A Comparison of Different Remote Sensors for Ramie Leaf Area Index Estimation
Round 1
Reviewer 1 Report
L102: Which company for ASD FieldSpec 3, View Spec Pro 6.2, A DJI Phantom 4, Pix4dmapper, PhenoAI
L132: Which equation to calculate Vis?
L133: How accuracy of LAI-2200 canopy analyzer for the dense Ramie canopy?
L141: How to split training and validation sets?
L142: Ground LAI should be the response variables, instead of input parameters
L143: Please describe details for Linear, RF, SVR and PLSR.
L148: The stability of model cannot evaluate with difference of R2 between training and validation sets.
L154: Add a section in the results to describe the distribution of observed LAI
L159: How to calculate the correlation coefficients? How to average correction coefficients?
L168: What’s results for “RFE was used to screen hyperspectral indexes”?
L178: PLER?
L180: RF is better than other methods from training sets
L239: There is no point to compare the reflectance if no correction for the band width.
Author Response
Dear reviewer,
Thank you very much for your guidance on this paper. According to your suggestions, I have used the audio track change mode in MS Word to make a large modification of the paper. Please see the attachment for specific instructions.
Author Response File: 
Author Response.doc
Reviewer 2 Report
The authors report an analysis on the detection of spectral activity and its correlation with leaf area index (LAI) in ramie crops. Specifically, the authors evaluated what is the best approach to determine LAI and establish the "health" status of the plant. For the analyses, they simulated a nitrogen deficit in the soil and measured the spectral activity with a hyperspectral sensor and a multispectral sensor at different stages of pyothy growth. Subsequently, the spectral data were used for some algorithms that established the parameters of LAI. The empirical data confirmed that both sensors are able to estimate LAI. The system developed by the authors can be a valuable aid for smart agriculture and future commercial approaches.
Author Response
Dear reviewer,
I am very happy to receive your comments, which are really precious to me. What I want to say is that you are a very responsible reviewer. Your suggestions have taught me a lot of knowledge, and your comments have also encouraged me. Thank you again.
I have revised my paper according to the reviewers’ suggestions, and hope to get a good result. As for the language, I have read through the whole text and made some amendments. If you think there are any problems with the language of this article, please feel free to contact me. Or I can apply to the journal for language editing.
Reviewer 3 Report
The manuscript (agronomy-2285111) compared the potential to estimate leaf area index (LAI) using hyperspectral and multispectral cameras, finding that the multispectral camera was more suitable for monitoring LAI in the field. The accuracy of the LAI estimation model was improved by using data from multiple growth stages. Introduction its ok. However, need better clarify objectives and hypothesis in manuscript. Material and methods, its fine but better explanation of statistical analysis is necessary. The results were well described but statistical analysis between camera and other method for analyses were necessary. However, discussion its not good. Please improve these sections for discussion with other works. Check old references and when possible, include in manuscript. English grammar and spelling were necessary, for Native language.
Points:
This is a simple work but with scientific merit in testing different analysis methods. However, I suggest to the authors to write and make it clear how these results can better assist agronomic aspects.
Figure 1. Bad quality. Conversion? Please, check and improve these Workflow.
Figure 2. Why this noise? Fieldspec its good equipment and this noise not “normal”.
L185. “Universal”?
L221. Why “texture”?
L249. What is statistical in box plot? Its mandatory!
Best regards,
Author Response
Dear reviewer,
Thank you very much for your guidance on this paper. According to your suggestions, I have used the audio track change mode in MS Word to make a large modification of the paper. Please see the attachment for specific instructions.
Author Response File: Author Response.doc
Reviewer 4 Report
This manuscript is pretty easy to follow. There are several comments.
What is a size of plots?
What is a experimental design of this plot study?
In line 106: change 'heat' to 'warm-up'.
Add more information about a calibrated white reflectance panel.
Calibrated white reflectance panel must be measured before and after plot measurements.
Add more information about 5 band multispectral sensor.
'Camera' and 'sensor' are using together in text. Reviewer recommend using 'sensor'.
Fig 1 needs higher resolution. Text inside of boxes cannot be read.
When show R2 and RMSE, reviewer guess the values are average of models. Average of models may useful for comparisons. However low significant models will not be useful. Show highest R2 and RMSE values of each model instead of average values. It is so confusing because values in text and values in figures do not match.
In Fig. 4 and 5, why significance (R2) of training results and validation results are different? Add exaplnation.
Line 168: What is RFE?
Line 240-241: What are PT4 and FS3?
Fig 8: Add labels of left and right figures.
Author Response
Dear reviewer,
Thank you very much for your guidance on this paper. According to your suggestions, I have used the audio track change mode in MS Word to make a large modification of the paper. Please see the attachment for specific instructions.
Author Response File: Author Response.doc
Round 2
Reviewer 1 Report
NA
Reviewer 3 Report
I recomend accept in present form these manuscript.
