Investigating the Potential of Crop Discrimination in Early Growing Stage of Change Analysis in Remote Sensing Crop Profiles

Round 1

Reviewer 1 Report

Point 1: In the chapter 2.1. please change “Their phenological period is shown in Table 1. “ to Their phenological periods are shown in Table 1.

Point 2: You wrote “the annual accumulated temperature is 2513.3℃, and the average precipitation is 402.9 mm”. The question is for what year or years? Please put the information into the text!

Point 3: I suggest to change the title of Table 1 to “Phenological periods of main crops in the studied region”.

Point 4: Table 1 correct “heanding” to heading.

Point 5: The short form for Normalized Difference Red Edge index is NDRE. Please change NRED1, NRED2, NRED3, to NDRE1, NDRE2, NDRE3 everywhere in the manuscript (also in the figures).

Point 6: Figure 4 and 5, please add title for Y axis. The X-axis is not aligned to the number of days, which distorts the curves.

Point 7: I suggest indicating the referred year in the title of all the figures and tables.

Point 8: The manuscript contains some language errors and typos, please correct them.

Dear Authors, I have marked some necessary corrections in the manuscript.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors illustrate early-stage crop discrimination methods based on feature bands selection and machine learning. The manuscript should be revised before publication in Remote Sensing.

In page 2, "...crops have unique growth and development rules. Over time, growth rates and spectral characteristics may differ. " references should be provided here.

In section 2.2, there are "220m" and "402.9 mm". The use of space before the unit should be the same.

Table 1 should be reformatted (oc- to- ber is inappropriate).

For section 2.2.2 "between two years", does it mean the first two year? It should be clearer here. 

The unit should be set to %, and the maximum accuracy for y-axis should be 100% in Figure 4.

Table 6, 7, 8 and 9 have formatting errors.

In addition, the manuscript could be improved by dropping the endless series of acronyms in the discussion and conclusion parts. Authors could take a bit more time and space to spell out methods.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Introduction

Page 2, paragraph 1, lines 2-4. The reason for the lack of information in early crop stages is not clearly explained. Expanding the description of the phenology detection problem in the crop's early stages is suggested.

 Although several examples of similar studies that support the work are presented, the objectives and contribution of the article are not clearly stated since they are limited to mentioning that the potential to discriminate crops based on their spectral characteristics is explored. The usefulness and practical contribution of such a study should be emphasized.

2.1 Study Sites

A more detailed description of the characteristics of the crops in the study area is needed. It is essential because one of the objectives is to study the interaction of crop type and spectral characteristics. Having crop statistics for the area where, among others, the following are described: area, type of crop, type of phenology, duration of the cycle, varieties, sowing, and harvesting dates, would improve the work.

Table 1. The source of the data should be cited.

2.2 Data Sources. Satellite images.

It is not described or justified why only six images were used to analyze a variety of crops. However, the large pool of images and the potential of the GEE platform to process large amounts of data and generate more reliable results could be exploited.

Table 2.

Some characteristics of the images (corrections applied, spectral characteristics of the bands, and width-height of each image) are missing.

3.1 feature construction.

The calculation process of the spectral indices is not correctly described, several bands have a different spatial resolution, but it does not describe how they were uniformized, if there was a normalization process, or if the scale of each index was used.

Results

The description of results could be improved by adding information about the partial results of the image processing described in the methodology. For example, it could start with the descriptive statistics of each crop's spectral characteristics and spectral signatures for both spectral bands and indices, allowing us to see the differences between crops quickly.

Table5.

The information shown in the table is not clearly explained. Therefore, describing or restating the information in more detail is suggested. For example, what is meant by the column headings, letters, numbers, or numbers of images reorganized is not understood.

Discussion

The discussion of the results is weak and scarce. It is presented as a summary of the results obtained without comparing or analyzing other similar works. For example, it would be interesting to discuss the reasons for obtaining better results compared with similar studies such as [24]. In addition, the relationship between the phenology of each crop and its spectral characteristics could be studied in depth to have more information on the study's findings and their relationship with other studies.

Conclusions

It is suggested not to start the conclusions as a summary of the results found and to give an idea of how the results found are helpful in the practical solution of the problem treated.

References

Some bibliographic references are wrongly cited and have errors. For example, citation [25] has no journal name, and citation [26] has no volume or number. It raises the curiosity that there may be more similar errors.

It is suggested to use references of papers from recognized journals or relevant Ph.D. theses. The use of master's theses is not recommended, e.g., reference [27].

Author Response

Please see the attachment.

Author Response File: Author Response.pdf