Ownership Patterns Drive Multi-Scale Forest Structure Patterns across a Forested Region in Southern Coastal Oregon, USA
Round 1
Reviewer 1 Report
General Comments:
This is a clever way to use lidar to assess forest structural conditions across land ownerships, in a way that I think forest landscape managers can utilize given the chosen response variables: mean match size and interspersion/juxtaposition. The application of PCA is also clever, as a way to exploit the information content of the selected lidar metrics. They found a very clear separation between production-oriented and structurally complex forests that aligns with private versus public land ownerships. The management and policy implications are well considered. I have just some minor issues with the presentation that should be easy to fix.
Specific Comments:
L65-69. Don’t private lands tend to be lower-lying, less steep and more productive than public lands? The original homesteaders claimed land that was more productive and valuable. However, you don’t make this point until much later (L511-522). Some minor reorganization is warranted.
Table 1. The “Private” label is vague and confusing because it indicates land parcels smaller than “Private Very Small”. Why would you not call the smallest case “Private Very Small”, followed by “Private Small”, then “Private Medium”, and “Private Large” as the largest case? These are just relative size labels within the “Private Non-Industrial” ownership category. Moreover, must what you label “Private” only be 50 ha? Can it not be <50 ha (see L486)?
L185. Please name the forest type covering 15% of the study area that you chose to omit.
L194. Canopy cover is more defensibly calculated from just 1st returns, not all returns. You could call it something different, such as canopy density.
L203-215. You don’t offer a citation for applying PCA to lidar metrics. Here’s one:
Silva, C.A., C. Klauberg, A.T. Hudak, L.A. Vierling, V. Liesenberg, S.S.P. Carvalho and L.C.E. Rodriguez. (2016) A principal component approach for predicting the stem volume in Eucalyptus plantations in Brazil using airborne LiDAR data. Forestry: An International Journal of Forest Research 89(4): 422-433. https://doi.org/10.1093/forestry/cpw016.
Fig. 2A. If you orient the x-axis labels vertically instead of horizontally like now, they won’t overlap.
L415. Make “stand” plural.
L440-454. Under the NWFP, hasn’t there been some effort by managers to create stand structural complexity (similar to old-growth) faster than it develops naturally, in part to create hopefully suitable habitat for spotted owls and marbled murrelets? However, you don’t get to this until section 4.2.1, which isn’t a smooth logical flow. Minor reorganization is warranted.
L503. This sentence doesn’t conclude. Section ____?
L587. Say “waiting” or “to wait” instead of “waited”?
L607-613. Couldn’t author initials be used instead of full names to shorten this section?
L761. Are ALL CAPS necessary?
Author Response
This is a clever way to use lidar to assess forest structural conditions across land ownerships, in a way that I think forest landscape managers can utilize given the chosen response variables: mean match size and interspersion/juxtaposition. The application of PCA is also clever, as a way to exploit the information content of the selected lidar metrics. They found a very clear separation between production-oriented and structurally complex forests that aligns with private versus public land ownerships. The management and policy implications are well considered. I have just some minor issues with the presentation that should be easy to fix.
Response: Thank you.
L65-69. Don’t private lands tend to be lower-lying, less steep and more productive than public lands? The original homesteaders claimed land that was more productive and valuable. However, you don’t make this point until much later (L511-522). Some minor reorganization is warranted.
Response: We have moved this point to a new paragraph in the Study Area section that describes land tenure history (lines 158-167).
“Complicated histories of land tenure dominate the American west, including our study area. First, private and public ownership is often divided by elevation. This bifurcation is the result of the federal government ceding more productive lands to private interests and retaining the land that was more difficult to utilize (because of its higher elevation and topographic complexity) in the early 1900s [64]. However, in our study area, there are instances of intentional public and private intermixing. These are the result of 19h century land grants to support the building of railroads in parts of the American west: here, the Oregon and California railroad lands (O&C) and Coos Bay Wagon Road railroad lands (CBWR). These land grants ceded every other square mile (259 ha) of federal land to railroad companies, resulting in a “checkerboard” pattern of public and private ownership [65].”
Table 1. The “Private” label is vague and confusing because it indicates land parcels smaller than “Private Very Small”. Why would you not call the smallest case “Private Very Small”, followed by “Private Small”, then “Private Medium”, and “Private Large” as the largest case? These are just relative size labels within the “Private Non-Industrial” ownership category. Moreover, must what you label “Private” only be 50 ha? Can it not be <50 ha (see L486)?
Response: While we do understand the confusion, these names are from the database from which these data originate. Since it is in use by others at TNC and their projects throughout the state, we wish to remain consistent with their naming standards. Yes, you are correct, lands can be <50 ha on “Private” lands and we have changed the text to reflect that.
L185. Please name the forest type covering 15% of the study area that you chose to omit.
Response: We have added the forest type, Oregon Coastal Tanoak (Line 202).
L194. Canopy cover is more defensibly calculated from just 1st returns, not all returns. You could call it something different, such as canopy density.
Response: Prior papers by co-authors of this manuscript have used all returns, notably Kane et al. 2013. We chose to use all returns here because we find it better approximates being in the field and evaluating canopy cover, as if you were looking up through all layers of the trees. Additionally, we have found a high R^2 (typically 0.98-0.99) when evaluating the relationship between canopy cover counted with first returns and all returns.
L203-215. You don’t offer a citation for applying PCA to lidar metrics. Here’s one:
Silva, C.A., C. Klauberg, A.T. Hudak, L.A. Vierling, V. Liesenberg, S.S.P. Carvalho and L.C.E. Rodriguez. (2016) A principal component approach for predicting the stem volume in Eucalyptus plantations in Brazil using airborne LiDAR data. Forestry: An International Journal of Forest Research 89(4): 422-433. https://doi.org/10.1093/forestry/cpw016.
Response: Added this reference to line 227.
Fig. 2A. If you orient the x-axis labels vertically instead of horizontally like now, they won’t overlap.
Response: While we did not change the x-axis label orientation, we did increase the spacing of the figure to ensure the text does not overlap.
L415. Make “stand” plural.
Response: Done.
L440-454. Under the NWFP, hasn’t there been some effort by managers to create stand structural complexity (similar to old-growth) faster than it develops naturally, in part to create hopefully suitable habitat for spotted owls and marbled murrelets? However, you don’t get to this until section 4.2.1, which isn’t a smooth logical flow. Minor reorganization is warranted.
Response: yes, a good point. We have added two sentences to this effect in lines 476-480.
“Alternatively, these patches could be the result of efforts by managers under the NWFP to create structural complexity faster than it develops naturally, with the aim of creating suitable habitat for spotted owls and marbled murrelets (see Section 4.2.1). Again, if future research bears out the younger mean age for complex stands, then this suggests a contrary strategy of active management to create structural complexity.”
L503. This sentence doesn’t conclude. Section ____?
Response: We received this comment from one other reviewer as well. However, when we checked this, we found that we finish the sentence with “Section 4.4”. The number 4.4 was in a new line by itself and perhaps this is why it has been missed. It now should be more visible (Line 530).
L587. Say “waiting” or “to wait” instead of “waited”?
Response: Changed to “to wait”.
L607-613. Couldn’t author initials be used instead of full names to shorten this section?
Response: Done.
L761. Are ALL CAPS necessary?
Response: Changed to lower case letters.
Reviewer 2 Report
Review manuscript forests-1020215
The authors deal with an important and interesting topic on the influence of ownership patterns on forest structures - their research can make an important contribution to the development of sustainable use modes for forests and provision of their ecosystem services.
After detecting some very early paper writing errors, that I know from my students, I guess this manuscript is developed based on a master thesis, covers very valuable work done and has very valuable results, but is not in a apropriate form for publication. I hope my comments will help to impove strongly the manuscript, I don#t want to frustrate you, your research is worth to be published.
Line 11ff. Abstract:
I would strongly recommend to completely rework this summary, please focus more on your results and insights. Very few introductory words are enough to shed light on the context and research needs. At the moment, the introduction takes up more than the half of the abstract, informations on your results are much too short. It should be the other way arround.
Introduction:
The introduction should be reworked and be much more precise and focused on issues related to your research, to your study region. And go deeper into the topic and write more precise. The writing in this manuscript is often writing is not scientific, this seems to be obscured by the extreme number of quotations.
Please revise carefully the number of citations you are using more than 109 citations on less than 20 pages, often more than siy different for one short sentence/idea. I understand that this is not an review or a text book for an university course. It is important to select the most relevant literature and don’t cite all. Please reduce this to maxima two or three. Due to this the use of multitude of citations it finally remains unclear and arbitrary. You give many references for generalities, a review article is enough. Please keep the audience in mind. People that read your target journal are aware of the basics. This multitude of citations gives the impression that the authors are not sure about the context. Please be more self-confident in your topic and sory telling. You should delete some more than 60 citations ana keep some 40-45 in the manuscript. Focus on the most relevant. Here applies as with many points much does not help much.
Check for redundancies. E.g. Line 39 and 43 Forest structure—the horizontal and vertical arrangement of biomass; forest structure is the arrangement of…
Related to this some parts of the introduction remain imprecise and produce more questions than answers. You should be more focused and clear with the meaning of your sentences. In contrast to authors that are no native speaker I feel here it is not a language issue it is a content issue. Please be more precise and not to vague and generic in your formulations. Some examples:
46f. a forest manager would not state this! Change structures of forests except fast growing timber plantations needs generations, they are not at all easy to manipulate.
51f. In many societies, management goals and practices are organized by different classes of owners [21]. What does this mean, what is society?
An other example on unprecise writing and citing line 60ff. 15 citations for a sentence where no concreate approach is mentioned, all information is general…
“The effects of forest ownership on forest conditions have received attention in recent years, in part due to emerging new forest owner types, ownership fragmentation, and questions related to sustainable yield of forest products [28]. The effects of forest ownership on ecological patterns have been explored using multiple approaches. For example, forest management, land cover, and land use are all critical components of ownership and have been used as lenses to examine ownership effects on forest cover, patch dynamics, structure, and ecosystem function and services [29–44].”
Other example which produce more confusion:
line 67ff. “This is typically because public lands have less intense land use regimes [45] (although this is not always the case, see [46, 47]) and were found to host more biodiversity [46].’
I miss more precise definitions on ownerships. The dichotomy between private and public is much to general.
Material and Methods
A general comment: the database for the work should be accessible to access the work, I miss an appendix or something. You mentioned e.g. in line 143 spatially esplicit data without reference) as well as other sources without references and following lines. I can believe that there are the data (and I’m sure they are somewere), but these important sources are not easy to find or access. The M&M section could have an overview table about the data used with precise citations. The data it self can be linked (or presented) in the appendix.
Line 147f. Why did you excluded owners with less than 10,000 ha? Based on what criteria the types where deifined. I guess these are types from the data base used etc.etc.etc.
To bring it to the point: Please rework the M&M section in order to provide a complete description to enable reproducibility of your methods
Line 118f again example of the above mentiond problem: “The forests of the Oregon Coast Range are some of the most productive in the world [58] and as a result are both economically and ecologically important.” Why ecologically? Be more precise and come to the point.
I like Table 1 and Fig.1 but please add also some fotographs, to give the reader an impression, skip gridlines, but provide access, letters/ numbers are much to small.
Who is Derek Churchill? Where he is working?
I miss the definitions of forest structure classes. I have the impression, that you did a lot of desk work but have not been so much in touch with the reality in field?
Line 134 … (not reported) can you cite some other diocument or grey literature, or personal communication
Check for redundancy.
In the chapter 2.5 again references are missing. Fragstats?
Result section
Please enhance this section. The results are not clearly presented y see a lot of confusing Figures, I miss tables wirth the (also statistical) data. In the current form it is not possible to assess the data.
Fig. 2 See comments on Figures above, In general I like the Figures, but please enhance and include the basics like axis.
Fig.3 and Fig.4 should be re-worked. Letters are much to small there is no chance to reasd the figures.
Better name chapter 3.1 Identified classes of forest structure?
The same holds for the other figures it is complicated to understand and interpret this section, with better (the more colourful is also not the better) and cleat Figures and some table on statistics could help to present the results better.
Discussion
I feel that a proper assessment of this needs a clear and transparent presentation of the results. Here, I stopped assessment due to the fact that the results have not presented in a appropriate form to be able to develop nor provide a judgement on this section.
Without going in detail, I observed a almost total lack of comparision with literature. After the overflow of citations in the introduction here an reflection of own data with literature is missing.
I insist that methods and results (although not well and clearly presented) might have the potental to be published, but not in the current form. I#m happy to review a strongly enhanced version of the manuscript.
Author Response
The authors want to thank the reviewer for the time they put in this review. We are sorry they are disappointed in the manuscript. We did our best to respond to the reviewer’s comments; however, there are instances where we found the comments to be too vague to provide guidance on how we should respond to improve the manuscript.
The authors deal with an important and interesting topic on the influence of ownership patterns on forest structures - their research can make an important contribution to the development of sustainable use modes for forests and provision of their ecosystem services.
After detecting some very early paper writing errors, that I know from my students, I guess this manuscript is developed based on a master thesis, covers very valuable work done and has very valuable results, but is not in a apropriate form for publication. I hope my comments will help to impove strongly the manuscript, I don#t want to frustrate you, your research is worth to be published.
Line 11ff. Abstract:
I would strongly recommend to completely rework this summary, please focus more on your results and insights. Very few introductory words are enough to shed light on the context and research needs. At the moment, the introduction takes up more than the half of the abstract, informations on your results are much too short. It should be the other way arround.
Response: Forests requires that Research Highlights, Background and Objectives, Materials and Methods, Results, and Conclusions be included in the abstract. Considering the need both to abstract these sections appropriately and to meet the word limit, we feel that the abstract addresses each section appropriately.
Introduction:
The introduction should be reworked and be much more precise and focused on issues related to your research, to your study region. And go deeper into the topic and write more precise. The writing in this manuscript is often writing is not scientific, this seems to be obscured by the extreme number of quotations.
Please revise carefully the number of citations you are using more than 109 citations on less than 20 pages, often more than siy different for one short sentence/idea. I understand that this is not an review or a text book for an university course. It is important to select the most relevant literature and don’t cite all. Please reduce this to maxima two or three. Due to this the use of multitude of citations it finally remains unclear and arbitrary. You give many references for generalities, a review article is enough. Please keep the audience in mind. People that read your target journal are aware of the basics. This multitude of citations gives the impression that the authors are not sure about the context. Please be more self-confident in your topic and sory telling. You should delete some more than 60 citations ana keep some 40-45 in the manuscript. Focus on the most relevant. Here applies as with many points much does not help much.
Response: We understand the concern. Forests is a well-regarded journal with high scientific standards so we feel that the number of citations is appropriate to their standards. Additionally, no review papers exist for this topic. We will follow the guidance of the editor on this point.
Check for redundancies. E.g. Line 39 and 43 Forest structure—the horizontal and vertical arrangement of biomass; forest structure is the arrangement of…
Response: This sentence now reads “Forest structure also controls the amount and configuration of carbon stored in a tree and forest [12].” (Lines 43-45)
Related to this some parts of the introduction remain imprecise and produce more questions than answers. You should be more focused and clear with the meaning of your sentences. In contrast to authors that are no native speaker I feel here it is not a language issue it is a content issue. Please be more precise and not to vague and generic in your formulations. Some examples:
46f. a forest manager would not state this! Change structures of forests except fast growing timber plantations needs generations, they are not at all easy to manipulate.
Response: We have changed this to “can be manipulated” in line 47. Additionally, this statement is supported by Franklin et al. 2002. From the introduction of Franklin et al. 2002, “Structural attributes of forest stands are increasingly recognized as being of theoretical and practical importance in understanding and managing forest ecosystems because:
- structure is the attribute most often manipulated to achieve management objectives following establishment of a forest stand.”
51f. In many societies, management goals and practices are organized by different classes of owners [21]. What does this mean, what is society?
An other example on unprecise writing and citing line 60ff. 15 citations for a sentence where no concreate approach is mentioned, all information is general…
“The effects of forest ownership on forest conditions have received attention in recent years, in part due to emerging new forest owner types, ownership fragmentation, and questions related to sustainable yield of forest products [28]. The effects of forest ownership on ecological patterns have been explored using multiple approaches. For example, forest management, land cover, and land use are all critical components of ownership and have been used as lenses to examine ownership effects on forest cover, patch dynamics, structure, and ecosystem function and services [29–44].”
Other example which produce more confusion:
line 67ff. “This is typically because public lands have less intense land use regimes [45] (although this is not always the case, see [46, 47]) and were found to host more biodiversity [46].’
Response: To clarify this statement, we have added “as compared to production focused private lands” in this sentence (lines 68-69).
I miss more precise definitions on ownerships. The dichotomy between private and public is much to general.
Response: We added “(e.g. federal and state)” to further define public lands in line 67 and “(public)” to line 73 to clarify the distinctions between public and private in this region.
Material and Methods
A general comment: the database for the work should be accessible to access the work, I miss an appendix or something. You mentioned e.g. in line 143 spatially esplicit data without reference) as well as other sources without references and following lines. I can believe that there are the data (and I’m sure they are somewere), but these important sources are not easy to find or access. The M&M section could have an overview table about the data used with precise citations. The data it self can be linked (or presented) in the appendix.
Response: Unfortunately, this database is a proprietary database and cannot be shared. We have added a sentence to this effect in line 149 to clarify.
“We are unable to share this dataset because of its proprietary nature.”
Line 147f. Why did you excluded owners with less than 10,000 ha? Based on what criteria the types where deifined. I guess these are types from the data base used etc.etc.etc.
Response: We excluded owners with less than 10000 ha because of how small of a proportion of land this would represent in our study area (less than 0.5%). We have added a sentence to this effect in lines 152-153.
“We excluded ownership classes with less than 10000 hectares in our study area because they would represent less than 0.5% of the landscape.”
To bring it to the point: Please rework the M&M section in order to provide a complete description to enable reproducibility of your methods
Response: We believe that within the restrictions of our data sources and with the addition of clarifying sentences, our methods are reproducible.
Line 118f again example of the above mentioned problem: “The forests of the Oregon Coast Range are some of the most productive in the world [58] and as a result are both economically and ecologically important.” Why ecologically? Be more precise and come to the point.
I like Table 1 and Fig.1 but please add also some photographs, to give the reader an impression, skip gridlines, but provide access, letters/ numbers are much too small.
Response: We are not sure what photographs we could add to a table of ownerships. We have increased text size in figure 1.
Who is Derek Churchill? Where he is working?
Response: We have added Derek Churchill’s affiliation (Washington State Department of Natural Resources) to the in-text citation, line 188.
I miss the definitions of forest structure classes. I have the impression, that you did a lot of desk work but have not been so much in touch with the reality in field?
Response: The use of lidar to define forest structure classes is well represented in the literature. See Kane et al. 2010a and b, Kane et al. 2013, North et al. 2017, and Kane et al. 2019 for the methods also used in this paper. For other uses of forest structure aside from these specific methods, see Moran et al. 2018, Listopad et al. 2015, and Smart et al. 2012. Kane et al. 2010 a and b using the methods used here, calibrated their structure classes to field data and demonstrated that lidar can be used as a primary resource in place of field data.
The forest structure classes are derived using our methods, so you will find their definitions in the results section.
Line 134 … (not reported) can you cite some other document or grey literature, or personal communication.
Response: We believe this information is not central enough to the story of this paper to warrant including here. We follow common practice by not reporting all results and additionally, the paper would be too long if we were to include this.
Check for redundancy.
Response: Unfortunately, this comment is too vague to respond to.
In the chapter 2.5 again references are missing. Fragstats?
Response: We have added a reference for the Fragstats manual (McGarigal 2015).
Result section
Please enhance this section. The results are not clearly presented y see a lot of confusing Figures, I miss tables wirth the (also statistical) data. In the current form it is not possible to assess the data.
Fig. 2 See comments on Figures above, In general I like the Figures, but please enhance and include the basics like axis.
Fig.3 and Fig.4 should be re-worked. Letters are much to small there is no chance to reasd the figures.
Response: We have increased the text size in these figures.
Better name chapter 3.1 Identified classes of forest structure?
The same holds for the other figures it is complicated to understand and interpret this section, with better (the more colourful is also not the better) and cleat Figures and some table on statistics could help to present the results better.
Response: These comments are difficult to evaluate because of their generality.
Discussion
I feel that a proper assessment of this needs a clear and transparent presentation of the results. Here, I stopped assessment due to the fact that the results have not presented in a appropriate form to be able to develop nor provide a judgement on this section.
Without going in detail, I observed a almost total lack of comparision with literature. After the overflow of citations in the introduction here an reflection of own data with literature is missing.
I insist that methods and results (although not well and clearly presented) might have the potental to be published, but not in the current form. I#m happy to review a strongly enhanced version of the manuscript.
We regret that the reviewer did not provide an assessment of the discussion.
Reviewer 3 Report
Having the experience of dealing with a social-ecological diagnosis of a 100 000 ha forested landscape (for the purpose of scenario building and and the initialization of forest growth models) , I can attest of the relevance of a methodology-enclined forestry research set at such a 'meso' scale. Articulating ownership knowledge with ALS data, this paper is a welcome addition to the litterature.
I must say the paper is an enjoyable read, as it is well written, well structured, accompanied by maps and graphs of a good semiological standard. See the end of my comments for a few typos and unclear phrases. The most difficult part may have to do with the specific context of land tenure history in the american West: the checkerboard pattern of public/private lands appearing in fig 1 is not evident for a non-US reader. Explanations are also wanted for some institutions eg. "O&C lands".
Besides those quibbles, the overall argument of the text is rather clear: overlaying proxies for forest attributes derived from a Lidar coverage with a typology of owners over a study site in coastal SW Oregon, the authors reach the conclusion that there is a bifurcation in the structure of forests under private and public ownership, the latter being associated with older, taller and more complex stands. They also scrutinize the variations within the broad ownership classes: non-trivially, the non-industrial private owners do not have attributes of more complex forest structures than the industrial ones. Those findings have several policy implications:
- Complex forest structures seem to be reached rather quickly, validating the management practices pursued under regional directives (the North West Forest Plan) for (§ 4.4.1.)
- The bifurcation between public forests and private industrial forests should lead to a prioritization of conservation easements in settings where those two types of properties are close: the checkered O&C patterns, located in NE & SE of the area, being good candidates for this (§ 4.4.2.)
- The spatial constraints put on clearcutting areas by the Oregon Forest Practices Act are not sufficiently binding to create real differences in landscape-level structures (§ 4.4.3.)
The added value of the research might have been not very decisive to reach the 3rd conservation conclusion, but that is not the case of the others.
Considering that the analysis finally leads to rather strong claims, I feel that there are three features of the paper that tend to weaken its arguments.
1. The choice of forest structure metrics and the PCA-based typology
A typology of six classes of forest structure is proposed, based on a PCA+hierarchical clustering of 4 Lidar-derived metrics. Three metrics are well-adapted to monitor the development of an even-aged stand, while the last one -rumple- capture the rugosity of the canopy, that may be associated with complex, irregular structures. Due to this selection of metrics, most of the logic of the classification revolves around age.
As a result, the classes 5 & 6 are those with the highest dominant heights and the higher values for the rumple index. What complicates the picture is also the result, if I read well another paper from one of the co-authors, of an interaction between p95 & the rumple index:
https://cdnsciencepub.com/doi/full/10.1139/X10-064
"Rumple therefore should be interpreted in light of canopy height. For example, a rumple value of 2.5 for a 30m 95th percentile height would indicate high structural complexity, while the same rumple for a 60m height would indicate low structural complexity."
A frustration may arise because, with only 6 classes, we do not have the tools to discern highly stocked even-aged forest stands from more complex, uneven-aged stands, as both are mostly conflated in classes 5 & 6
The commentaries in the aforementioned paper add nuance to this question, as eg old-growth forests may have, due to the prevalence of dead broken crowns, a lidar signature that is not so easy to discern from younger even-aged stands. So I understand this is not a trivial issue.
But such nuance is a bit missing here, as the "complex forest structures" are somehow taken at face-value in the remainder of the text. This should be improved or the conclusions a bit toned down.
[Another choice, made earlier by the co-authors, is the 30 m resolution for the the lidar metrics. I understand well the advantages (compatibility with inventory plots, landsat data...) but could the rumple index, obviously rather scale-dependant, have been computed at coarser levels?]
2. A lack of history?
There are idiosyncrasies guiding the spatial patterns of the ownership classes (eg. the private non-industrial properties are mostly located in the SW of the area). I was very convinced by the explanations of § 4.3 but they come late in the paper, and remain partial.
Eg, an important role is given to the areas burnt around 1900, as they might be the dominant event driving forest (re)growth. I had the curiosity to check the Interagency Fire Perimeter History GIS files: large fires seem indeed to be rare in the vicinity, but still happened during the XXth century. North of Elkton, there is a 'wildfire' patch of 18000 ha for the year 1966, now largely dominated by class 4 (harvestable) stands.
Accordingly, it is not clear in the text how different the forest composition was a century ago. For what it's worth, The Natural Vegetation of Oregon map from 1915 is dominated by 'mesophytic coniferous forest', but it also points to noticeable patches of grasslands and chapparal in the area, ~ 50 000 ha near Coos Bay.
This is not to say that the text should be turned into the historical ecology paper that it is not, but I feel more context should be provided in § 2.1 & 2.2, eg being more explicit in the dominant locations of each owners categories.
https://www.fs.usda.gov/Internet/FSE_DOCUMENTS/fseprd716479.pdf
https://www.fs.usda.gov/detail/umatilla/learning/history-culture/?cid=fsbdev7_016133
3. Unclear definition and use of landscape ecology metrics
The reminder of the 3 dimensions of the analysis (individual patch level, within ownership class, between classes) was particularly welcome line 503. Such level of clarity is not always reached earlier in the paper, espacially when it comes to the landscape ecology metrics.
For exemple, the Area-weighted mean patch size is defined with confusing terms: I guess xij should be replaced by aij/N, and n (nb of patches in each class) and N (total number of patches in the landscape) be explained. To be fair, the FRAGSTATS manual is itself rather laconic (my source was p.87 in Botequilha et al. Measuring landscapes: A planner's Handbook)
More problematic was the status of the IJI metric: its formula seems a bit simpler in FRAGSTATS, as well as in the landscapemetrics R vignette
I am afraid it was not totally clear to me whether it was computed onwership-class-wise, If so how? one against the 12 others?
To fully understand the results of the regression models underlying figs 5 & 6, the explanations on formulae should be more explicit in this regard.
***
The improvements in these 3 directions are in my opinion needed. I recommend to accept the paper, and consider these changes as minor revisions, that I would be happy to read.
best regards
l43-44: rewrite?
l67-68: indeed, not always the case in Europe
l162: explanation on 'matrix lands'
fig1: km instead of miles for the legend. the maps are a bit small
l211: include the screeplot?
l236 & 241 Cf. above
Figure 3A: improve the resolution? (or vector eps files, don't know whether that's standard for MDPI)
l446: a rather strong statement
l 460: convert acres to ha
l503: unfinished phrase
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Round 2
Reviewer 2 Report
The authors have unfortunately ignored most of my well-intentioned advice and answer vague and generalities on your part.
For me there are mayor problems with the quality of the manuscript, which were not addressed in the second version.
Among them the issue, that the authors cited more than 110 references in a simple research paper. The statement that there is no review on this is invalid. The authors of course are free to write a review, but this should be a another manuscript and for an review a methodology should be provided. all this is not done and it seems that the authors pushed only things in after the principle: the much the better. But an important scientific skill is to decide between important and less important things.
The M&M section is not transparent, the research is not replicable based on the information provided.
The result section is also poor, here tables with data are missing, there are only figures with low quality, they are colorful, but often lack the basic elements of a diagram, again largely untransparent.
The Discussion has the same problem like the introduction (see above) and parts have no references to literature.
My main concerns have not been addressed. I cannot propose the paper for publication in order to maintain scientific standards.
Author Response
We are dividing our response into multiple sections. We first include our response to the reviewer’s second review. Following this, we include comments from this reviewer’s first and second reviews where we illustrate our difficulty in addressing this reviewer’s concerns. Whenever the reviewer made clear requests, we followed the reviewer’s recommendations and include those as well.
Second Review
The authors have unfortunately ignored most of my well-intentioned advice and answer vague and generalities on your part.
For me there are mayor problems with the quality of the manuscript, which were not addressed in the second version.
Among them the issue, that the authors cited more than 110 references in a simple research paper. The statement that there is no review on this is invalid. The authors of course are free to write a review, but this should be a another manuscript and for an review a methodology should be provided. all this is not done and it seems that the authors pushed only things in after the principle: the much the better. But an important scientific skill is to decide between important and less important things.
Response: We reviewed prior Forests articles and see that it is common in this journal to have far fewer citations than in our paper. However, many of these articles are very specific and do not cross discipline boundaries as our study does. From a brief examination of articles from other journals this paper could be appropriate for, such as Landscape Ecology and Forest Ecology and Management, we find that there were frequently 100+ citations. We maintain that the number of citations is appropriate given the interdisciplinary scope of this paper and that it is not scientifically unsound to have these citations.
The M&M section is not transparent, the research is not replicable based on the information provided.
Response: Unfortunately, we do not know how to improve the M&M section based on this comment.
The result section is also poor, here tables with data are missing, there are only figures with low quality, they are colorful, but often lack the basic elements of a diagram, again largely untransparent.
Response: We are unsure what tables the reviewer proposes we include. We are also unsure how to improve the figures based on this comment.
The Discussion has the same problem like the introduction (see above) and parts have no references to literature.
Response: The only sections that do not review the literature are those summarizing the results of this study. Lines XXX are devoted to synthesizing these results with prior literature.
My main concerns have not been addressed. I cannot propose the paper for publication in order to maintain scientific standards.
Comments we found too vague to respond to
“The introduction should be reworked and be much more precise and focused on issues related to your research, to your study region. And go deeper into the topic and write more precise. The writing in this manuscript is often writing is not scientific, this seems to be obscured by the extreme number of quotations.”
Response: We could spend time rewriting the introduction but reviewer 2 could respond that those changes were not what they meant. This comment does not tell us what reviewer 2 would like to see changed and how.
“Please revise carefully the number of citations you are using more than 109 citations… Without going in detail, I observed a almost total lack of comparision with literature. After the overflow of citations in the introduction here an reflection of own data with literature is missing.”
Response: As we point out in the introduction, this is novel work enabled by newly available airborne lidar data and a precise (but proprietary) property ownership database. By definition, novel research will limit the amount of directly applicable prior studies to cite. Despite this, we have cited by a quick count 36 prior papers in her discussion. This is close to the maximum number of citations that reviewer 2 recommends for the entire paper.
In addition, it is traditional in reviews for the reviewer to cite specific papers that they think should be incorporated in a manuscript, not give vague statements that prior literature is not cited.
“The result section is also poor, here tables with data are missing, there are only figures with low quality, they are colorful, but often lack the basic elements of a diagram, again largely untransparent.”
Response: Which tables and which data? The authors could add a dozen tables and none of them might address this reviewer’s concerns.
“The M&M section is not transparent, the research is not replicable based on the information provided.”
Response: Which specific methods aren’t replicable?
“Please enhance this section. The results are not clearly presented y see a lot of confusing Figures, I miss tables wirth the (also statistical) data. In the current form it is not possible to assess the data.”
Response: The authors are unsure what changes we might make to address this comment.
Clear requests which we followed
Check for redundancies. E.g. Line 39 and 43 Forest structure—the horizontal and vertical arrangement of biomass; forest structure is the arrangement of…
Response: In line 43, we used “Because forest structure is the arrangement of biomass…” as a transition phrase to help explain the following phrase of that sentence and to tie in to the previous use.
Related to this some parts of the introduction remain imprecise and produce more questions than answers. You should be more focused and clear with the meaning of your sentences. In contrast to authors that are no native speaker I feel here it is not a language issue it is a content issue. Please be more precise and not to vague and generic in your formulations. Some examples:
46f. a forest manager would not state this! Change structures of forests except fast growing timber plantations needs generations, they are not at all easy to manipulate.
Response: We have changed this to “can be manipulated” in line 47. Additionally, this statement is supported by Franklin et al. 2002. From the introduction of Franklin et al. 2002, “Structural attributes of forest stands are increasingly recognized as being of theoretical and practical importance in understanding and managing forest ecosystems because:
- structure is the attribute most often manipulated to achieve management objectives following establishment of a forest stand.”
line 67ff. “This is typically because public lands have less intense land use regimes [45] (although this is not always the case, see [46, 47]) and were found to host more biodiversity [46].’
Response: To clarify this statement, we have added “as compared to production focused private lands” in this sentence (lines 68-69).
I miss more precise definitions on ownerships. The dichotomy between private and public is much to general.
Response: We added “(e.g. federal and state)” to further define public lands in line 67 and “(public)” to line 73 to clarify the distinctions between public and private in this region.
Line 147f. Why did you excluded owners with less than 10,000 ha? Based on what criteria the types where deifined. I guess these are types from the data base used etc.etc.etc.
Response: We excluded owners with less than 10000 ha because of how small of a proportion of land this would represent in our study area (less than 0.5%). We have added a sentence to this effect in lines 153-154.
“We excluded ownership classes with less than 10000 hectares in our study area because they would represent less than 0.5% of the landscape.”
