Quantifying Hail Damage in Crops Using Sentinel-2 Imagery
Round 1
Reviewer 1 Report
Dear Authors,
Congratulations for the work done, it is a high quality paper that fits well with the target of the journal.
Below, I present a few minor points to be considered by the Authors:
- Why are hailstorms defined as “unpredictable” events? I do agree that these phenomena might be “difficult” to observe, model and predict. However, they are not “unpredictable”.
- In the Introduction, Authors should be making reference to the effects of global warming on hailstorms and how the area of study will be impacted.
- Section 2.2.5 Validation – please add the reasons why the RMSE has been used (similarly to the explanation given to “r”). LN271: low or lowest RMSE?
- Table 4 – I would suggest to add a new column to include the “target” or purpose for using those indices (why the Authors have selected these indices for this particular study). Remember that all the components of each formula needs to be defined in the text. Revise the MSAVI2 formula (I think that “x” needs to be deleted)
- Figure 5 (wheat): why is the 84% damage line (blue) above the 4% damage (red) before the hailstorm event? This does not happen for canola or lentil
Format and presentation:
- References need to follow the Instructions for Authors. There are many examples in the text in which instead of mentioning a specific work or author, a number is used. See LN76 (“[14] calculated….” ), LN203 (“[33] measured”), LN424, LN433, LN441, LN444.
- The abbreviations should be defined first time used in the abstract and main text. Please carefully review the text to make sure this happens. For example, (CDN) (LN39), or “ha” (hectare) LN108.
- Given the considerable amount of abbreviations, it is recommended adding an “Abbreviations” section after the Conclusions to include all of them.
- Figure 1 – needs to add in each picture (a), (b), (c), (d). Figure 1d needs to be improved as it is not possible to differentiate the colours. All Figures ((a), (b), (c), (d)) needs to include the NORTH and scale
Author Response
Dear Reviewer,
Thank you very much for your constructive comments and suggestions. We accepted all your suggestions edited the manuscript carefully. Please find our work and attached file as well as in the new version of the manuscript.
Thank you very much for your contributions.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The work is very interesting and properly written.
Please check the formula number 3, an additional parenthesis is needed.
Author Response
Thank you for pointing this out. The error (a parenthesis) was revised for formula number 3.
Author Response File: Author Response.docx
Reviewer 3 Report
The study quantifies the hail damage in crops using Sentinel-2 imagery. The study found a good correlation between the time-series changes in vegetation indices and ground estimates of hail damage that may allow for a more accurate assessment of the extent and severity of hail damage to crop land. From my perspective, the article is within the theme of the journal and interesting for potential readers. Thus, it could be considered for publications if the following concerns can be solved properly:
- Section 2.2.3: The criteria to define the two dates (points: a and b) before and after the event should be explained.
- Fig. 4 needs to be improved. Sample borders should be clearer. Fig. 4A is before the hailstorm, so the damage percentage numbers should be removed. The NDVI colour scheme should have the same range for both A and B. Are the NDVI values from 0 to 1? What is the size of the sample? And also, please check the scale bar.
- Equations (3) & (4): are x and y different from Pi and Oi?
- The caption of Table 6 mentioned 14-days AUC while it was explained as 12-days AUC in the text. Please correct it.
- From the definition in Table 4, GNDVI and NDWI should yield the same evaluation results with an opposite sign, so please correct their corresponding results interpreted in the text.
- Figs. 6 & 7: Please check if the captions of the two figures are correct, e.g., the terminology “the actual (insurance-inspected) crop damage” vs “ground-estimated crop damage”, are they the same? 32-days in both figures? Fitting equations and R2 of Fig. 7?
- If I understand correctly, the site number of lentil sites is 5 (Table 1) but the scatter plots in Figs. 6 & 7 have 6 data points. Are they correct?
- Fig. 8: see the comments for Fig. 4, also label a and b as mentioned in the text.
- The practical applications of this research in the context of insurance should be discussed.
- What are the limitations of this study and the need for further research?
Author Response
Dear Reviewer,
Thank you very much for your constructive comments. We accepted all your suggestions and edited the manuscript carefully. Please find our work and attached file as well as in the new version of the manuscript.
Thank you very much for your helps.
Author Response File: Author Response.docx
Round 2
Reviewer 3 Report
The authors have addressed all comments and hence the manuscript can be considered for publication.
