Characterizing Spatiotemporal Patterns of Mangrove Forests in Can Gio Biosphere Reserve Using Sentinel-2 Imagery
, Sangay Gyeltshen 2, Chau P. T. Nguyen 1, Duy X. Tran 3, Tung H. Luu 4 and Man B. Duong 4Round 1
Reviewer 1 Report
It needs a major revision for the following points:
1) The section of Introduction is too short to be made a literature review and show why this research is significant;
2) Validation is also importatnt for NDVI derived from satellite data, but there is no information how the validation was made and what the accuracy of the data classified from NDVI is.
3) References are not sufficient and please update if possible.
In short, it needs a revision as suggested above.
Author Response
Dear Reviewer,
Please find enclosed the response to the reviewers’ comments as well as the new manuscript version for our paper (ID: applsci-815831) “Characterizing Spatiotemporal Patterns of Mangrove Forests in Can Gio Biosphere Reserve using Sentinel-2 Imagery”, to be considered for publication in Applied Sciences.
We have done our best to address the reviewers’ suggestions and concerns in the answering letter as well as in the revised manuscript. All the changes in the manuscript have been made in red to outline them.
The entire manuscript has been revised extensively, and we hope that it is now commensurate with the high standard of the Applied Sciences.
If you need any further information or clarification, please do not hesitate contacting us.
Looking forward to hearing from you.
Sincerely,
Author (s)
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper aimed to address spatiotemporal patterns of mangrove forests from Sentinel-2 imagery using the state-of-the-art Google Earth Engine technologies. As the topic itself is interesting and falls in a good realm of remote sensing applications, it appears that this study is more appropriate to be submitted to a remote sensing related journal. However, I see there are numerous mangrove-remote sensing related papers have been published in the past decade. In order to be considered for publication in a remote sensing focused journal, the methodologies used the study has to be innovative enough and the work more comprehensive. Given the fact that deforestation of mangrove trees is a global environmental concern, this study might be of interest to landscape conservation communities and stakeholders and be helpful to government agencies for their spatial decision making.
This paper aimed to address the change spatial-temporal patterns of mangrove by characterizing land use/cover change over the years. instead of using popular machine learning classification methods (e.g., Random Forest, CART, SVM, etc.), the method used in this study for identification of mangrove forests is to threshold NDVI values (>0.3) using a mangrove referencing map. NDVI thresholding only is not the most robust way as NDVI varies with different species and varies among the same species with different phenological phases or growing seasons. As shown in Figure 5, significant land use/cover transitions occurred between the forest and other landcovers. For example, if there are other forest species, other than mangrove in the study area, the change of NDVI might not be solely from Mangrove.
The environment where mangrove forests grow is complicated as it is mixed with wetlands and temporary inundation water, which could be better identified from SAR imagery, especially for mangrove standing in water. In that the authors had accessed Sentinel-2 MSI data collection from Google Earth Engine, they could acquire Sentinel-1 SAR imagery as well. It would be more interesting to integrate both SAR and multi-spectral to better identify the mangrove forests.
One of limitations of this paper is that it addressed the spatial-temporal pattern of mangrove forest simply from its decrease in area. With time-series NDVI, a Mann-Kendall trend analysis and significant test can be easily conducted with Google Earth Engine. The trend analysis using NDVI of the same seasons over the years will reflect whether and where NDVI has a significant trend of increase or decrease over time.
No strict accuracy assessment was conducted on the mangrove's distribution. Although Overlap Similarity Algorithm was introduced, it is not used as commonly as Kappa etc.
Author Response
Dear Reviewer,
Please find enclosed the response to the reviewers’ comments as well as the new manuscript version for our paper (ID: applsci-815831) “Characterizing Spatiotemporal Patterns of Mangrove Forests in Can Gio Biosphere Reserve using Sentinel-2 Imagery”, to be considered for publication in Applied Sciences.
We have done our best to address the reviewers’ suggestions and concerns in the answering letter as well as in the revised manuscript. All the changes in the manuscript have been made in red to outline them.
The entire manuscript has been revised extensively, and we hope that it is now commensurate with the high standard of the Applied Sciences.
If you need any further information or clarification, please do not hesitate contacting us.
Looking forward to hearing from you.
Sincerely,
Author (s)
Author Response File: Author Response.docx
Reviewer 3 Report
Review of the paper:
Characterizing Spatiotemporal Patterns of Mangrove Forests in Can Gio Biosphere Reserve using Sentinel- 2 Imagery
The paper deals with an interesting topic related to spatio-temporal patterns of mangrove forests within a Biosphere Reserve.
Reviewer comments:
- In the section Introduction, lines 45-55, provide more details about this fragile ecosystem-mangrove (relevance, threats).
- In the section Introduction the authors have to define a research question/ research hypothesis. The authors present only main objectives.
E.g. Could the multi-temporal NDVI derived from Sentinel through Google Earth Engine provide valuable instruments for mapping and monitoring the mangrove patterns?
- In the discussion section there should be a subchapter showing that the methods used by the authors improve this type of analyse.
Author Response
Dear Reviewer,
Please find enclosed the response to the reviewers’ comments as well as the new manuscript version for our paper (ID: applsci-815831) “Characterizing Spatiotemporal Patterns of Mangrove Forests in Can Gio Biosphere Reserve using Sentinel-2 Imagery”, to be considered for publication in Applied Sciences.
We have done our best to address the reviewers’ suggestions and concerns in the answering letter as well as in the revised manuscript. All the changes in the manuscript have been made in red to outline them.
The entire manuscript has been revised extensively, and we hope that it is now commensurate with the high standard of the Applied Sciences.
If you need any further information or clarification, please do not hesitate contacting us.
Looking forward to hearing from you.
Sincerely,
Author (s)
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
My comments are well revised and it is agreed to be accepted for publication.
