Exemplifying Stratified Deforestation in Four Protected Areas in Madagascar

Round 1

Reviewer 1 Report

The reviewer finds the manuscript well written and contributes to the well-known problem of deforestation in Madagascar. The manuscript is in line with the purpose of the journal and is worthy of publication. However, there is still information and analysis that should be included to improve the study.

 

Comments and Suggestions for Authors

The manuscript is a study of deforestation trends in four different protected areas on the island of Madagascar. The study shows that deforestation rates are higher in large and newly established protected areas. However, the causes leading to this deforestation could not be fully explained. The manuscript contributes to the current issue of deforestation in high biodiversity areas. Several improvements that can enhance the manuscript are recommended below.

 

Major and Minor Issues

Title

The title should be revised as it currently refers to deforestation processes, but the authors do not have data to explain the processes that lead to deforestation over time, so this is not appropriate.

 

Abstract

The abstract is confusing, especially the methods and results should be formulated more clearly. L28: Do you mean “prove”? Prove is a verb, while proof is a noun

L29: site-specific current

The hyphen is missing in many cases, please double check it.

The full name of the abbreviations should be given the first time they are mentioned, e.g., Convention on Biological Diversity (CBD).

 

Introduction

The introduction is easy to read and understand, but the objective, hypothesis and research questions are missing. Theoretical aspects on protected areas inside/outside are not mentioned, although the authors analyze forest cover inside and outside PAs.

Also, key research is not included, please check:

Global patterns of forest loss across IUCN categories of protected areas, Leberger et al. 2019

Predicted climate shifts within terrestrial protected areas worldwide, Hoffmann et al. 2019

 

 

Methods

Table 1. Does the color have a meaning? If so, explain it clearly, as well in Table 2.

Simplify Table 2. The total area is the same in most cases. As it is now, it looks confusing and the reader cannot easily find the most important information. Why are some rows colored?

L92: A buffer has a certain size. So what do you mean by over 10 km?

L93-100: Explain why you choose them in a certain order. I don't understand why you choose a buffer of over 10 km.

Are there no other small roads apart from those mentioned in the table, and what about forest fragmentation?

Although some causes of deforestation were mentioned in the discussion, no data were presented that could be analyzed. For example, it should be analyzed whether human settlement, population density and distance to the road play a role.

See: “Deforestation processes in south-western Madagascar over the past 40 years: what can we learn from settlement characteristics?” https://doi.org/10.1016/j.agee.2014.06.008

 

Results

L160-173: Some assumptions are mentioned, but they have no basis. These assumptions should be mentioned in the introduction.

Fig. 3. The idea of the diagram is good, but the colors are similar to those used above (see Table 1, 2), so the reader cannot make a difference between the buffer sizes. Again, what is >10km? 20, 100km? Also, the data for the other PAs should be included, if not here, at least in the supplementary materials. Why did you use a black background for the figures?

Table 2 also shows that the total area in Behara is higher than in Tsaratanana (except 2015-17). The cases shown by the authors in Figure 3 correspond to the cases where deforestation was highest.

Although there are clear patterns of deforestation, these are not explained with a proper analysis.

 

Discussion

The first paragraph needs to be restructured, because it is confusing, it should better contain general findings, more theoretical arguments for or against, findings from other regions, followed by detailed information about each PA (what has been done).

The discussion focuses mainly on the four study regions, but what about all of Madagascar? There are other publications on other areas that should be considered. These could support the findings and give a broader overview.

See: “Analysis of deforestation patterns in the central Menabe, Madagascar, between 1973 and 2010”, Zinner et al. 2013.

 

Data availability:

Delete the current statement, it is useless! Please add a link to the data used and the script. This is necessary for transparency and reproducibility.

Comments for author File: Comments.pdf

Author Response

Comments and Suggestions for Authors

The reviewer finds the manuscript well written and contributes to the well-known problem of deforestation in Madagascar. The manuscript is in line with the purpose of the journal and is worthy of publication. However, there is still information and analysis that should be included to improve the study.

 

Comments and Suggestions for Authors

1. The manuscript is a study of deforestation trends in four different protected areas on the island of Madagascar. The study shows that deforestation rates are higher in large and newly established protected areas. However, the causes leading to this deforestation could not be fully explained. The manuscript contributes to the current issue of deforestation in high biodiversity areas. Several improvements that can enhance the manuscript are recommended below.

Thank you for your detailed review and the various comments which allowed us to improve the current manuscript. The contribution would have better fit as a “case report” while the editors moved this contribution to “article”. We have designed this study as case study approach to describe the ongoing deforestation in these four PAs, to test a new approach (buffers), and to highlight the various challenges that oftentimes are very case specific. This contribution is not intended to identify global patterns that apply at Madagascar level. We acknowledge the need to extend the analysis to the entire network of protected areas, but this would require a completely new article. In this manuscript we would like to introduce a new approach for the analysis of a stratified deforestation analyses that is based on zones of a given width extending from the reserve boundary in- and outwards. We exemplify this approach with four different protected areas.

Major and Minor Issues

Title

2. The title should be revised as it currently refers to deforestation processes, but the authors do not have data to explain the processes that lead to deforestation over time, so this is not appropriate.

We removed “processes” from the title and added “stratified”, which now reads: “Exemplifying stratified deforestation in four protected areas in Madagascar”

 

Abstract

1. The abstract is confusing, especially the methods and results should be formulated more clearly. L28: Do you mean “prove”? Prove is a verb, while proof is a noun

The methodology section within the abstract reads now: “We have considered forest versus other land cover within the PAs in “buffers” at a distance of 500m, 2.5km, 5km, 10km from the border of the PA. These buffers were set from the border towards the center of the PA (inside the PAs), and from the border outside the PAs.”

We have corrected “proof” and the last sentence reads now: “All four cases prove to be very challenging to manage. Future conservation activities require tailored interventions to account for site-specific current and potential future threats as detailed in this contribution.”

 

2. L29: site-specific current

The hyphen is missing in many cases, please double check it.

Corrected as suggested, also on l. 284.

 

3. The full name of the abbreviations should be given the first time they are mentioned, e.g., Convention on Biological Diversity (CBD).

Corrected

 

Introduction

4. The introduction is easy to read and understand, but the objective, hypothesis and research questions are missing. Theoretical aspects on protected areas inside/outside are not mentioned, although the authors analyze forest cover inside and outside PAs.

We have included two sentences explaining these aspects, before the four PAs are mentioned: “The pressure on PAs is assumed to be highest close to its border; therefore, a national decree for the management of PAs defines a buffer zone 2.5 km outside the limits of the PAs—so-called “zone de protection”—which allows communities to access some natural resources in order to better protect the PAs. To estimate whether PAs provide the best protection against deforestation, as compared to land outside PAs, and to evaluate the role of the outside buffer zone in ensuring their protection, we analyze the evolution of forest cover according to the distance to the limits of PAs, inside and outside these PAs.”

 

5. Also, key research is not included, please check:

Global patterns of forest loss across IUCN categories of protected areas, Leberger et al. 2019

Predicted climate shifts within terrestrial protected areas worldwide, Hoffmann et al. 2019

Thank you for these interesting references but our current contribution does not touch these aspects. Further analysis will be conducted to consider the IUCN categories and more global aspects, but these considerations need to include/ be extended to the entire network of PAs. Again, we are presenting a case based study and thus refrain from any attempt to generate global (Madagascar-wide) patterns.

 

Methods

6. Table 1. Does the color have a meaning? If so, explain it clearly, as well in Table 2.

We have removed the color in Table 1.

 

7. Simplify Table 2. The total area is the same in most cases. As it is now, it looks confusing and the reader cannot easily find the most important information. Why are some rows colored?

We have simplified Table 2, the areas are now on a single line; we have removed the color from Table 2, and shaded some cells with the following explanation in the caption: “shaded cells indicate formal protection”

 

8. L92: A buffer has a certain size. So what do you mean by over 10 km?

Thank you, we agreed that this is confusing. We have 4 buffers inside the PAs, and for the largest PAs another core area beyond the 10km buffer. In Madagascar, few PAs have such core areas. The longest distance from the border to the core point is 27.6 km. For our case studies, the longest distance is 13.5 km in Tsaratanana. Throughout the text, we now refer to “core area” instead of the Buffer > 10km. We have included a new figure (Figure 2) in the methodology which illustrates these buffers.

The sentence reads now: Selecting the buffers: “Four buffers were chosen from 500m, 2.5km, 5km, and 10km for the inside of a protected site, and the core area beyond the 10km buffer for the largest PAs; the same four buffers were applied reciprocally for the outside (Figure 2).”

 

9. L93-100: Explain why you choose them in a certain order. I don't understand why you choose a buffer of over 10 km.

We have replaced the term “buffer of over 10km” with “the core area”. We started with a buffer of 2.5 km and incremented the distance of further buffers by doubling the distance at each step. We have added “, by doubling the 2.5 distance of the initial buffer” in the explanation. The sentence reads now: “The 5km and 10km buffers were chosen to get more information as one moves further from the protected area borders, by doubling the 2.5km distance of the initial buffer.

 

10. Are there no other small roads apart from those mentioned in the table, and what about forest fragmentation?

There are small roads and trails; only the larger roads are mentioned, i.e., those connecting the rural places to main towns. Other roads are mostly non accessible to cars, and their consideration would bring complexity unrelated to the present study. Therefore, we avoid entering such details. Vieilledent et al. 2018 have done a Madagascar scale analysis of deforestation processes; these authors have described in detail fragmentation. For our four cases, we are only interested in the total forest cover change of a PA (actually, we increase the resolution by introducing the buffers); fragmentation is a consequence of deforestation (or other disturbances) and thus not in the scope of our study.

 

11. Although some causes of deforestation were mentioned in the discussion, no data were presented that could be analyzed. For example, it should be analyzed whether human settlement, population density and distance to the road play a role.

These analyses are beyond the scope of the current contribution. The main objectives of this study is to describe deforestation over the period of 30 years. With four cases only it is not meaningful to link any other factors to deforestation to possibly gain an understanding of drivers and mechanisms of the deforestation process. We do acknowledge the importance of considering these features for future analyses with the consideration of the entire network of terrestrial PAs. The focus of this analysis is on the evolution of deforestation, not on the processes and mechanisms of change per se.

 

12. See: “Deforestation processes in south-western Madagascar over the past 40 years: what can we learn from settlement characteristics?” https://doi.org/10.1016/j.agee.2014.06.008

Thank you for this reference. Ranobe-PK32 lies in SW Madagascar, to the North of the area considered in this reference. The deforestation in Ranobe-PK32 has its own specificities which are better outlined in Gardner et al. 2016, i.e., ref. #51 cited here. We want to avoid confusion and concentrate on the four PAs without comparing to other PAs. To further avoid confusion, we have deleted references to other sites at the end of the introduction.

 

Results

13. L160-173: Some assumptions are mentioned, but they have no basis. These assumptions should be mentioned in the introduction.

We have included the two assumptions in the introduction, in the following sentences: “The pressure on PAs is assumed to be highest close to its border; therefore, a national decree for the management of PAs defines a buffer zone 2.5 km outside the limits of the PAs—so-called “zone de protection”—which allows communities to access some natural resources in order to better protect the PAs. To estimate whether PAs provide the best protection against deforestation, as compared to land outside PAs, and to evaluate the role of the outside buffer zone in ensuring their protection, we analyze the evolution of forest cover according to the distance to the limits of PAs, inside and outside these PAs.”

 

14. 3. The idea of the diagram is good, but the colors are similar to those used above (see Table 1, 2), so the reader cannot make a difference between the buffer sizes. Again, what is >10km? 20, 100km? Also, the data for the other PAs should be included, if not here, at least in the supplementary materials. Why did you use a black background for the figures?

We have changed the figure and adopted the color patterns used for the buffers as in the new Figure 2 in the Methodology section.

We have included a figure for the annual deforestation rates for the two remaining sites, as Figure 6.

We believe that the background color adds visibility and will offer optional figures; we will provide the figures with a white background for the editors to make a choice.

 

15. Table 2 also shows that the total area in Behara is higher than in Tsaratanana (except 2015-17). The cases shown by the authors in Figure 3 correspond to the cases where deforestation was highest.

We agree that our figures and tables were confusing, therefore we used the annual deforestation rates in both old Figures 2 and 3 (new Figures 4 and 6)

 

16. Although there are clear patterns of deforestation, these are not explained with a proper analysis.

We understand your concern; yet, as mentioned previously, this contribution should be considered as a case study report which is based on four cases only. Given the nature of this contribution, we believe that the description of deforestation is scientifically sound and robust. Any attempt to scale up or draw global/ general conclusions would be simply wrong / too speculative.

 

Discussion

17. The first paragraph needs to be restructured, because it is confusing, it should better contain general findings, more theoretical arguments for or against, findings from other regions, followed by detailed information about each PA (what has been done).

The four PAs cannot justify the elaboration of theories as any kind of extrapolation would not be robust. Further analyses are needed that have entire network of PAs in scope; this, however, is beyond the purpose of the current study. To clarify the goal of the contribution, we have added the following sentence at the beginning of the discussion: “This study exemplifies the variation in deforestation patterns using four protected areas in Madagascar. For the time being, sample size is too small, and the four cases cannot be used to derive general patterns or causes for deforestation. Rather, the examples should be considered as an indication of caution against broad generalizations.”

 

18. The discussion focuses mainly on the four study regions, but what about all of Madagascar? There are other publications on other areas that should be considered. These could support the findings and give a broader overview.

Thank you, we do work on a follow up of this contribution. We agree, when considering the full Madagascar scale it makes sense to compare with previously published studies.

 

19. See: “Analysis of deforestation patterns in the central Menabe, Madagascar, between 1973 and 2010”, Zinner et al. 2013.

Yes, this is an important reference, and the central Menabe is highly threatened. We have included this reference early in the discussion after reference 34, and within the last § of the discussion, where we have added some text. The sentence before last in the discussion reads now: “The Anosy region, and the nearby Androy region, has been experiencing extended drought since decades with associated food insecurity and famines leading to migration waves of people towards the west [57–62] and further towards the north, resulting in high deforestation rates in the Menabe [35].”

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present an interesting topic that is the evolution of land cover on and outside of protected areas in order to evaluate the efectiveness of the management. Based on forest PA's the authors evaluate the evolution of deforestation during 30 years either on the PA's and in buffer areas of them. The methodology is adequate and replicable and the results are well presented with the necessary tables and figures. The extension and quality of the bibliography wis correct and the authors cite the most important studies dealing with the goals of the study.

I'd like to do some minor remarks to contribute that the manuscript become a paper for forests journal.

1. Abstract- There is a mistake with word "cornerstone", in line 16 is "cornerstone" and in 34 "corner stone", please clarify this.
2. You use the word "gazzeted" many times, I'm not sure if you mean "protected". Sorry but my english is improveable.
3. Table 1. It contains a lot of interesting and necessary information but maybe is not necessary to colour the header if the columns of the PA's.
4. Table 2 is also interesting but I'd suggest do not use colours to set the onset of protection.
5. Figure 3 is so nice (congratulations !!). Maybe, to a better reading, the information regarding of border distances could be vertical instead of horizontal.
6. The word "coup d'etat" (lines 197 and 266) that is a French word should be in italics. The same for the name of the Ministry (line 236).
7. Line 202, add the number of reference.

It's all; as you can see they are minor change proosals.

I've enjoyed reading the manuscript. Congratulations for the study.

Author Response

Comments and Suggestions for Authors

1. The authors present an interesting topic that is the evolution of land cover on and outside of protected areas in order to evaluate the efectiveness of the management. Based on forest PA's the authors evaluate the evolution of deforestation during 30 years either on the PA's and in buffer areas of them. The methodology is adequate and replicable and the results are well presented with the necessary tables and figures. The extension and quality of the bibliography wis correct and the authors cite the most important studies dealing with the goals of the study.

Thank you very much.

 

I'd like to do some minor remarks to contribute that the manuscript become a paper for forests journal.

2. Abstract- There is a mistake with word "cornerstone", in line 16 is "cornerstone" and in 34 "corner stone", please clarify this.

Corrected to cornerstone

 

3. You use the word "gazzeted" many times, I'm not sure if you mean "protected". Sorry but my english is improveable.

We mean “formerly protected” with a decree.

 

4. Table 1. It contains a lot of interesting and necessary information but maybe is not necessary to colour the header if the columns of the PA's.

We have removed the color from Table 1 and Table 2.

 

5. Table 2 is also interesting but I'd suggest do not use colours to set the onset of protection.

Given that two PAs considered were not protected before 2015, we believe that it is an important aspect to consider, especially because we look at the efficiency of PAs to protect the forests. We have simplified Table 2 and have included a single line for the areas.

 

6. Figure 3 is so nice (congratulations !!). Maybe, to a better reading, the information regarding of border distances could be vertical instead of horizontal.

Thank you, and as much as some of us liked it too, other readers and reviewers found it difficult to read. We have added a figure in the methodology and applied the pattern of the latter in a new version of this figure. We also changed the deforestations rates to annual deforestation rates to ease comparison.

 

7. The word "coup d'etat" (lines 197 and 266) that is a French word should be in italics. The same for the name of the Ministry (line 236).

Coup d’état is also an English word, therefore we have not put it in italics. The Ministry has first capitals and is used as a name, not a description, therefore we have not changed it.

 

8. Line 202, add the number of reference.

Good spotting, thank you. (Vieilledent et al. 2018) has been changed to [18]

 

It's all; as you can see they are minor change proosals.

I've enjoyed reading the manuscript. Congratulations for the study.

Thank you so much

 

Author Response File: Author Response.docx

Reviewer 3 Report

Please make sure all dash, space, a hyphen, en dash, and capital words would be appropriate throughout the manuscript.

Please make sure the font size is in all figures, tables, and text.

Please articulate the novelty and the significance of the paper in the abstract, discussion, and conclusion sections.

The title can be improved.

Abstract:

Line 17: main forest types?

Line 30: this report?

Why do use data from 1990 till 2017? any specific reason?

1. Introduction

Please avoid the merging of sentences.

Introduction can be more elaborate with citations.

2. Methods

I would like to suggest that please write the database and methodology section separately. Lots of ideas are merging.

There is no detailed information about the months, resolution of the dataset. Please describe in detail.

2.1. Case studies

Please describe more about study sites.

2.2 Assessing deforestation

Line 83-85: please cites such statements throughout the manuscript.

I do not understand this section. Author(s) just written pre-processing of the dataset. Further analysis is needed to assess the deforestation in the protected areas.

Recommendation articles: https://doi.org/10.3390/environments4020034

https://www.sciencedirect.com/science/article/pii/S0378112702003353?via%3Dihub

3. Results

The results section is still weak. Need to demonstrate with figures.

There is no accuracy assessment in the manuscript.

Author Response

Comments and Suggestions for Authors

1. Please make sure all dash, space, a hyphen, en dash, and capital words would be appropriate throughout the manuscript.

Done

 

2. Please make sure the font size is in all figures, tables, and text.

Done

 

3. Please articulate the novelty and the significance of the paper in the abstract, discussion, and conclusion sections.

Done

 

4. The title can be improved.

Done, the title reads now: “Exemplifying stratified deforestation in four protected areas in Madagascar”

 

Abstract:

 

5. Line 17: main forest types?

We have added “occurring on the island” and the sentence reads now: “Protected areas (PAs) are a cornerstone for conservation biodiversity. Madagascar as a hotspot for biodiversity, has a network of 114 terrestrial protected areas covering the main forest types encountered on the island.” More information on forest types is given in the following sections and are shown in Figure 1.

 

6. Line 30: this report?

We have changed “report” to “contribution”.

 

7. Why do use data from 1990 till 2017? any specific reason?

We use the latest analyses performed for the forest at the national level, which was published in 2018. These aspects are explained in the methodology and should not be detailed in the abstract.

 

1. Introduction

 

8. Please avoid the merging of sentences.

Thank you, we have avoided the merging of sentences.

 

9. Introduction can be more elaborate with citations.

We have cited the most relevant and recent papers totaling 21 articles.

 

2. Methods

 

10. I would like to suggest that please write the database and methodology section separately. Lots of ideas are merging.

We have added some text to the introduction, in accordance with the following sections. But this contribution does not look at the entire network of PAs in Madagascar and should not be regarded as exemplifying the network. Extrapolation from four cases only would not be appropriate.

 

11. There is no detailed information about the months, resolution of the dataset. Please describe in detail.

These details are given in the references we cite (Vieilledent et al. 2018). The forest cover has not been analyzed in the current contribution; thus, any technical aspects are available the reference we refer to.

 

2.1. Case studies

 

12. Please describe more about study sites.

Given the scope of this study (describing deforestation in and outside four PAs), we believe that the current description of the four case sites is detailing the relevant aspects to gain an understanding in the challenges for managing the sites. Four gigantic volumes on PAs have been published recently (Goodman et al. 2020, ref. # 21). Figure 1 and table 1 present a detailed compendium of the four PAs allowing for a robust contextualization.

 

2.2 Assessing deforestation

 

13. Line 83-85: please cites such statements throughout the manuscript.

It is not fully clear what is expected here. We provide the sources and references for the four PAs in table 1.

 

14. I do not understand this section. Author(s) just written pre-processing of the dataset. Further analysis is needed to assess the deforestation in the protected areas.

We have not redone the analysis of deforestation per se, but used the work published in 2018 by Vieilledent et al. re. forest cover. This is clearly explained in the methodology.

 

15. Recommendation articles: https://doi.org/10.3390/environments4020034 https://www.sciencedirect.com/science/article/pii/S0378112702003353?via%3Dihub

Thank you, these references apply at the global scale while our contribution has a format of case studies.

 

3. Results

 

16. The results section is still weak. Need to demonstrate with figures.

We have changed old Figures 2 and 3 (now new Figures 3 and 4), and we believe that it is more clear now.

 

17. There is no accuracy assessment in the manuscript.

As we have not analyzed the remotely sensed information in our study but we only used already processed information from Vieilledent et al 2018, we do not need to provide an accuracy assessment in our case study paper. This has been done by Vieilledent et al. 2018. To make it clear, we have added the following sentence in the methodology (ll. 136—137 in the track-changed version): “Since we only used already processed information from Vieilledent et al. [18], we do not provide an accuracy assessment in our case study paper.”

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have improved the manuscript and it is suitable for publication.

Author Response

Thank you very much

Reviewer 3 Report

I am sorry to say but I could not find the novelty of this research. Author (s) just analyzed the secondary data from previous study (Vieilledent et al.)

 

Line 82, author mentioned driving forces leading to deforestation. However, line 275-278, author mentioned that “sample size is too small, and the four cases cannot be used to derive general patterns or causes for deforestation” – I cannot understand the actual purpose of this study. There is lack of discussion to understand the actual cause/driving forces of deforestation in Protected areas.

 

Why author selected only and these four protected areas?

 

There is lack of clarity in abstract what the main objectives of this research.

 

Please make sure the database and methodology should be separated.

 

Introduction can more elaborate.

Author Response

Comments and Suggestions for Authors

Line numbering as the version resubmitted without track change

1. I am sorry to say but I could not find the novelty of this research. Author (s) just analyzed the secondary data from previous study (Vieilledent et al.)

 

We made it very clear in our contribution that we have used Vieilledent's data:

Methodology, ll. 124 – 133: “The 1990 and 2000 data were taken from the original data of Harper et al. [23] and resampled by Vieilledent et al. [18] at a resolution of 30 m using a nearest-neighbor interpolation from which they labelled the unclassified data due to cloud cover of 2000 from the 2000 tree cover of Hansen et al. [24] To obtain the forest cover map for the years 2005, 2010, 2015 and 2017, we used the data from Vieilledent et al. [18], which was the result of a combination of the forest cover map of the year 2000 with the annual tree cover loss maps at 30 m spatial resolution [24,25]. For each year, we have given specific names to each of the resulting features to reflect the time of change. Since we only used already processed information from Vieilledent et al. [18], we do not provide an accuracy assessment in our case study paper.”

We clearly outlined the use of this external database, which is a common and legitimate practice. Though we have resampled the data in a novel approach by stratifying deforestation; the buffers approach is central and key to our study which we showcase using four protected areas.

Introduction, ll. 71 – 73, we mention:” Here we report on the driving forces leading to deforestation in four so far relatively under researched sites and with little management.”

i.e., that we make it clear that we consider only these four protected areas and we report only on these four cases. We have 114 terrestrial protected areas in Madagascar and our intention in this contribution was not to look at the Madagascar scale deforestation per se, nor the mechanisms of deforestation. These approaches have been published for Madagascar in various contributions which we all cite in our introduction, in the following:

Introduction, ll. 60 – 64: “While deforestation continues unabated (e.g., [15,18]), studies conducted by Eklund et al. [19] or Yesuf et al. [20] concluded that PAs in Madagascar can slow down deforestation. While Madagascar represents nowadays a network of protected areas across its main forests, there are only few sites that receive the bulk of attention from research, tourism, and NGOs [15,16,21].”

 

2. Line 82, author mentioned driving forces leading to deforestation. However, line 275-278, author mentioned that “sample size is too small, and the four cases cannot be used to derive general patterns or causes for deforestation” – I cannot understand the actual purpose of this study. There is lack of discussion to understand the actual cause/driving forces of deforestation in Protected areas.

We do mention “driving forces leading to deforestation” but we also add “in four so far relatively under researched sites and with little management.” Which makes it very clear that we are not trying to identify a pattern applicable to the 114 protected areas. We have also changed the title to emphasize the objective of this study, which is now “Exemplifying stratified deforestation in four protected areas in Madagascar”

In this contribution, the main purpose is to present a new approach on how to describe deforestation in Madagascar. This is key for a biodiversity hotspot to better understand management challenges as it allows to highlight where the main pressures (deforestation) are taking place in and around a PA. The management of parks and reserves partly rely on external buffers zones (cf. decree), as mentioned in the following:

Introduction, ll. 65 – 71: “… a national decree for the management of PAs defines a buffer zone 2.5 km outside the limits of the PAs—so-called “zone de protection”—which allows communities to access some natural resources in order to better protect the PAs. To estimate whether PAs provide the best protection against deforestation, as compared to land outside PAs, and to evaluate the role of the outside buffer zone in ensuring their protection, we analyze the evolution of forest cover according to the distance to the limits of PAs, inside and outside these PAs.”

This contribution is the first to report on stratified deforestation pattern, inside and outside the park boundaries for Madagascar (and we believe also globally); we showcase the approach identifying the changing deforestation according to buffers in four protected areas. This alone grants novelty (also referring to Revier3’s point 1).

 

3. Why author selected only and these four protected areas?

In our previous revisions, we have highlighted in several occasions, that this contribution would have better fit as a case report and that extrapolation is not possible:

See our answer to reviewer 1: “The contribution would have better fit as a “case report” while the editors moved this contribution to “article”. We have designed this study as case study approach to describe the ongoing deforestation in these four PAs, to test a new approach (buffers), and to highlight the various challenges that oftentimes are very case specific. This contribution is not intended to identify global patterns that apply at Madagascar level. We acknowledge the need to extend the analysis to the entire network of protected areas, but this would require a completely new article. In this manuscript we would like to introduce a new approach for the analysis of a stratified deforestation analyses that is based on zones of a given width extending from the reserve boundary in- and outwards. We exemplify this approach with four different protected areas.”

This aspect is clear in the following sentence:

Abstract, ll. 71 - 73: “Here we present a case study approach reporting on four PAs from the humid forests, dry western forests, and southwestern dry and spiny forests and thickets.”

We have chosen this study design—case study approach—to showcase the novel approach on how to describe deforestation for PAs by use of zones in- and outside of PAs. This allows for a systematic and automated process to analyse any available data sets (as we have done for this study using Vieilledent’s). By design, we can link spatial changes (thanks to stratification) to actual events (e.g., charcoal production from outer buffers towards the core in the course of 30 years).

 

4. There is lack of clarity in abstract what the main objectives of this research.

We have added the following sentence “Exemplifying stratified deforestation in four protected areas in Madagascar”. Please note the word “stratified” which is central and refers to the buffers.

 

5. Please make sure the database and methodology should be separated.

We have the following structure in our methodology, which underlines the chronological work accomplished for this study:

• Case study, together with Figure 1 and Table 1 detailing the four protected areas
• Assessing deforestation

- Database used for the protected areas

- Shapefiles, i.e., the database for the regions of concern for the four protected areas in a Geographical System

- Buffers, explaining the novel methodology implemented here, with Figure 2

- Analysis of the forest cover in the buffers through time

The last two §, i.e., buffers and analysis of forest cover are the methodology versus the preceding § which are linked to databases.

The chronological order followed in the methodology allows to clearly present the novelty of the approach.

 

6. Introduction can more elaborate.

We have improved our introduction, following review round 1, and have added the following: “The pressure on PAs is assumed to be highest close to its border; therefore, a national decree for the management of PAs defines a buffer zone 2.5 km outside the limits of the PAs—so-called “zone de protection”—which allows communities to access some natural resources in order to better protect the PAs. To estimate whether PAs provide the best protection against deforestation, as compared to land outside PAs, and to evaluate the role of the outside buffer zone in ensuring their protection, we analyze the evolution of forest cover according to the distance to the limits of PAs, inside and outside these PAs.”

 

Author Response File: Author Response.docx