Vegetation Type Mapping in Southern Patagonia and Its Relationship with Ecosystem Services, Soil Carbon Stock, and Biodiversity

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript "Vegetation Types Mapping in Southern Patagonia and its relationship with ecosystem services, soil carbon stock, and biodiversity" explores the spatial patterns of vegetation types (VTs) in Santa Cruz Province, Southern Patagonia, Argentina. The study uses optical Earth observation data to determine VTs related to climatic, topographic, and spectral variables. It assesses the spatial relationship between VTs, potential biodiversity, ecosystem services, and soil organic content. The study involves extensive field sampling and employs Google Earth Engine for modeling and validation. It also examines the relationship between VTs and ecosystem functions like biodiversity conservation and natural resource management.

 

The paper presents a comprehensive and detailed exploration of vegetation types in Southern Patagonia. However, there are areas that could be improved or require attention:

 

Clarity and Conciseness: The manuscript is extensive and could benefit from a more concise presentation of information. Some sections might be overly detailed for a journal manuscript and could be streamlined for clarity.

 

Methodological Details: While the study is comprehensive, some methodological details, especially in terms of data analysis and environmental predictors, could be elaborated further for better clarity.

 

Statistical Analysis: The manuscript could be strengthened by including more robust statistical analyses to support its findings.

 

Integration with Broader Environmental Context: The paper could benefit from a discussion on how its findings relate to broader environmental and ecological studies, particularly focusing on climate change and its impact on vegetation patterns.

 

Comparative Analysis: Including comparisons with similar studies in other regions could provide a broader context and enhance the manuscript's contribution to the field.

 

Use of Recent Literature: Incorporating more recent studies could ensure the paper's relevance and topicality, especially in the fast-evolving field of ecological research.

Line 45:  add this recent reference:Fathy et al. (2023) Assessing geochemical and natural radioactivity impacts of Hamadat phosphatic mine through radiological indices. PLoS ONE 18(8): e0287422. https://doi.org/10.1371/journal.pone.0287422.

Overall, the manuscript offers valuable insights into vegetation types mapping in Patagonia, but addressing these points could improve its overall impact and contribution to the field.

Comments on the Quality of English Language

Moderate editing of English language required

Author Response

The manuscript "Vegetation Types Mapping in Southern Patagonia and its relationship with ecosystem services, soil carbon stock, and biodiversity" explores the spatial patterns of vegetation types (VTs) in Santa Cruz Province, Southern Patagonia, Argentina. The study uses optical Earth observation data to determine VTs related to climatic, topographic, and spectral variables. It assesses the spatial relationship between VTs, potential biodiversity, ecosystem services, and soil organic content. The study involves extensive field sampling and employs Google Earth Engine for modelling and validation. It also examines the relationship between VTs and ecosystem functions like biodiversity conservation and natural resource management.

Thanks for the comments.

Clarity and Conciseness: The manuscript is extensive and could benefit from a more concise presentation of information. Some sections might be overly detailed for a journal manuscript and could be streamlined for clarity.

Thank you for the comments. We followed also comments from Reviewers 2 and 3 that requested methodological details.

 Methodological Details: While the study is comprehensive, some methodological details, especially in terms of data analysis and environmental predictors, could be elaborated further for better clarity.

The methodological section has been improved following the advice from Reviewer 2 and 3.

Statistical Analysis: The manuscript could be strengthened by including more robust statistical analyses to support its findings.

The manuscript contains main statistical analysis used in international literature. We improved the analysis following the advice of Reviewer 2 by adding in Figure 2 simple linear regression lines with statistics.

Integration with Broader Environmental Context: The paper could benefit from a discussion on how its findings relate to broader environmental and ecological studies, particularly focusing on climate change and its impact on vegetation patterns.

Thanks for the comments. We have incorporated in the Discussion section this aspect with a new reference (Afuye et al. 2021).

Comparative Analysis: Including comparisons with similar studies in other regions could provide a broader context and enhance the manuscript's contribution to the field. Use of Recent Literature: Incorporating more recent studies could ensure the paper's relevance and topicality, especially in the fast-evolving field of ecological research.

Thanks for the comments. We have incorporated in the Discussion section this aspect with new references (Aghababaei et al. 2021; Pech-May et al. 2022; Marín Del Valle and Jiang 2022).

Line 45:  add this recent reference: Fathy et al. (2023) Assessing geochemical and natural radioactivity impacts of Hamadat phosphatic mine through radiological indices. PLoS ONE 18(8): e0287422. https://doi.org/10.1371/journal.pone.0287422.

The reference has been added.

Reviewer 2 Report

Comments and Suggestions for Authors

There is only text in the outputs. It is necessary to divide them into several points and add numerical information from the results. Conclusions should be a concise and concentrated summary of the results.

Comments for author File: Comments.pdf

Author Response

1. The manuscript, Mapping Vegetation Types in Southern Patagonia and Its Relationship to Ecosystem Services, Soil Carbon Stocks and Biodiversity, examines spatial patterns of vegetation types associated with climatic, topographic and spectral variables in the province of Santa Cruz (Southern Patagonia, Argentina). to improve understanding of land use at a regional scale. In addition, the authors examined the spatial relationships between vegetation types, potential biodiversity, and soil organic matter content in the study region.

 

Thanks for the comments.

 

2. What is original and relevant to the field in this study is an accurate map of land cover and vegetation types, which provides important information for ecological research, biodiversity conservation, vegetation management and restoration, and regional policy decision-making.

 

Thanks for the comments.

 

3. The article, in comparison with other published material, shows that potential biodiversity was higher in bushland compared to broadleaf forest. In the present study, classification of land cover and vegetation types in Santa Cruz Province using multi-temporal Landsat imagery and climate and topographic variables on Google Earth Engine achieved high levels of accuracy for most of the 19 classes considered in the evaluation.

 

Thanks for the comments.

 

4. In terms of methodology, the authors should consider increasing the image resolution from 30 to 10 m. Such a site may contain very different types of vegetation. The authors write that bare soil was sometimes classified as shrubland, or ephemeral bodies of water, or grassland.

 

We appreciate the reviewer's comment. For the regional scale of the study, converting the map from a resolution of 30 to 10 m implies multiplying the volume of information by 9 (a 30x30m pixel fits 9 10x10m pixels). Since it is a very large surface (244,458 km²), this causes the processing capacity in Google Earth Engine cloud-based computing platform to be exceeded.

 

5. The conclusions are not fully consistent with the evidence and arguments presented. They are text only and provide general language and suggestions for continued research to improve knowledge related to the impacts of grazing and logging on grassland vegetation and ecosystem services. The conclusions should be divided into several paragraphs and specific numerical information should be added from the results. Conclusions should be a concise and concentrated summary of the results.

 

We improved the Conclusions following the Reviewer’s advice.

 

6. There are no comments on the tables, but on Figure 2 it would be nice to add regression lines.

We incorporated in Figure 2 three simple linear regression lines and the statistics. Also, we incorporated this in the text.

 

Reviewer 3 Report

Comments and Suggestions for Authors

1. Only Random Forest has been used to access the performance of classification. More ML algorithms need to be used and the results need to be compared with RF algorithm to state why RF performed better than others.

2. Discussion part needs to be enhanced

3. Conclusion needs to be more detailed

Comments on the Quality of English Language

It is fine.

Author Response

Only Random Forest has been used to access the performance of classification. More ML algorithms need to be used and the results need to be compared with RF algorithm to state why RF performed better than others.

The algorithm Random Forest used for the classification process that consists of an ensemble of decision trees that delivers the modal class of the total set of results has been used worldwide. This classification algorithm’s success for land cover and VT classification has been used in broader environmental and ecological studies around the world [47, 56,57,58]. As a result of this, we determined an Overall Accuracy of 90.4% and a Kappa coefficient of 0.87, and User’s (UA) and producer’s (PA) accuracies values for most land cover categories were high-er than 0.8, highlighting a very good performance of the land use cover map obtained in the present work.

 

Discussion part needs to be enhanced

Discussion has been enhanced following specific advice from other reviewers.

 

Conclusion needs to be more detailed

We improved the Conclusions following the Reviewer’s advice.

 

Reviewer 4 Report

Comments and Suggestions for Authors

In this manuscript, the authors used random forest supervised classification method to classify the land cover type of Santa Cruz province from environmental variables and tested the correlation between land cover type and biodiversity, soil organic carbon and ecological service.

Major comments:

The predictors used in the random forest test are not very clear from the descriptions in the method section. As far as I understand, the image map and the environmental covariates are used to predict the land cover type, while the true land cover type is determined by the 59285 sampled sites through empirical survey. It would be helpful if the authors could provide some descriptions about the predictors and the true land cover type in the random forest test.

Additionally, I am curious that which predictor, be it an image pixel or an environmental variable, has the greatest predicting power to land cover type? It may be out of the scope of this study but I would be excited to learn about it.

Minor comments:

Line 43: Do you mean "to provide the basis for understanding"? Also, I am confused by the human activity pressure on anthropogenic processes. Aren't the two things supposed to be closely-related, if not identical?

Line 89: Where is item (i)?

Line 194: The citations appear to be in parentheses rather than brackets. 

Comments on the Quality of English Language

The use of preposition and clause in this manuscript poses some difficulty for readers to understand the intention of the authors. However, the main message of the sentences can still be grasped by looking at the context.

Line 51-56: The grammar structure appears to be rarely encountered, if allowed, in academic literature. It may be helpful to separate the sentence to shorter sentences to make the structure of each sentences clear.

Line 110: What does "de aridity index" mean in English?

 

Author Response

Major comments:

The predictors used in the random forest test are not very clear from the descriptions in the method section. As far as I understand, the image map and the environmental covariates are used to predict the land cover type, while the true land cover type is determined by the 59285 sampled sites through empirical survey. It would be helpful if the authors could provide some descriptions about the predictors and the true land cover type in the random forest test.

Yes, first we obtained covariate maps generated or uploaded to the Google Earth Engine and together with the 59,285 sampled sites through field survey (corresponding to 19 major categories of land cover) we fit a land cover type map. We used the Random Forest approach to generate an optimum decision tree for classifying the data based on a set of observations. The predictor variables are well described in section “2.2 Environmental predictors for land cover map”.

Additionally, I am curious that which predictor, be it an image pixel or an environmental variable, has the greatest predicting power to land cover type? It may be out of the scope of this study but I would be excited to learn about it.

Instead it is out of the scope of this study, by calculating the variables importance as mean decrease in prediction accuracy before and after permuting a variable the main variables were spectral bands B2 (blue), B3 (green) and B4 (red), followed by altitude.

Minor comments:

Line 43: Do you mean "to provide the basis for understanding"? Also, I am confused by the human activity pressure on anthropogenic processes. Aren't the two things supposed to be closely-related, if not identical?

This paragraph has been clarified.

 

Line 89: Where is item (i)?

This has been corrected.

 

Line 194: The citations appear to be in parentheses rather than brackets. 

The cite 42 appear in brackets.

 

Comments on the Quality of English Language

The use of preposition and clause in this manuscript poses some difficulty for readers to understand the intention of the authors. However, the main message of the sentences can still be grasped by looking at the context.

The manuscript has been revised by a native English colleague from Australia to guarantee English grammar and text clarity.

 

Line 51-56: The grammar structure appears to be rarely encountered, if allowed, in academic literature. It may be helpful to separate the sentence to shorter sentences to make the structure of each sentences clear.

This has been corrected.

 

Line 110: What does "de aridity index" mean in English?

This has been corrected.

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript is improved than the last version