Monitoring the Impact of Heat Damage on Summer Maize on the Huanghuaihai Plain, China
Round 1
Reviewer 1 Report
I have accepted the paper with minor suggestion. Some comments attached in the PDF file.
Comments for author File: 
Comments.pdf
Author Response
Thank you for your review of this article. We have enhanced the quality of Figure 1 according to your suggestions and added a north arrow.
Reviewer 2 Report
This paper proposes and evaluates a new remote sensing index LSHDI (land surface heat damage index) to reflect the high temperature and heat damage of land surface vegetation. This index performs well in the research area given in the article and its added value is detailed.
1. The topic is well introduced and the main objectives are clearly stated, and the theory that support the study is described with full detail.
2. Please provide details of MODIS dataset and ground measurement data used as basic materials of this study in Abstract.
3. The quality of the figures of this manuscript can be improved, such as increasing the size of figures and letters.
4. Please unify the font in the figures, for example, the font "times series" in Figure 6 is clearly inconsistent with the font below, please check the full text.
5. Figures looks a bit much, it is recommended to reduce it by a few.
This paper is well written, and English language and style can be improved, and maybe considered for publication after making some modifications.
Author Response
Thank you for your review and approval of this article. Here are some responses of your suggestions:
- Thank you for your positive review.
- We have supplemented some of the MODIS surface reflection data, MODIS land surface temperature products, ground observation data and statistical data used in this article in the abstract section.
- We have made quality improvements to the entire figures, including but not limited to increasing the size of figures and letters.
- We have checked the figures throughout the full text and have unified the font.
- We have simplified the unnecessary descriptions in some sections.
Reviewer 3 Report
Dear authors,
Your manuscript suggests an interesting method to detect stress in vegetation, yet your application (corn fields) and the validation (utilizing soil sensor dataset) are not appropriate:
It does not make sense to present any relationships between the MODIS pixels and the corn fields, as these fields are below 250 widths and usually their width will be around 100-150m. Thus, the MODIS reflectance data are really from mixed land covers and cannot be attributed to a specific crop. To make it worst, the MODIS thermal bands are 1000m, so their relationships to a specific field are not accurate.
It is even more meaningless to present any relationships between the MODIS pixels and the soil moisture sites, as the last indicates spatial changes in the area of 10-50cm around the sensor. Further, although you wrote in line 535 “Soil moisture can reflect the damage to vegetation affected by temperature and water”, the linkage between soil moisture and crop water stress is not straightforward, and most of the time there is a lag between the soil moisture and the crop moisture. In other words, if the crop is not watering (either irrigation or rain), the soil moisture will drop after the first day, while the crop will start to present stress after 3 or 5 days (for example Celedón, José M., et al. "Sensitivity and variability of two plant water stress indicators: exploring criteria for choosing a plant monitoring method for avocado irrigation management." Chilean Journal of agricultural research 72.3 (2012): 379).
To validate MODIS products, you can use flux towers data and for corn fields, you must use Landsat, ASTER, or ECOSTRESS data, with a pixel size of less than 100m.
In addition, the descriptions of your Methods need to answer the following questions:
What is the pixel size of your product? Are the thermal bands split into 500m or the 500m optic pixels were scaled up to 1000m?
How did you use the land cover map?
How do you know the land cover was not changed between 2010 and 2018?
How did you know where the cornfields were? In lines 230-231 you wrote that planting areas are from the Bureau of Statistics, yet it is not clear if this dataset shows each field or just statistics in the level of administration area.
Please also correct several minor issues:
Line 184 – you wrote: “The data used in this study include Landsat” yet I didn’t find any other mentions for these images.
Line 257, Figure 2, upper-right box – the word Reflectance is not written correctly.
Lines 296-297 – you wrote that minimum and maximum temperatures were 0.05% and 95.5%, respectively. Are these numbers correct? Make more sense that you will use 0.05% and 99.05% or 5% and 95%
Lines 395-396 please provide 2-3 references for your statement: “The Mann–Kendall method is a commonly used monitoring method in the field of remote sensing.”
Line 459 – Is the modeled data in Figure 7 equal to the validation data? If so, please indicate it in the text. Else please explain what this dataset is.
Lines 480-481, equations 17-18 – these NDVI and NDWI equations should be in the Method, not in Results, and specifically in the Data preprocessing Section.
Line 489, Figure 10 - what are the red and blue dots (in the NDVI-LST & NDVI-LSWI plots) represent?
Line 505, Figure 12 – how the soil moisture values, with units of percentage, are above 100%?
Line 534, Figure 15 – the legend in both graphs is not accurate.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 4 Report
Dear Authors,
Please find my commends and recommendations in the attached file!
Kind regards!
Comments for author File: Comments.pdf
Minor editing of English language required!
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 3 Report
Dear authors,
There are two main issues that are still not resolved in your current version.
The first issue is easy: although you answered my questions in your Respond file, I did not see them in the corrected version. It includes the answers to the following:
What is the pixel size of your product?
How did you use the land cover map?
How do you know the land cover was not changed between 2010 and 2018?
The second issue is more complicated, as you can not validate the 500m pixel dataset with crop fields or even with soil moisture sensors. Your product can be used to assess heat damage on a large scale as your study area, but not on specific plots. Thus, your title should be changed, for example, to “Monitoring the Impact of Heat Damage on the Huanghuaihai Plain, China”. Further, you need to explain that you do not have a good dataset for validation, as you mentioned in the Respond file, and clarify that the datasets you used (soil sensors and crop field) are not validation but a reference for the stability of this method. As I wrote before, validation of MODIS should be performed with Flux-tower data, or if you want to use soil sensors, you should use Landsat and ASTER.
no comments
Author Response
Dear reviewer，
Thanks for your suggestions for this paper! We have follow your advice to explained and supplemented the pixel size of LSHDI，the usage of land cover map and explained the little change for land cover between 2010 and 2018 in study area in the text.
In addition, we have corrected the relevant description in the paper, stating that soil moisture is no longer used as validation data, but rather as a reference to explained the stability of LSHDI. And explained there is no reliable disaster verification dataset that has been published in the world at present. Meanwhile, in the Discussion section, it was also explained in detail that using data with higher spatial resolution would be better.
