Early Onset Yellow Rust Detection Guided by Remote Sensing Indices

Round 1

Reviewer 1 Report

The author has done a good study in understanding the remote sensing in wheat rust. However, there is always some concern in remote sensing and in achieving good disease management practices.

"There is a scope to extend the work by incorporating other relevant vegetation indices 512 such as Leaf Area Index(LAI) and disease-specific indices which were not considered in 513 of this study. Not only more disease-relevant parameters "but also other parameters like soil 514 types, planting procedure, topography, and other geomorphological parameters could 515 also be included in the models to assess performance improvements".

The concern is that the author has not explained the epidemiological aspects in a real forecasting tool, which sometimes misleading using machine learning. It is always important to include the biophysical expert to add their expertise when developing such a study. Secondly, the implication of this study with relevance to farmers has not been mentioned. How does the farmer is using this study, and what percent of his disease management is taken care of through this approach?

As the manuscript is narrated well, it should not be merely a source of citation, rather it should be an impact-oriented discussion and conclusion. I hope I have not made a too candid opinion. Because, there are millions of smallholder farmer, they still have not been accessed to such machine learning forecasting systems. So, we need to see how such a study would cater on a larger scale, where disease can be the good spreading point.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer

We thank you very much for taking the time to read through our manuscript and giving us your valuable feedback. Please see the attachments consisting of our responses you your comments.

Author Response File: Author Response.pdf

Reviewer 2 Report

The study focuses on utilization of several data sources, including ground measurements, remote sensing images and phonological features, as well as weather data, for building a model for the early detection and disease signal onset when it comes to wheat yellow rust. Many useful and agricultural practice-based insights could be found in this study, resulting in a proposed solution for timely action in yellow rust management in order to prevent yield loss. Minor revisions are suggested according to the specific comments given below, as well as suggested changes given in the attached version of the manuscript.

Several sentences or phrases should be rephrased, as given in the specific comments in the attached manuscript version.

The software(s) and libraries used for data analysis, interpolation and visualization should be given in the Materials and Methods section. Methodology used for data interpolation should be explained in more details in each section where the data interpolation was performed.

Units (e.g. m) should be written with a space between the value and the unit (60 m) across the manuscript. The same applies for other units as well, except °C.

Data presented in Table 1 could include specific bands present in the images acquired from different sources to better explain the given priorities.

Abbreviations given in Table 2 should be explained in table footnotes.

“Significant differences” – when comparing parameters used as regression model metrics statistical analysis should be performed to claim there hadn’t been significant differences between the tested models for each parameter.

X-axis title is missing in the column 1 of the Figure 8, please add.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer

We thank you very much for taking the time to read through our manuscript and giving us your valuable feedback. We read through all your comments and made all the necessary changes in the manuscript. Please see the attachments consisting of our responses you your comments.

Author Response File: Author Response.pdf

Reviewer 3 Report

1. I found many mistakes in writing and format, even in the references also didn't follow the format in MDPI.

2. Please check line by line all the sentences and spacing, bracket and make it a free error in typo, eg:
Agriculture Organization(FAO) - Organization SHOULD BE SPACE, PLEASE CHECK ALL OF THEM!, IOT - Should be IoT.

3. The title, it is correct to use RS indices, or maybe can change to VI?

4. Have you explained what is EMS? In Line 66 page 3.

5. A combination of in-situ observations, weather data, 123 remote sensing and phenology data for yellow rust prediction has also not yet been 124 explored - are you sure? has any proven or supporting doc?

6. Please mentioned FL before you used it! Check others too!

7. Do you have any experimental designs?

8. Why is figure 3 only showing up to 2017, I thought your data is up to 2018?

9. In table 1, why do you need to add the start date? What does it's mean with the start date here? 

10. Section 2.4.1 any other pre-processing for the satellite imagery? 

11. Why at MAE Or RMSE no equation No? Please check the format.

12. Do you have a photo of g Leaf level yellow rust?

13. Why did you choose the window sizes of 15,31 and 42 days?

14. Why do I think missed the weather data? Where it is and how you are integrated with the RS data?

15. Please improve the writing and discussion. 

Author Response

Dear Reviewer

We thank you very much for taking the time to read through our manuscript and giving us your valuable feedback. We read through all your comments and made the necessary changes in the manuscript. Please see the attachments consisting of our responses you your comments.

Author Response File: Author Response.pdf