Exploring the Potential of PRISMA Satellite Hyperspectral Image for Estimating Soil Organic Carbon in Marvdasht Region, Southern Iran
, Ruhollah Taghizadeh Mehrjardi 2 and Farhad Samadzadegan 1
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsSee the attached document
Comments for author File: 
Comments.pdf
Author Response
Dear Reviewer,
Please see the attached answer file.
Regards,
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsGeneral Comment
Satellite hyperspectral imaging was discussed to predict the spatial variability of SOC over a semiarid farm area in Iran, and some potential predictability was proposed. The content of materials have value to be interested. Though there are many insufficient explanations and mistakes found, such as the text has too much local abbreviations and not easy to read through. Distinction of "Methods" and "Results" were fuzzy. Critical discussion of the results was redundant and out of focus. A list of abbreviations and their explanations are required at the top of the text, at least. The material should be rejected or major revision.
Specific Comments
1. Title: A proper noun "PRISMA Satellite" implies to focus on a specific tool of method, or it is a common terminology in the RS people. The reviewer feels it strange.
2. Abstract: Too many first-look abbreviations are found. The readers could not follow it. For example, SG-FOD, TV, VisuSrink, HFC, PCA, ICA, LGBM, CBRT, RF, M#1, S#1 and so on. They need some explanations at least, and unnecessary abbreviations should be restricted to use.
3. Introduction has also many abbreviations and redundant, not easy to read.
4. Line 137 and 138 on page 3: The position of reference number is not correct. Usually it puts at the end of text. [34] is not used for the subjective. Revise them.
5. Figure 2. No captions on vertical and lateral axis, and the scales are different.
6. Lines 393 to 423 on page 12. Section 3.1 should be moved in chapter "Material and Methods". This is showing the experimental site and conditions.
7. Lines 414 to 444 on page 13 and 14. Section 3.2 and 3.3 should be moved into chapter "Material and Methods". This is also showing the experimental conditions on the spectra used.
8. Lines 447 to 467 on page 14 and 15. Section 3.4.1, The difference of prediction to the reference data should be a main target on discussion. Difference among scenarios are somehow out of the scope. There are some measures for "Accuracy", not only R^2. When discussing on R^2, scores more than 0.6 will be the issue of interest. The other sections of results should be reconstructed in the same way
Author Response
Dear Reviewer,
Please see the attached answer file.
Regards,
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThis study took SOC as the main research object and proposed prediction models with different processing methods and machine learning algorithms on PRISMA image. The results showed in the manuscript indicated that the highest accuracy when features obtained from PCA transformation were added to the results from the TV algorithm, yielding an R2 of 81.74%. However, I recommend a supplement of the innovation and necessity of the study at the end of the introduction section. Additionally, I suggest rearranging the order of each section. The current structure of the manuscript may leave readers feeling confused. Another important issue is that, the distribution of sample points is not uniform and cannot represent the soil in the entire study area.
Several detailed recommendations were as follows:
1. The abstract is not standardized, please rewrite it. The important conclusions of the article are not covered in the abstract, and there are too many methods.
2. The abbreviation that first appears in the abstract should indicate its full name.
3. In lines 137 and 138, the citation method of references needs to be modified.
4. The author mentioned that few people use PRISMA HS data for DSM, so what is the reason why the author chose this data? What are its advantages compared to other data?
5. What is the input data when selecting sampling positions using conditional Latin hypercube sampling method?
6. In method part, how did the author solve the problem of mixed pixels?
7. I suggest placing sections 3.1 and 3.2 in the Materials and Methods part.
8. The spatial distribution of organic carbon varies greatly based on different scenarios and ML algorithms. Please further discuss the reasons.
Author Response
Dear Reviewer,
Please see the attached answer file.
Regards,
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsComments for author File: Comments.pdf
Author Response
Dear Reviewer,
Please see the attached answer file.
Regards,
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe material revised was in a good shape understandable. It has the worth to be published.
Author Response
Dear Reviewer,
Please see the attached answer file.
Regards,
Author Response File: Author Response.pdf
