Satellite-Based Frost Damage Detection in Support of Winter Cover Crops Management: A Case Study on White Mustard
Round 1
Reviewer 1 Report
Dear authors and editor,
the manuscript “Satellite-based frost damage detection to support winter cover crops management: a case of study on white mustard” aims to assess the suitability of Sentinel-2-derived vegetation indices to quantify the frost-affected area and the frost window on a cover crop. Although the first of these objectives does not fill specific gaps (a lot of literature is available on the topic), the second may represent a good starting point for future development. The manuscript is poorly written, both from a technical and a style point of view. However, there are other concerns that I would like to stress to the authors:
1. I believe there is a problem that you need to solve about the experimental design. You have based your analysis on only 1-year of observation, which is not recommended in agronomical studies. The problem could be amended by normalisation of your data on a multi-annual base (e.g., refer to https://doi.org/10.2111/REM-D-11-00153.1)
2. The Introduction is quite general and does not stress the relevance of your study. On the one hand, you should provide information on the current methodology for frost damage assessment on cover crops. On the other hand, you must explain the relevance of determining the frost window.
3. The material and methods paragraph on satellite-based frost detection must be improved. How many images did you use’ On which dates? How did you consider other factors influencing the spectral response for the vegetation?
4. The statistical approach to assess the suitability of the VIs to quantify the frost damage is rather poor: in Figure 2, it seems that the regression lines are influenced by a quite irregular pattern of the ground-based GI.
5. The Results could be more synthetic in some parts but are fairly reported.
6. To address future developments, the Discussion should give a more critical analysis of the Results. For example, the explanation of the response of the different spectral areas to frost can be found in https://doi.org/10.3390/rs12111896
7. Make sure to introduce the acronyms the first time you use them and keep on using them throughout the manuscript
8. English writing and editing must be deeply revised.
Here are some specific comments:
Line 62: you should add that the spatial resolution of satellite-based VIs is much lower, though
Equation 1: GI should be defined
Line 126: additional information is needed for the QA60 band
Table 2: references are needed for the VIs
Figure 1: the colours of the lines should be changed as it is not easy to see the ground-based time window
Line 304: “EVI is in grey.”??
Line 322: if I have understood, the discontinuity was on December 22
Author Response
Please see the attachment.
Sincerely,
Martina Corti
Author Response File: 
Author Response.docx
Reviewer 2 Report
1. It is a clear and understandable title, which is suitable for research purposes
2. The abstract should include some future work at the end as a case study at the end of the abstract
3. I would like to suggest that the Sentinel-2 vegetation indices be updated with some literature, since Sentinel-2 vegetation indexes are used for detection, but there is no literature review for it.view for it.view for it.
4. As for the Materials and Method section, I would recommend starting with the subsection without further explanation. Therefore, I would recommend explaining the section before starting the next section or subsection.
5. Section 2.1.1 cites, and you should make sure it cites to what?
6. It is imperative to note that the section "Ground measurement of frost damage" will be providing some graphical calculations to help to understand the process.
7. There is no clear correlation between the delay high accuracy in figure 1 and the delayed low accuracy in figure 2.
8. There are no clear images in all the results; they need to be put in a clear format
9. There is a need for further research in the conclusion as well
10. There are some references that need to be updated
Author Response
Please see the attachment.
Sincerely,
Martina Corti
Author Response File: Author Response.docx
Reviewer 3 Report
Dear authors.
The article is of scientific interest and well written. However, the applicable methodology, particularly the number of field surveys (20) is too small to obtain statistical significance with satellite indices. In the opposite case it can be considered as a randomness.
Moreover, it seems to me more expedient to accept the given methodology for a separate field. When you apply a linear regression between all GI indices from all plots and the independent variables (satellite indices), that can also give biased results. For example, the blue circles in Figure 2 clearly stand out from the rest.
Since this is the main purpose of the paper, I cannot recommend it for publication.
Other minor comments in the attached file.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Sincerely,
Martina Corti
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Thank you to the authors for considering most of my suggestions and their explanation in their reply. I think the quality of the manuscript has considerably improved. I recommend the publication of the manuscript.
Reviewer 3 Report
I am satisfied with the authors' answers. I recommend the article for publishing.
