady increase (Fig. 5) reflecting the interaction among subdomains in forest health research.”

C3/ Line 362 – delete “the” after “from”.

R3/ L 360, the reviewer is right, it was deleted “the” after “from”.

C3/ Line 383 – “peripheral” instead of “peripherical”.

R3/ L 381, the reviewer is right, it was replaced “peripheral”.

C3/ Line 404 (Figure 6) – this is a good set of word clouds. I could quibble with the chronological list of events on the right, but I think you captured the main elements of interest.

R3/ L 403, we agree with the reviewer and thank him for his vision, to summarise graphically and temporally this relating events was not an easy task.

C3/ Line 519 – “increase” instead of “increased”.

R3/ L 518, the reviewer is right, it was replaced “increase”.

C3/ Line 543 – delete “the” before “molecular”.

R3/ L 542, the reviewer is right, it was deleted “the” before “modelcular”.

C3/ Lines 559-561 – I have an alternate explanation for the lack of attention to EBV in forest health research: many forest health experts are tasked with and therefore most interested in maintaining forests (i.e., forest cover and forest land use) in general. In short, they're foresters / practitioners first and forest ecologists second. Furthermore, and perhaps most especially in North America, a small share of the professionals who deal with forest health are in research, so it takes some time for conceptual frameworks that arise from research to percolate throughout the larger forest health community. Going back to Pereira et al. 2013, the examples of EBVs that they presented do come up among forest health researchers (and practitioners), but there remains little agree about the most meaningful forest health indicators, whether those indicators are EBVs or not. Sometimes, embracing changes to long-standing conceptual frameworks is like trying to turn the Titanic, unfortunately.

R3/ This is indeed a very interesting explanation. We have tried to include it in the discussion in L566-569: “Novel conceptual frameworks might take time to percolate into a research field, particularly as forest health has greatly focused on maintaining forest cover and biomass rather than on the plenty services they provide”.

C3/ Line 560 – replace “be very benefited” with “benefit”.

R3/ L 560, the reviewer is right, it was replaced “benefit”.

C3/ Line 564 – see comment above. People working on forest health in North America (and elsewhere) do talk about ecosystem services, but overwhelmingly the conversation has become about one thing in particular: carbon. I think this is because it can be quantified in both ecological and economic terms and, of course, is highly relevant in the face of a warming climate. It goes without saying that there are many other important ecosystem services, but carbon storage/sequestration seems to be top of mind for a lot of forest scientists and policymakers. On a related note, the concept of forests as habitat didn’t really come up in your results, either, despite the fact that wildlife habitat is an ecosystem service of forests that policymakers and the public seem to value rather highly.

R3/ thank you for the comment and suggestion. We have included this rationality in L566-569 (see response above).

C3/ Supplementary Material

Figure S1 – what is the difference between a “database” and a “register”? Please clarify. Also, replace “form” with “from” in the figure caption.

R3/ A "database" is the repository to query, and a "record" is one of the documentary results of the database query. We have pointed out this issue in the PRISMA explanatory text.  

C3/ I noticed some misspellings in Table S1 (e.g., “termal stress”, “quenching coeficients”, “multiciteria decision-analysis”, “time serie”). Some of them I can understand, such as having appearances of “ips tpographus” as well as “ips typographus” (the correct spelling), but for others I’m not sure whether they had any bearing on subsequent analyses. Can you check through the table for errors?

R3/ The reviewer is right, we have reviewed and adjusted the table.

C3/ Figure S2 – instead of a linear function, why not try to fit the data with an exponential function since you mentioned this in the main text?

R3/ The reviewer's suggestion was implemented, the function explaining the trend of the data was changed to an exponential function and the equation was recalculated.

C3/ Figure S3 – it's noteworthy that the most relevant authors are all focused heavily on climate drivers of forest health, including drought (and hotter drought under climate change). The highest-ranked author who works primarily on biotic agents is Fettig (a forest entomologist). Regardless, I won't disagree that these are important names in forest health research over the last few decades. Maybe it’s worth discussing the factors that might enable the primarily climate-focused authors to be more prolific, e.g., their research may be less dependent on time- and labor-intensive field and lab work. This isn’t a major concern, but I don’t think you can fully discount how much the contributions of a few individual researchers can shape the history of a research topic/discipline.

R3/ Thank you very much for the comment and definitely we agree that the pattern found is possibly linked to the variability in complexity of different research fields. In relation to these results, we decided to leave them in the supplementary and focus our discussion on the temporal shifts and trends which we believe will be more interesting for the readers.

Reviewer 2 Report

Comments and Suggestions for Authors

Evolution and paradigm shift in forest health research: A review on global trends and knowledge gaps by Acosta-Muñoz et al is a well written manuscript with an interesting topic.

Introduction: 30-36: summarize this paragraph and paste it into the next one. Otherwise, it is too general and the ideas are repeated.

However, this section is well written, the reader is introduced to the chosen topic.

The aims of the study should be better emphasized but not scattered in the two paragraphs 106-116 and 117-123.

Materials and Methods: This section describes how the authors conducted their research. I appreciate that the study has been carried out rigorously and can be reproduced.

147: On what basis were the most relevant articles chosen? 

Results: all relevant results were reported concisely and objectively, in a logical order.

213, 214: at least Veblen, 1983 and Allen and Breshears, 2010 are bibliographical references

Discussion: It focused on explaining and evaluating what the authors found, showing how it relates to the literature review and making an argument in support of the conclusion. 

Conclusions: 614-618: the concept of One Health is not new but has reemerged as a concept to both better understand the triple threats to health and to better address these contemporary challenges using new approaches.

Comments on the Quality of English Language

Moderate editing of English language required.

Author Response

Comments - Reviewer 1

C1/ Evolution and paradigm shift in forest health research: A review on global trends and knowledge gaps by Acosta-Muñoz et al is a well written manuscript with an interesting topic.

Introduction: 30-36: summarize this paragraph and paste it into the next one. Otherwise, it is too general and the ideas are repeated.

However, this section is well written, the reader is introduced to the chosen topic.

R1/ We appreciate the reviewer for checking the manuscript. Specially, pointing out mistakes and areas for improvement in the document.

L 30-38 a summary of the original first paragraph has been made and merged with the second paragraph to be more concise, as recommended by the reviewer.

C1/ The aims of the study should be better emphasized but not scattered in the two paragraphs 106-116 and 117-123.

R1/ L 102-112 the two paragraphs have been reduced to one paragraph, avoiding the dispersion of the objectives to make them clear and concise, as recommended by the reviewer.

C1/ Materials and Methods: This section describes how the authors conducted their research. I appreciate that the study has been carried out rigorously and can be reproduced.

147: On what basis were the most relevant articles chosen? 

R1/ L 140 the criteria considered for the selection of the most relevant articles was the WoS impact value. This now has been specified in L140.

C1/ Results: all relevant results were reported concisely and objectively, in a logical order.

213, 214: at least Veblen, 1983 and Allen and Breshears, 2010 are bibliographical references

R1/ L 208-211 the text has been adjusted and the bibliographical references of the reviews indicated have been incorporated.

C1/ Discussion: It focused on explaining and evaluating what the authors found, showing how it relates to the literature review and making an argument in support of the conclusion. 

Conclusions: 614-618: the concept of One Health is not new but has reemerged as a concept to both better understand the triple threats to health and to better address these contemporary challenges using new approaches.

R1/ L 622 – 627 We appreciate the reviewer's suggestion. We agree that the concept of One Health is not new, but as you indicate it is a concept that has evolved. That is the reason why we consider important to highlight it in the conclusions. However, and in accordance with your instructions, the clarity of the wording has been improved.

Reviewer 3 Report

Comments and Suggestions for Authors

The article presented for review is a very interesting literature analysis of research in the field of defining forest health. The authors initially precisely defined the objectives of the research undertaken, then correctly presented the procedure and described the results. This is a thorough analysis. It seems that there was no explanation in the description of the methodology, whether you researched what forests (from what region of the world) the individual researchers were writing about, e.g. whether if the author(s) or the journal came from the USA, they were writing about US forests. It often happens that scientists are interested in forests from other parts of the world and then it cannot be concluded in the discussion that they are mainly interested in forests from a given country. Another important element that should be addressed is the digitization of collections. The history of your analysis begins in the 1960s, although the first publication was from 1936, then the publications were only in paper form, the digital archiving of these publications took place much later. There are still many countries in which scientific publications have only been partially archived, could this have influenced the results?

Author Response

Comments - Reviewer 2

C2/ The article presented for review is a very interesting literature analysis of research in the field of defining forest health. The authors initially precisely defined the objectives of the research undertaken, then correctly presented the procedure and described the results. This is a thorough analysis. It seems that there was no explanation in the description of the methodology, whether you researched what forests (from what region of the world) the individual researchers were writing about, e.g. whether if the author(s) or the journal came from the USA, they were writing about US forests. It often happens that scientists are interested in forests from other parts of the world and then it cannot be concluded in the discussion that they are mainly interested in forests from a given country.

R2/ We appreciate the reviewer's comment. Our review is global without limiting the area or forest of scope. We have included this clarification in L133: “The temporality and geographical scope of the search was not limited, as the aim was to consider all global existing records in the databases from their origin to the present”

The results provided about countries (e.g. Fig 2) consider the country of affiliation of the researchers, not the country of study. This information has now been provided in L156 and represents well the “ecosystem” of publication. As the reviewer suggests most of researchers will study their closest ecosystems but it is not always the case. Thus, from this piece of information we can’t infer specifically anything about the region or forest ecosystem of study. In any case, the results obtained showed that given the nature of global research, the majority included work from the northern hemisphere, in particular in the USA and Europe, at least until the middle of the 20th century, which represents a bias in the results. It is true that in recent decades numerous articles have emerged in the southern hemisphere, particularly in tropical forests and temperate forests. However, in the review as a whole we understand that the results reflect trends on a global scale.

C2/ Another important element that should be addressed is the digitization of collections. The history of your analysis begins in the 1960s, although the first publication was from 1936, then the publications were only in paper form, the digital archiving of these publications took place much later. There are still many countries in which scientific publications have only been partially archived, could this have influenced the results?

R2/ we agree with the reviewer that the studies corresponding to the first decades of the 20th century may be underrepresented due to the fact that most of them are not digitized. However, when carrying out cross-reviews with the first digitized works, particularly from the 50s and 60s of the last century, we have been able to verify that there are few references to this type of events and that the majority of studies were published from the 80s. Therefore, our main results (Fig. 3-5) focus on the main period of publication (from 80s) where we have enough information to reflect well the evolution of the concept of forest health.