Change Detection of Amazonian Alluvial Gold Mining Using Deep Learning and Sentinel-2 Imagery
, Kangning Cui 2, Victor Paul Pauca 1, Sarra Alqahtani 1, Miles Silman 3,4, Raymond Chan 2, Robert Jame Plemmons 1, Evan Nylen Dethier 5, Luis E. Fernandez 3,4,6 and David A. Lutz 5Round 1
Reviewer 1 Report
The comments can be found in the attached PDF.
Comments for author File: 
Comments.pdf
Author Response
We appreciate the thoughtfulness and thoroughness of the two reviews. The suggestions have been addressed and we feel they have significantly improved the manuscript. A detailed response to reviewer1 is found in pdf document.
Sincerely,
Author Response File: Author Response.pdf
Reviewer 2 Report
This is a very competent manuscript. It is well written and presented. The tables and figures are professional and informative. There is a clear statement of intent, logical flow in organization and a very extensive (72) set of appropriate references in complete and consistent format.
The topic is unusual and interesting. The science is very complex but as the authors briefly mention, the methods might be extended to other change detection situations. The authors might acknowledge that in some communities, similar change detection is a visual process. The authors’ decisions to use multiple study sites and three methods of performance evaluation were very appropriate.
It is a journal policy but the extensive number of figures and tables in the appendix is unusual. They might be reduced or some incorporated into the primary text.
As in most manuscripts, there are editorial suggestions for consideration by the authors, several of which follow:
- It is increasingly common and really an editorial decision but the extensive use of personal pronouns (we, our) is not comfortable for some readers.
- Some of the keywords are very general.
- Land cover, land-cover, land use, land-use? Perhaps land use land cover (LULC)?
- Figure 1 needs country identification and scale. There are a number of other figures that would benefit from an indication of scale or size.
- Line 71, mixed or mined?
- Lines 151 and 44, data are plural.
- Line 189, )?
- Many of the figure and table captions are very long.
- Line 259, Figure. Remove period.
- Line 345, al-though.
- Do not separate figures and captions.
- In the text Figure is both bold and not?
- Line 401, fi-ne?
- Reference 26 appears incomplete.
In summary and as stated, this is an excellent but complex manuscript. This reviewer appreciated the competence of the authors in both the science and the presentation.
Author Response
We appreciate the thoughtfulness and thoroughness of the two reviews. The suggestions have been addressed and we feel they have significantly improved the manuscript. A detailed response to reviewer2 is found in the pdf document.
Sincerely,
Seda
Author Response File: Author Response.pdf
