Quantitative Study of the Effect of Water Content on Soil Texture Parameters and Organic Matter Using Proximal Visible—Near Infrared Spectroscopy
Round 1
Reviewer 1 Report
My most important criticism of this paper is that the description of methods does not discuss how you know the "true" measure of clay, silt, sand, and OM in each sample. In order to train the spectral data using PLS, you need some reference measurement of these quantities. I assume you have this, but it is not explained in the paper. It is necessary for you to describe these "truth" measurements so that the reader understands how your spectroscopy results can be considered accurate.
In general, the paper needs to be heavily edited for clarity. Many paragraphs have clumsy wording that takes 2-3 readings to figure out what you mean.
I don't understand what you mean by "hyperspectral" in this paper. My understanding is that hyperspectral imaging is a remote sensing technique that provides simultaneous spatial/spectral characterization of a 2D scene. I think your use of "hyperspectral" just means "spectroscopy". You should at least clarify what you mean by "hyperspectral" so that it is clear to readers with lab and remote sensing backgrounds.
I think the paper would be stronger if you had considered soils with less sand and more silt/clay. You acknowledge that sandy soils are overrepresented, but you don't really justify why it is appropriate to bias your results in this way. This might also help strengthen you conclusions regarding low sand samples in your set.
In the introduction, your use of the word "model" is confusing. Do you mean statistical modes to describe your spectral results, or do you mean models developed by soil scientists that are built on accurate parameters such as those you are studying in this paper. Please be more clear about this.
Section 2.1.2 needs a better start. You should start the section by saying that your objective is to acquire spectra of your soil samples at varying WC levels. Then discuss how you do this. It was unclear on first reading where the spectroscopy fit into the method. Just a simple, more effective intro sentence would help.
Section 2.2.6. I suggest using a different word than "discard", which suggests you are eliminating the set permanently, rather than just setting it aside to be a validation set.
Section 2.3. Item 5 in your procedure (lines 299 and 320) is unclear. What does it mean to repeat for "1 K times"? Previous reference to k has been lower case. Also, it seems like you are defining k=10 in this procedure, so it might be clearer to just say "10 times".
Section 3.2.2. Do you have an explanation why the bands at the end of the visible range were most correlated with WC? it would help strengthen your case if you did.
The discussion of your results comparing R2, NASH, etc. would benefit from a summary table, where perhaps you can highlight the largest deviations between models considering WC and models that don't.
In your conclusion section, you mention estimates having "decent" accuracy (e.g., line 592). What does "decent" mean? Evaluation of how good accuracy results are often depends on what the user needs. You need to define what good, or decent, means.
Author Response
To reviewer
Author Response File: 
Author Response.docx
Reviewer 2 Report
1. Figure 4 is completely uninformative. In Figure 4, it is better to present a functional scheme, including the optical part of the illumination of the surface of the samples and the collection of reflected radiation.
2. Line 172: Explain the phrase "to reduce the signal-to-noise ratio".
3. How was the spectral reflection data processed in the overlap range of two spectrometers (900-1038nm)?
4. In the graphs (Figure 6) along the abscissa axis, the wavelength interval should be a multiple of 2, 5, 10 or 25. In Figure 6A, it is possible to limit the ordinate axis to 60 or 65%, which will increase the informative area of the figure. The same remarks can be attributed to Figure 7 and others.
5. What is the point of showing in Figure 6A absolutely all the reflection curves of the samples?
6. In lines 583-592 of the Conclusions section, instead of conclusions, there is a repetition of the purpose and description of the research.
Author Response
To reviewer
Author Response File: Author Response.docx
Reviewer 3 Report
Quantitative Study of The Effect of Water Content on Soil Texture parameter and Organic Matter Using Proximal Visible-Infrared Spectroscopy
1. Very interesting research entitled “Quantitative Study of The Effect of Water Content on Soil Texture parameter and Organic Matter Using Proximal Visible-Infrared Spectroscopy”.
2. Correct the structure of the article (it's just a suggestion). See attached file.
3. Indicate the type of document on line 1 (see example). See attached file.
4. On line 13:
Says: [email protected] (A. E.-A.); Tel.: +1 418 654 3819.
Should say: * Correspondence: [email protected] (A. E. A.); Tel.: +1 418 654 3819.
5. On page 3, there are 2 footnote references to “vegetation1” – line 104. Delete, they are not used in this journal.
6. On lines 139-140 it says dice “Although Lekshmi [18] suggested that the dry state is reached for soil heated to 105°C for 24 hours,”. Is the soil not damaged by such a high temperature for 24 hours?
7. In lines 168-170 it is about “6) A software for acquisition, reading, and storage of spectra that allows to adjust the speed, the number of acquisitions, and other parameters allowing the optimization of the quality of the signal and to reduce the noise.”. Indicate the name of the software and show the results obtained.
8. Reduce the size of figure 4. It is shown very large and crude.
9. In line 180 the algorithm is treated “The K-means is a non-hierarchical data partitioning algorithm based on moving centers”. I suggest including the algorithm and referencing where it was taken from. I suggest that the algorithms included in this article use the following format: (See attached file).
10. The steps indicated on lines 288-299, can be written as an algorithm.
11. In line 349 it says “[5% - 10%[, [10% - 15%[, [15% - 20%[, [20% - 25%[,” and must say “[5% - 10%], [10% - 15%], [15% - 20%], [20% - 25%],”.
12. Define some future works on this research.
13. Very good bibliography. I hope you can consult more bibliography.
The article has good content and very interesting.
Authors are requested to make all indicated corrections.
Comments for author File: 
Comments.pdf
Author Response
To reviewer
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Overall, the latest edit is an improvement over the first draft I reviewed.
I still would prefer a little more detail on how the soil texture parameters were analyzed, but at least now there's a reference to the methods used.
I don't like the addition of the Figures 6 and 7 which describe the algorithms used. I felt the previous description using a numbered list in the text was sufficient. My preference is to remove these and keep the numbered lists.
Also, the added discussion of why the bands at the end of the visible range were highly correlated with WC is confusing, with perhaps some important words missing. Please review and rewrite.
The added summary table in the conclusion is very helpful and makes the differences between considering and not considering WC easier to see.
Finally, I still do not like the use of the phrase "decent accuracy" in the conclusion section. I recommend either defining what "decent" means in the context of these results, or using more precise language. A reader might see this and assume the authors are using intentionally vague language, or at the very least some readers might interpret "decent" as good and some might interpret it as "average". I disagree with the authors that the term "decent" is implicitly explained by the results that follow. I suggest using the word "satisfactory" or "sufficient", if that's what you mean.
Author Response
To reviewer
Author Response File: Author Response.docx
Reviewer 3 Report
I acknowledge that they made all indicated corrections.
Very good article.
Best regards.
Author Response
To reviewer
Author Response File: Author Response.docx
