Mapping Large-Scale Mangroves along the Maritime Silk Road from 1990 to 2015 Using a Novel Deep Learning Model and Landsat Data
Round 1
Reviewer 1 Report
Dear authors,
The paper intituled "Large-scale Mangroves Along the Maritime Silk Road From 1990 to 2015 Using a Novel Deep Learning Model and Landsat Data" is an interesting paper.
However, there are many issues that should be solved before publication. For instance, it is lacking a framework showing the full track of your study. It will make your objectives and methods more clear. Also, the legend of the figures is very poor detailed and needs more information.
I made several comments along with your manuscript that is available in the attached pdf file. Please, consider my critical comments to improve your work.
Best regards
Comments for author File: 
Comments.pdf
Author Response
The authors would like to express thanks to the anonymous reviewers for their voluntary work and the constructive comments to improve this manuscript. Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the reviewers’ comments are marked in red, the revised manuscript are highlighted in blue ).
Author Response File: Author Response.pdf
Reviewer 2 Report
The study has tested performance of Capsules-Unet in mapping mangroves at MSR-scale. The mangrove mapping effort at large spatial scale with a consistent and reproduceable remote sensing method is always welcomed. However, the manuscript has enormous scope for further improvements. Some general comments are:
1. References and information provided in the introduction section, in many places, were not correct. Please see sticky noted in the attached file.
2. Reorganize method section. Provide more details on how you implemented Capsules-Unet model in this study.
Comments for author File: Comments.pdf
Author Response
The authors would like to express thanks to the anonymous reviewers for their voluntary work and the constructive comments to improve this manuscript. Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the reviewers’ comments are marked in red, the revised manuscript are highlighted in blue ).
Author Response File: Author Response.pdf
Reviewer 3 Report
The article provides an interesting analysis of depp learning techniques for mangrove classification and its application for land use change detection.
It covers a large area of study, and results demonstrate potential of the tested technologies.
In general, the article is presented with clarity, althoough I would reccomend an English review of the manuscript.
Some suggestions are also provided below.
Specific comments.
Abstract
I would suggest to shorten the opening paragraph stressing the importance of mangroves (lines 12-16), and particularly, the particular project related to this study (line 15-16), which might be removed or simplified, and focus instead in the abstract more on briefly stressing the research needs and novelty of the study.
The potential to replicate the methodology elsewhere could also be stressed in the asbtract.
line 58. availability "of"
Line 60. Perhaps it would be convenient to add some references at the end of the first sentence.
Also, I would add "For example" at the begginning of the second sentence in line 60.
Line 66. Please add "," after 2019.
Line 70. Perhaps "perform" or similar is better than "complete".
Line 72. Currently "a" growing number...
In the introduction, I would suggest to mention some specific background in deep learning techniques for mangrove mapping, and stress how the current approach differs from those previous studies.
Please rewrite and review English in lines 14-16.
Line 148. Please specify the method for atmospheric correction.
Line 151. Please revise the Landsat wavelengths, particularly for NIR and green.
Line 159. Please replace "is" with "were".
Line 166. Why were most training samples obtained from 2015? This might limit classification accuraccy for other years.
Figure 2. Please remove "the".
Line 185-188. Please mention the spatial precission of the GPS utilized for field surveys.
Line 188. PLease refer to figure 1 for the field validation. It would probably help to add "(1) and (2)" to identify upper and lower figures in figure 1.
Line 221. indicator.
Line 223. I would suggest to split this lengthy sentence using "." after precission and recall. Then add a sentence starting "Precission is defined as"
Lines 221-228. Please define the acronym FN (false negatives) and refer to eq. 1 and 2 in the text.
Tables 1 and 3. Please define the acronyms UA and PA.
Line 274. Hainan "Island".
For the discussion, it would be interesting to compare the observed rates of mangrove area loss against other multitemporal previous anaysis from the literature in the area of study.
Future research needs could be mentioned in the discussion.
For the conclusion, perhaps it is better to remove the sentence about climate change, since it was not directly adressed by the study.
Stressing the advantadges of the deep learning approach demonstrated here and needs for future research could be mentioned in the conclusion.
Author Response
The authors would like to express thanks to the anonymous reviewers for their voluntary work and the constructive comments to improve this manuscript. Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (the reviewers’ comments are marked in red, the revised manuscript are highlighted in blue ).
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Dear Authors,
I am glad about the improvement in your work. I believe that all my comments were answered.
Best regards
Reviewer 2 Report
The revised version looks OK. I suggest to publish the manuscript.
