The Spatiotemporal Dynamics of the Landscape of the Itombwe Nature Reserve and Its Periphery in South Kivu, the Democratic Republic of Congo

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I think the introduction is too long and much of this content is mentioned in the Discussion or should be there. Please identify the problems in the introduction and in the Discussion detail them by referring to relevant publications. The conclusions are, in my opinion, appropriate, but should be verified if they do not appear in the introduction.

The changes described in 3.2. should be compared with direct human activity, for example in the discussion, but the changes should be named and the discussion should refer to the changes so defined (shown in the diagrams). An interesting result would be to indicate whether the forest in the reserve has the natural potential to recover without human intervention? It is important to know whether the shrinkage of the forest is solely due to agricultural expansion and war damage or also due to climate change?

Repetitions of content appear, e.g. the work of Cabala et al. [63] (567, 615). Please reduce repetitions, they stretch the content excessively.  Does nature, while remaining under less human pressure, have the potential to recover its biodiversity, can areas with savannah spontaneously become overgrown with forest, or are these changes irreversible? What happened between 2007 and 2013 to cause such a large change in the area of forest in the reserve?

A lot of relevant content and conclusions by other authors are given in the discussion, but there is no acknowledgement here that the authors of this paper agree with them or not. In the conclusions at the end such statements are made and that is fine. They should be generalised here, but some reference of one's own opinion in the discussion should also be there.

Comments on the Quality of English Language

(28) which was employed to obtain robust landscape replications from 83 2.5 km plots (i.e., 6.25 km²) sampled in each zone. -  better is:

 which was employed to obtain robust landscape replications from 83 plots (i.e., 6.25 km²) sampled in each zone.

(131) (DR Congo) but elsewhere (DRC)

(135) Nature Re-serve - Nature Reserve

The Albertine Rift (197) is this the same as Albertine Graben (204) ? Please standardise the naming.

(222) In the key for the Fig.1 is used different phrases thanin  text i.e.  RN_ Itombwe - Itombwe Nature Reserve, Hinterland - periphery, buffer zone. Please standardise the naming.

(222) lake of water - better phrase is ‘water bodies and pathways’

(247) analyses were conducted at two levels: at the reserve level (1990 to 2024) and at the periphery level (1990 to 2024). Why is the date range repeated?

Figure 1 and Figure 2 - Please standardise the font size

(326) - (m²)

Author Response

Comment 1: I think the introduction is too long, and much of this content is mentioned in the Discussion or should be there. Please identify the problems in the introduction and the discussion and detail them by referring to relevant publications. The conclusions are, in my opinion, appropriate but should be verified if they do not appear in the introduction.

Response: We agree with this comment thereby it is acknowledged that the initial introduction was excessively protracted and comprised details that were more suitable for the discussion. To address this issue, the introduction has been streamlined to prioritise essential background information and research objectives. The introduction has been synthesized to remove detailed descriptions of the underlying drivers of deforestation and anthropogenic pressures, such as the impacts of armed conflict, agricultural encroachment, and governance issues (as reported in [1-5, 15, 18-23]). These factor analyses are now developed in the discussion section. (pages 2 and 3, line 48-106) 

Comment 2: The changes described in 3.2. should be compared with direct human activity, for example in the discussion, but the changes should be named, and the discussion should refer to the changes so defined (shown in the diagrams). An interesting result would be to indicate whether the forest in the reserve has the natural potential to recover without human intervention. It is important to know whether the shrinkage of the forest is solely due to agricultural expansion and war damage or also due to climate change.

Response: - 

• We agree with this comment thereby in the revised version, changes are directly compared to human activities such as agricultural encroachment, war-related damage, and illegal resource exploitation (e.g., [15, 18–23]), as clearly illustrated in our diagrams (Figure 3). (page 9, line 290 -292). This comparison provides a framework for understanding how human actions are driving landscape transformation.
• According to this comment: does the forest of the Itombwe Nature Reserve have an intrinsic capacity for natural regeneration in the absence of human intervention? : We agree with this comment; thereby, the fluctuations indicate that the reserve's forest lacks an intrinsic capacity for recovery without human intervention. It is only through the implementation of conservation measures stemming from human actions, such as agricultural expansion, charring, and war damage, that the reserve's forest can be restored. (page 9, line 303-307)
• It is important to know whether the shrinkage of the forest is solely due to agricultural expansion and war damage or also due to climate change? Response: We agree with this comment thereby, the decline observed in the forest is much more related to direct human activities (agricultural expansion, charcoal production, war damage) than to climate change factors. refer to (page 9, line 303-307)
• Comment 3. Repetitions of content appear, e.g. the work of Cabala et al. [63] (567, 615). Please reduce repetitions, they stretch the content excessively.
• Response: We agree with this comment thereby, we've made an effort to reduce repetition, starting with the introduction. (pages 2 and 3, line 48-106) 
• Comment 4. Does nature, while remaining under less human pressure, have the potential to recover its biodiversity, can areas with savannah spontaneously become overgrown with forest, or are these changes irreversible?
• Response: We agree with this comment thereby, given the documented fluctuations in the savannah, it is inevitable that the area will undergo conversion to forest without human intervention, such as the promotion of agroforestry and silviculture. This assertion is substantiated by the observation that substantial transformations of savannahs tend to favor agricultural areas. (page 10, line 323-327)
• Comment 5: What happened between 2007 and 2013 to cause such a large change in the area of forest in the reserve?
• Response: We agree with this comment thereby the observed peak of forest loss between 2007 and 2013, after the reserve's establishment, suggests that the local population engaged in anarchic resource exploitation as a form of defiance against the reserve's managers. This finding indicates that the designation of this forest as a reserve had not yet garnered the support of the local population, who regarded this decision as a form of expropriation. (page 23, line 532-536)
• Comment : Comments on the Quality of English Language

  (28) which was employed to obtain robust landscape replications from 83 2.5 km plots (i.e., 6.25 km²) sampled in each zone. -  better is: 

   which was employed to obtain robust landscape replications from 83 plots (i.e., 6.25 km²) sampled in each zone. Response: We agree with this comment thereby, we already standardised (page 6, line 220)

  (131) (DR Congo) but elsewhere (DRC): Reponse: We agree with this comment thereby, we already standardised (page 3, line 110 and elswhere)

  (135) Nature Re-serve - Nature Reserve: Response: We agree with this comment thereby, we already standardised  (Elswhere)

  The Albertine Rift (197) is this the same as Albertine Graben (204) ? Response: We agree with this comment thereby, we standardised (page 3, line 127 and 134)

  (222) In the key for the Fig.1 is used different phrases thanin  text i.e.  RN_ Itombwe - Itombwe Nature Reserve, Hinterland - periphery, buffer zone. Please standardise the naming. Response: We agree with this comment thereby, we already standardised (page 4, line 150)

  (222) lake of water - better phrase is ‘water bodies and pathways. Response: We agree with this comment thereby, we already standardised (page 4, line 150)

  (247) analyses were conducted at two levels: at the reserve level (1990 to 2024) and at the periphery level (1990 to 2024). Why is the date range repeated?: Response: We agree with this comment thereby, we already Corrected (page 6, line 174) 

  Figure 1 and Figure 2 - Please standardise the font size: Response: We agree with this comment thereby, we already standardised (page 4, line 150 and page 7, line 238)

   

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for submitting your manuscript, "Spatiotemporal Dynamics of the Landscape of the Itombwe Nature Reserve and Its Periphery in South Kivu, Democratic Republic of Congo." I appreciate the time and effort that clearly went into assembling this work. The long-term monitoring of land cover changes in a geopolitically and ecologically complex region is both challenging and important. 

That said, after a thorough reading of the manuscript, I must be honest in expressing that I have several serious concerns regarding its structure, analytical rigor, and overall readiness for publication in a journal like Land. While the subject matter is clearly relevant to the journal's scope, the current version lacks the clarity, scientific sharpness, and statistical robustness expected for publication at this level. 

Introduction

The Introduction is overwritten and loses focus early on. Key ideas are repeated across multiple paragraphs, especially those related to conflict and resource exploitation, which weakens their overall impact. The text drifts between general background and specific details without a clear narrative. Most importantly, the study’s objectives and rationale only become apparent toward the end of the section, making it difficult for the reader to follow the logic or understand the novelty of your approach. Furthermore, the introduction of the landscape metrics (PLAND and LPI) appears abruptly, without sufficient integration into the scientific context or justification for their choice. This section needs substantial tightening and restructuring to present a focused, logical, and clearly justified research aim.

Materials and Methods

Your methodological choices are appropriate in theory. However, in practice, the section lacks transparency and critical detail. The Random Forest classifier is introduced without stating its configuration (e.g., number of trees, cross-validation approach), and it is unclear how training and validation points were selected or distributed across years and land cover classes. The justification for selecting 83 plots is not explained, nor is the spatial sampling strategy across the reserve and periphery. Of most concern is the application of ANOVA to compare PLAND and LPI values across years and zones. This approach treats each year as an independent group, ignoring the temporal structure of the data. In time series, observations across years are typically autocorrelated, and standard ANOVA assumes independence between groups — an assumption probably violated here. No attempt was made to assess or account for this autocorrelation, and no diagnostic tests for normality or homogeneity of variances were reported. The presence of numerous extreme outliers and visibly skewed distributions in the results further undermines the appropriateness of ANOVA in this context.

Results

In Figures 3, 4 and 5, the number and scale of outliers are striking, yet these are not acknowledged or discussed in the text. Many distributions appear skewed, with long tails and inconsistent variances between years and zones, further suggesting that parametric testing may not be appropriate. Additionally, the figures are overwhelming. This not only makes interpretation difficult but also limits the communicative power of what could be a strong dataset. 

Comments on the Quality of English Language

The writing is overly wordy, repetitive, and often lacks clarity. Transitions between sentences and paragraphs are weak.

Author Response

Comment1. The Introduction is overwritten and loses focus early on. Key ideas are repeated across multiple paragraphs, especially those related to conflict and resource exploitation, which weakens their overall impact. The text drifts between general background and specific details without a clear narrative. Most importantly, the study’s objectives and rationale only become apparent toward the end of the section, making it difficult for the reader to follow the logic or understand the novelty of your approach. Furthermore, the introduction of the landscape metrics (PLAND and LPI) appears abruptly, without sufficient integration into the scientific context or justification for their choice. This section needs substantial tightening and restructuring to present a focused, logical, and clearly justified research aim. 

Response: We agree with this comment thereby,the general information on deforestation, conflict and resource exploitation has been condensed, as it was found to be redundant in several paragraphs. The salient points on the aforementioned issues, which have been previously discussed and substantiated by references [1-5, 15, 18-23], have been summarised in the introduction. This approach was adopted to minimise redundancy and ensure the text is concise (page 2, lines 48-75).

- We agree with this comment thereby, the new version reorganizes the presentation, moving from a global context (forest extent of the Congo Basin and diversity of ecosystems in the DRC) to a more focused description of the challenges specific to the Itombwe Nature Reserve. We have introduced the problem progressively: first, the general context, then the anthropic impacts and finally the lack of detailed studies linking landscape dynamics in the reserve and periphery; this is even the reason for our study. (page 2, lines 76-86).

- We agree with this comment thereby in contrast to the initial version; the objectives and rationale of our approach are now clearly delineated in the introduction. It is evident from the outset that the objective of the present study is to address a lacuna in the extant literature concerning spatiotemporal dynamics, using a comparative analysis of the reserve and its periphery, utilising remote sensing methodologies and landscape metrics. This revised version incorporates the presentation of landscape measures (indices). It is briefly explained that PLAND quantifies the extent of each habitat and LPI assesses its spatial configuration, thereby specifying their complementarity in answering the research questions (page 2, lines 87-97 and page 3, lines 98-102). 

Comment 2: Methodology: Your methodological choices are appropriate in theory. However, in practice, the section lacks transparency and critical detail. The Random Forest classifier is introduced without stating its configuration (e.g., number of trees, cross-validation approach), and it is unclear how training and validation points were selected or distributed across years and land cover classes. The justification for selecting 83 plots is not explained, nor is the spatial sampling strategy across the reserve and periphery. Of most concern is the application of ANOVA to compare PLAND and LPI values across years and zones. This approach treats each year as an independent group, ignoring the temporal structure of the data. In time series, observations across years are typically autocorrelated, and standard ANOVA assumes independence between groups — an assumption probably violated here. No attempt was made to assess or account for this autocorrelation, and no diagnostic tests for normality or homogeneity of variances were reported. The presence of numerous extreme outliers and visibly skewed distributions in the results further undermines the appropriateness of ANOVA in this context.

Response: - We agree with this comment; thereby, we have now explicitly stated that the Random Forest classifier was configured with 100 trees and that the maximum depth was determined through preliminary cross-validation (see Section 2.2.3). (page 6, 211-216).

- We agree with this comment; thereby, we have clarified that sampling points were identified using high-resolution false-color composites available via Google Earth Engine, ensuring that points were distributed proportionally across land cover classes. Approximately 70% of the points were randomly assigned for training and 30% for validation. (page 5, lines 200-206).

- We agree with this comment thereby; Our stratified random sampling approach was designed to ensure a balanced representation of all cardinal and intercardinal directions around the reserve. We have shown that the selection of 83 plots was based on preliminary analyses aimed at achieving statistical accuracy while capturing landscape heterogeneity, thus providing a reliable basis for comparing the reserve and its periphery. (page 6, lines 218-237).

Comment 3: Results

In Figures 3, 4 and 5, the number and scale of outliers are striking, yet these are not acknowledged or discussed in the text. Many distributions appear skewed, with long tails and inconsistent variances between years and zones, further suggesting that parametric testing may not be appropriate. Additionally, the figures are overwhelming. This not only makes interpretation difficult but also limits the communicative power of what could be a strong dataset. 

-  We agree with this comment; thereby, as illustrated in Figures 3, 4, and 5, we meticulously refined our analyses by assessing the normality of our data. This process revealed that the AVOVA test was not necessary. Consequently, we resorted to non-parametric tests, utilising the Kruskal-Wallis test. Furthermore, the outliers were interpreted after the results presentation (see Figures 3, 4, and 5). Furthermore, we have made the figures more legible by correcting them and removing the cumbersome figures (p-value values) that were displayed each time in the initial figures. (Page 18, lines 457-469).

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

Thank you for the thorough and extended revisions you have made. Your manuscript is now considerably stronger.

Author Response

 First  Reviewer. Comment 1: I think the introduction is too long, and much of this content is mentioned in the Discussion or should be there. Please identify the problems in the introduction and the discussion and detail them by referring to relevant publications. The conclusions are, in my opinion, appropriate but should be verified if they do not appear in the introduction.

Response: We agree with this comment thereby it is acknowledged that the initial introduction was excessively protracted and comprised details that were more suitable for the discussion. To address this issue, the introduction has been streamlined to prioritise essential background information and research objectives. The introduction has been synthesized to remove detailed descriptions of the underlying drivers of deforestation and anthropogenic pressures, such as the impacts of armed conflict, agricultural encroachment, and governance issues (as reported in [1-5, 15, 18-23]). These factor analyses are now developed in the discussion section. (pages 2 and 3, line 48-106) 

Comment 2: The changes described in 3.2. should be compared with direct human activity, for example in the discussion, but the changes should be named, and the discussion should refer to the changes so defined (shown in the diagrams). An interesting result would be to indicate whether the forest in the reserve has the natural potential to recover without human intervention. It is important to know whether the shrinkage of the forest is solely due to agricultural expansion and war damage or also due to climate change.

Response: - 

• We agree with this comment thereby in the revised version, changes are directly compared to human activities such as agricultural encroachment, war-related damage, and illegal resource exploitation (e.g., [15, 18–23]), as clearly illustrated in our diagrams (Figure 3). (page 9, line 290 -292). This comparison provides a framework for understanding how human actions are driving landscape transformation.
• According to this comment: does the forest of the Itombwe Nature Reserve have an intrinsic capacity for natural regeneration in the absence of human intervention? : We agree with this comment; thereby, the fluctuations indicate that the reserve's forest lacks an intrinsic capacity for recovery without human intervention. It is only through the implementation of conservation measures stemming from human actions, such as agricultural expansion, charring, and war damage, that the reserve's forest can be restored. (page 9, line 303-307)
• It is important to know whether the shrinkage of the forest is solely due to agricultural expansion and war damage or also due to climate change? Response: We agree with this comment thereby, the decline observed in the forest is much more related to direct human activities (agricultural expansion, charcoal production, war damage) than to climate change factors. refer to (page 9, line 303-307)
• Comment 3. Repetitions of content appear, e.g. the work of Cabala et al. [63] (567, 615). Please reduce repetitions, they stretch the content excessively.
• Response: We agree with this comment thereby, we've made an effort to reduce repetition, starting with the introduction. (pages 2 and 3, line 48-106) 
• Comment 4. Does nature, while remaining under less human pressure, have the potential to recover its biodiversity, can areas with savannah spontaneously become overgrown with forest, or are these changes irreversible?
• Response: We agree with this comment thereby, given the documented fluctuations in the savannah, it is inevitable that the area will undergo conversion to forest without human intervention, such as the promotion of agroforestry and silviculture. This assertion is substantiated by the observation that substantial transformations of savannahs tend to favor agricultural areas. (page 10, line 323-327)
• Comment 5: What happened between 2007 and 2013 to cause such a large change in the area of forest in the reserve?
• Response: We agree with this comment thereby the observed peak of forest loss between 2007 and 2013, after the reserve's establishment, suggests that the local population engaged in anarchic resource exploitation as a form of defiance against the reserve's managers. This finding indicates that the designation of this forest as a reserve had not yet garnered the support of the local population, who regarded this decision as a form of expropriation. (page 23, line 532-536)
• Comment : Comments on the Quality of English Language

  (28) which was employed to obtain robust landscape replications from 83 2.5 km plots (i.e., 6.25 km²) sampled in each zone. -  better is: 

   which was employed to obtain robust landscape replications from 83 plots (i.e., 6.25 km²) sampled in each zone. Response: We agree with this comment thereby, we already standardised (page 6, line 220)

  (131) (DR Congo) but elsewhere (DRC): Reponse: We agree with this comment thereby, we already standardised (page 3, line 110 and elswhere)

  (135) Nature Re-serve - Nature Reserve: Response: We agree with this comment thereby, we already standardised  (Elswhere)

  The Albertine Rift (197) is this the same as Albertine Graben (204) ? Response: We agree with this comment thereby, we standardised (page 3, line 127 and 134)

  (222) In the key for the Fig.1 is used different phrases thanin  text i.e.  RN_ Itombwe - Itombwe Nature Reserve, Hinterland - periphery, buffer zone. Please standardise the naming. Response: We agree with this comment thereby, we already standardised (page 4, line 150)

  (222) lake of water - better phrase is ‘water bodies and pathways. Response: We agree with this comment thereby, we already standardised (page 4, line 150)

  (247) analyses were conducted at two levels: at the reserve level (1990 to 2024) and at the periphery level (1990 to 2024). Why is the date range repeated?: Response: We agree with this comment thereby, we already Corrected (page 6, line 174) 

  Figure 1 and Figure 2 - Please standardise the font size: Response: We agree with this comment thereby, we already standardised (page 4, line 150 and page 7, line 238)

  Second Reviewer: 

  Comment 1: I think the introduction is too long, and much of this content is mentioned in the Discussion or should be there. Please identify the problems in the introduction and the discussion and detail them by referring to relevant publications. The conclusions are, in my opinion, appropriate but should be verified if they do not appear in the introduction.

  Response: We agree with this comment thereby it is acknowledged that the initial introduction was excessively protracted and comprised details that were more suitable for the discussion. To address this issue, the introduction has been streamlined to prioritise essential background information and research objectives. The introduction has been synthesized to remove detailed descriptions of the underlying drivers of deforestation and anthropogenic pressures, such as the impacts of armed conflict, agricultural encroachment, and governance issues (as reported in [1-5, 15, 18-23]). These factor analyses are now developed in the discussion section. (pages 2 and 3, line 48-106)

  Comment 2: The changes described in 3.2. should be compared with direct human activity, for example in the discussion, but the changes should be named, and the discussion should refer to the changes so defined (shown in the diagrams). An interesting result would be to indicate whether the forest in the reserve has the natural potential to recover without human intervention. It is important to know whether the shrinkage of the forest is solely due to agricultural expansion and war damage or also due to climate change.

  Response: -

  • We agree with this comment thereby in the revised version, changes are directly compared to human activities such as agricultural encroachment, war-related damage, and illegal resource exploitation (e.g., [15, 18–23]), as clearly illustrated in our diagrams (Figure 3). (page 9, line 290 -292). This comparison provides a framework for understanding how human actions are driving landscape transformation.
  • According to this comment: does the forest of the Itombwe Nature Reserve have an intrinsic capacity for natural regeneration in the absence of human intervention? : We agree with this comment; thereby, the fluctuations indicate that the reserve's forest lacks an intrinsic capacity for recovery without human intervention. It is only through the implementation of conservation measures stemming from human actions, such as agricultural expansion, charring, and war damage, that the reserve's forest can be restored. (page 9, line 303-307)
  • It is important to know whether the shrinkage of the forest is solely due to agricultural expansion and war damage or also due to climate change? Response: We agree with this comment thereby, the decline observed in the forest is much more related to direct human activities (agricultural expansion, charcoal production, war damage) than to climate change factors. refer to (page 9, line 303-307)
  • Comment 3. Repetitions of content appear, e.g. the work of Cabala et al. [63] (567, 615). Please reduce repetitions, they stretch the content excessively.
  • Response: We agree with this comment thereby, we've made an effort to reduce repetition, starting with the introduction. (pages 2 and 3, line 48-106)
  • Comment 4. Does nature, while remaining under less human pressure, have the potential to recover its biodiversity, can areas with savannah spontaneously become overgrown with forest, or are these changes irreversible?
  • Response: We agree with this comment thereby, given the documented fluctuations in the savannah, it is inevitable that the area will undergo conversion to forest without human intervention, such as the promotion of agroforestry and silviculture. This assertion is substantiated by the observation that substantial transformations of savannahs tend to favor agricultural areas. (page 10, line 323-327)
  • Comment 5: What happened between 2007 and 2013 to cause such a large change in the area of forest in the reserve?
  • Response: We agree with this comment thereby the observed peak of forest loss between 2007 and 2013, after the reserve's establishment, suggests that the local population engaged in anarchic resource exploitation as a form of defiance against the reserve's managers. This finding indicates that the designation of this forest as a reserve had not yet garnered the support of the local population, who regarded this decision as a form of expropriation. (page 23, line 532-536)
  • Comment : Comments on the Quality of English Language

    (28) which was employed to obtain robust landscape replications from 83 2.5 km plots (i.e., 6.25 km²) sampled in each zone. - better is:

    which was employed to obtain robust landscape replications from 83 plots (i.e., 6.25 km²) sampled in each zone. Response: We agree with this comment thereby, we already standardised (page 6, line 220)

    (131) (DR Congo) but elsewhere (DRC): Reponse: We agree with this comment thereby, we already standardised (page 3, line 110 and elswhere)

    (135) Nature Re-serve - Nature Reserve: Response: We agree with this comment thereby, we already standardised (Elswhere)

    The Albertine Rift (197) is this the same as Albertine Graben (204) ? Response: We agree with this comment thereby, we standardised (page 3, line 127 and 134)

    (222) In the key for the Fig.1 is used different phrases thanin text i.e. RN_ Itombwe - Itombwe Nature Reserve, Hinterland - periphery, buffer zone. Please standardise the naming. Response: We agree with this comment thereby, we already standardised (page 4, line 150)

    (222) lake of water - better phrase is ‘water bodies and pathways. Response: We agree with this comment thereby, we already standardised (page 4, line 150)

    (247) analyses were conducted at two levels: at the reserve level (1990 to 2024) and at the periphery level (1990 to 2024). Why is the date range repeated?: Response: We agree with this comment thereby, we already Corrected (page 6, line 174)

    Figure 1 and Figure 2 - Please standardise the font size: Response: We agree with this comment thereby, we already standardised (page 4, line 150 and page 7, line 238)

Author Response File: Author Response.docx