Identification of a Threshold Minimum Area for Reflectance Retrieval from Thermokarst Lakes and Ponds Using Full-Pixel Data from Sentinel-2

Round 1

Reviewer 1 Report

The present manuscript introduces an interesting comparison between satellite data and UAS-data. The comparison between the two datasets is extremely challenging and the manuscript highlights the difficulties encountered by the authors to exploit high resolution imagery from UAS to improve potential of Sentinel-2 data. I personally see a lot of potential in the manuscript that somehow follow along the line of the challenges traced in our recent manuscript on Remote Sensing.

Manfreda, S., M.F. McCabe, P. Miller, R. Lucas, V. Pajuelo Madrigal, G. Mallinis, E. Ben Dor, D. Helman, L. Estes, G. Ciraolo, J. Müllerová, F. Tauro, M.I. De Lima, J.L. De Lima, F. Frances, K. Caylor, M. Kohv, A. Maltese, M. Perks, G. Ruiz-Pérez, Z. Su, G. Vico, B. Toth, On the Use of Unmanned Aerial Systems for Environmental Monitoring. Remote Sensing, 2018.

 

In the present form, the manuscript provides a comparison of the two dataset, but it is lacking in identifying the proper workflow in the use Sentinel-2 data. I would strongly recommend to provide a final section providing the results of land classification based on Sentinel-2 where the experience gained in the manuscript is put into practice.The authors may try to use an unsupervised classification and the supervised classification, where the knowledge gained is adopted to improve sentinell-2 classification. What is the impact of this study for lake classification and definition. 

 

I have also some additional comments that should also be taken into consideration:

 

·     The manuscript is not fluent and it requires a significant rewriting effort. Methodology and results are not well described. I suggest to use a professional English editing before publication.

·     Graphs and figures are introduced without giving a full description in the text. See for instance figure 5, 6, 7  and 10.

·     Figure 5 is not described in the text. It not possible to understand how it was derived. Is it a transect? If yes provide a map identifying the relative position.

·     The final maps based on WorldView are not relevant for the scope of the manuscript.

 

 

Minor comments

Figure 1: check size of labels in the figure.

Line 150: remove the full stop 

Line 365: a parenthesis is missing

Table 2: the correlation values can be reported using two decimals instead of one. This may help identifying the relative differences.

Author Response

Dear reviewer 1

Thank you very much for your comments and suggestions, which have contributed to clarify the manuscript. We have done several changes following both reviewer suggestions, and also conducted a thorough revision by a native English speaker. We hope that our replies and modifications included in the text will put the manuscript in good conditions to be accepted for publication. Attached you will find a Word file providing a point-by-point response to your comments.

Sincerely,

Pedro Freitas and co-authors

Author Response File: Author Response.pdf

Reviewer 2 Report

General comments: the manuscript by Pedro Freitas et al. presents a new methodology for remotely determining the threshold of the minimum area of thermokarst lakes in the permafrost zone from Sentinel-2 images and examples of its application for lakes in northern Canada. The relevance and importance of these studies are due to the need to monitor small thermokarst lakes and ponds in hard-to-reach permafrost areas. In the reviewed work, the analysis of the effect on the error of the remote determination of the boundaries of lakes caused by vegetation and shading of water by tall trees and shrubs was carried out. The analysis is based on the combined use of Sentinel-2 images and ultra-high resolution data obtained from an unmanned aerial vehicle. With the use of correlation analysis, the threshold minimum area of lakes in the studied territory was established for using Sentinel-2 images. It is shown that the Sentinel-2 images can be used for remote monitoring of lakes and ponds with sizes from 350 m2. In general, I believe that the manuscript is well written enough and, in particular, very detailed in the context of the literature. I believe that basically the methodology and interpretation of results are sound, and have only minor comments for consideration in revision  of the manuscript. Specific details are given below.

Comments:

Lines 215-219: It is not clear how WorldView-4 scenes with 750px resolution were used to ensure a good overlap between the UAS orthomosaic and the satellite imagery. It is proposed to the authors to include in the text the necessary explanations.

Lines 228, 242 and 254: It is not clear why sections 3.1.1, 3.1.2 and 3.1.3 respectively appear in the text after section 3.4.

Line 236: Not specified how 3 values were determined: scale: 5; shape: 0.0004; compactness: 0.5.

Line 247: It is not clear why the text mentions "the  salt and pepper effect". Authors are advised to include in the text the necessary explanations.

Line 258: Authors should indicate what the numbers “19” and “15” mean in the text so that the reader has no questions.

Line 263: Here you should indicate the total number of lakes and ponds surveyed, from which the 12 ponds mentioned.

Line 273: The authors' statement that the differences in the characteristics of the Sequoia and S2 sensors "will not affect the overall comparison" is not justified. Authors are invited to state considerations on this subject.

Line 300: It is not explained why the analysis uses 20% of the lake area, and not 30 or 10%. This may raise doubts about the results.

Line 360: It should clarify exactly which two methods are mentioned by the authors.

Lines 368-369: Indicated that "Bold shows values significant at p <0.05.". However, in Table 2, such values were not found.

Table 2. The table includes data for the Super red edge and Super NIR channels, which were not previously mentioned either in Table 1 or in the text. It is proposed to the authors to make appropriate additions and corrections to the article.

Figure 13. Authors should provide a justification for the choice of location and size of the study site.

 

Author Response

Dear reviewer 2

Thank you very much for your comments and suggestions, which have contributed to clarify the manuscript. We have done several changes following both reviewer suggestions, and also conducted a thorough revision by a native English speaker. We hope that our replies and modifications included in the text will put the manuscript in good conditions to be accepted for publication. Attached you will find a Word file providing a point-by-point response to your comments.

Sincerely,

Pedro Freitas and co-authors

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I would like to thank the authors for their effort in reviewing the manuscript. All my comments have been properly addressed and I think that the manuscript can be published in the present form.