Application of Sentinel-1B Polarimetric Observations to Soil Moisture Retrieval Using Neural Networks: Case Study for Bare Siberian Chernozem Soil
Round 1
Reviewer 1 Report
Only minor revisions were requested, and all were corrected/explained adequately and, where explanations were needed, provided.
Author Response
Dear Reviewer,
Thank you very much for your positive comment. We reread the article again and corrected some typos.
Best regards,
Muzalevskiy Konstantin
Reviewer 2 Report
This is a re-submission of a previous paper. The authors have tried there best to address all the comments from the reviewer in the best way. Field measurements remain an issue to all retrieval soil moisture study due to limited points but there nothing that authors could do more.
The authors improves the understanding of the paper and remove sentences that was causing confusion. Soil moisture retrieval remained a important topic. This paper is a interesting contribution to understudy area.
Author Response
Dear Reviewer,
Thank you very much for your positive comment.
Best regards,
Muzalevskiy Konstantin
Reviewer 3 Report
The authors have responded to all of my comments. Thank you very much. However, it is necessary to clarify the following point:
- Figure 2 is interesting but difficult to interpret as a function of soil moisture or roughness. It is necessary to specify the roughness state of the plot for the 7 days indicated in the figure and add this information in the description of the in situ data. In other words, it is necessary to know if on May 21 the plot was plowed or … with high, medium or smooth roughness (same pour other dates). Without this information figure 2 is difficult to interpret. In [20], alpha and entropy depend on both parameters soil moisture and soil moisture.
Author Response
Dear Reviewer,
I have responded to your comment and made corrections to the manuscript (see attached file).
Best regards,
Muzalevskiy Konstantin
Author Response File: Author Response.pdf
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
This manuscript addresses an interesting study using polarimetric Sentinel-1B for developing an algorithm of soil moisture prediction with artificial neural networks (ANNs) model.
English writing structure and grammar are all right, and only minor corrections are requested.
Abstract – line 16 – The correct term is an artificial neural network (ANN), and it is not a technology but a computational model. Please correct it all over the manuscript.
Line 21 and 22 – It will be better to express the depth in m instead of cm. 0-0.05 m and express RMSE in % instead of cm3 cm-3. Elsewhere avoid using cm as a distance measurement or cm3 for volume.
Line 63 and elsewhere - (see Figure 1) – remove see, just (Figure 1).
Line 107 – “In the case of the model [13]”, it will be better to give more information about which model you are referring to instead of only giving the reference. I understand that you are referring to equation 1, but it is not very clear.
Line 138 – Provide more information about the software Matlab, version and developer (Mathworks, MA USA).
Line 163-164 – Why this phrase is between parentheses?
Reviewer 2 Report
General comments
Muzalevskiy K. and Zeyliger A. submitted the article "Application of Sentinel-1B polarimetric observations to soil moisture retrieval using neural networks. Case study for bare Siberian Chernozem soil" to remote sensing journal mdpi. The authors proposed two NN retrieval algorithms to extract soil moisture data. Then, the results of the predicted soil moisture is compared to in-situ measurement on Siberian Chernozem soil. It is a well structured article, very concise that goes straight to its point. The article proposed two ANNs method The Introduction is sufficient to understand its component such as the sentinel 2 data, the in situ measurement and the debate around previous result on soil moisture retrieval. It is also available because rare results are available in the region proposed by the authors. The figure supported the claim of the authors being high quality making the article relatively solid and convincing. I would recommend to publish the article with minor revision (SOme captions could be improved, labelling and few clarification needed)
Questions
Question 1
Line 128 you mentioned that "there are significant random deviations " between left and right of the equation 1. Would you mention the most probable cause of that discrepancy? i.e. Uncertainty in cross polarized ration ....etc ...
Question 2
Would you clarified why you using "40 neurons ? 40" for the layer 1 (Line 174). Based on Figure 4b, it is difficult to understand why another number was not chosen ? Is that the optimal R2 value that minimized Extend p parameter error ?
Question 3
Based on Figure 5, the range of soil moisture investigated stands from 0.05-0.3 cm 3 cm -3 and the type of soil is unique (surface being small). Would you expect to have similar results for soil with another nature (sandy, loam, silt) ? Does high soil moisture value may present a larger error ?
Abstract
Try to highlight the importance of soil moisture satellite measurement … large coverage, impact on hydrological and climate model, direct application for agriculture ...
Detail comments
Line 95
Remove the Blank, "HGT). As"
Line 108
Define ? VH ⁄ ? VV
Line 111
Define kr too.
Line 121
of Equation (1)
-->
of Equation 1
Line 185
and H, see Figure 3)
-->
and H (see Figure 3),
KLine 223
"equal to about 37°" , do you mean higher than about 37°
Line 229
M.K. finished
-->
M.K. completed ?
Line 230
M.K. conceptualization, designed the framework of this study
-->
M.K. conceptualized and designed the framework of this study
Figures
Figure 3
Slight overlap of in/out ...
Figure 4
Seems to invert a) and b) as in caption a) concerns L140 and b) ? ? 1 ,? ? 1 ? 10 2 and ? 1 ? 2 but that the reverse in the Figure. Please clarify it.
Clarify the meaning of the 2 dashed vertical lines in the caption, it seems that on Figure 6b, it is where p=1.5, what about Figure 4a) ?
Figure 5
Precise what the meaning of the green hollow in the caption.
Figure 6
Please precise the unit of the color, it seems that cm 3 cm -3 ..
You may consider having the difference (Figure 6b) in % rather than in cm 3 cm -3 as we can see the largest absolute error.
You may also precise that the "dot" represents .. ?soil moisture in-situ measurements ?
Reviewer 3 Report
See my comments in the pdf file
Comments for author File: Comments.pdf