Mapping Growing Stem Volume Using Dual-Polarization GaoFen-3 SAR Images in Evergreen Coniferous Forests
Round 1
Reviewer 1 Report
This is an interesting article which outlines the potential of dual pol SAR to map GSV. However, a few points remained unclear
Major comments:
1. only 25 samples with ground truth were used. This is not enough to apply ML algorithms. Please provide a reason, why at all, you consider the ML techniques for this study and not just model based.
2. It is not clear how you separated the samples which were used for feature selection and learning from those samples which were used to validate the results
3. The english style must be significantly improved and checked for grammar mistakes
Figures 2.C and 6: Improve the resolution of plots
Table 1: feature Shv or Svv? No VV-pol is used
Feuture 7 and 8: only one of them should be used.
Some (not all) english style remarks:
1. line 20 and several other: Please avoid to begin a sentence with "And ..."
2. line 23: "saturation phenomenon also obviously occured"
3. line 78-79: rewrite the sentence
4. in many lines: check if you intended to use the word “different” instead of “difference”
5. The conclusion has to rewritten
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
Dear authors,
thank you for the very interesting and relevant research.
The study of the potential of dual polarization GaoFen-3 SAR
images for mapping Evergreen Coniferous Forests is so important for the development of forest monitoring.
You can find some comments and remarks.
Lines 1-2 - The title of the article is too long. Authors can think about improving the title. For example, "Evaluating the potential of dual polarization Gaofen-3 SAR images for mapping Evergreen Coniferous Forest" or "Using of dual polarization Gaofen-3 SAR images for mapping Growing Stem Volume in Evergreen Coniferous Forest".
L. 24 - m3/ha
L. 30-31 - Authors should improve the Keywords. Please, change "GF-3" and "Forest growing stem volume" to another keyword because the title contains those words. Also, GF-3 is an acronym in the article, the keyword is GaoFen-3.
Planted Chinese Pine or planted Chinese pine?
L. 84, 94, 217 - You can describe the acronym for "Random Forest" only once.
L. 109-252 - Authors have to add a description of data of ground samples.
L. 116 - Please improve the Celsium degree.
L. 118 - m3. What does mean "annual growth of GSV"? Does it mean "current/average increment"?
L. 125 - Why did you study only 42 ground samples?
L. 126 - Were samples with a size of 25 m × 25 m for young and overmature stands?
L. 143 - Two dots.
L. 153 - Figure 2C (or c?) has too small characters.
L. 177-178 - Figure 3. - GSV of Samples don't connect with the next stage of the framework.
L. 302 - Table 4 - RMSE and rRMSE are too large.
L. 306 - add space.
L. 319 - figure 6 has too small characters, marks and symbols.
L. 365-368 - figure 6 has too small characters, marks and symbols.
L. 379 - m3/hm2
L. 405-408 - figure 9 has too small characters, marks and symbols.
L. 445-448 - It seems, the values of rRMSE from all base models in Figure 10 should not be connected with the line.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
The manuscript considers the possibility of studying the growing stem volume using Synthetic Aperture Radar (SAR) imaging using GF-3 data as an example. The manuscript is based on a sufficient amount of experimental data. At the same time, a number of corrections need to be made.
1) The purpose of the manuscript is evaluating the capability GF-3 SAR images for mapping forest GSV. At the same time, there is only one figure in the manuscript (Figure 8), which presents the results of GSV mapping. It is difficult to understand from Figure 8 what is the accuracy of GSV mapping using the proposed approach. Maps should be added to the manuscript showing the deviations of the calculated GSV values from the actual values. Or add maps with actual and calculated GSV values.
2) In the manuscript, it is necessary to unify the abbreviation for hectare. In some places it is designated as "ha" (for example, in line 24 or in figure 6. In other parts of the manuscript it is designated as "hm2" (for example, in line 317 and others).
3) Line 95: It is necessary to correct the abbreviation "SVR" to "SVM" (support vector machine).
4) Line 255. It is necessary to check whether there is an error in the number of analyzed features (seventy-four alternative features).
5) Line 274. In the explanation to Figure 4, in order to improve its readability, it is necessary to provide a decoding of each feature indicated on the vertical axis of the graph.
6) Line 317. Perhaps the correct option was: "than 300 m3"?
7) Line 420. It is necessary to correct the abbreviation "SVR" to "SVM".
8) In the manuscript, it is necessary to change the structure of sentences that begin with the words "To reduce ..." or "To estimate ...", etc. Examples of such sentences: Line 154, Line 182, Line 246, Line 311, Line 325, Line 355, Line. 409.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
thanks for submitting the revision. I am still advising some (possibly professional) English editing
line 79, "severely depended on" should be "severely dependent on."
line 98: typo "is"
line 125: "This ground measured work was conducted" should be changed to "The ground measurements were conducted"
line 134: "the GSV of all measured samples are ranged" should be changed to "the GSV of all measured samples ranges"
line 474: change "did not consider" to "were not considered"
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
Dear authors,
thank you for improving the article.
Author Response
Thanks for your work on this manuscript !