Estimation of Chlorophyll Content in Wheat Based on Optimal Spectral Index

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The present manuscript estimating Chlorophyll content with different light spectra of 400-1000 nm and its polynomial regression, random forest, decision tree and artificial neural network algorithm, and the optimal vegetation index combinations.

As these techniques are being used for a long time, authors have compared spectra based estimation techniques for identifying the best stage and method for accurate and reliable spectra based estimations.

The manuscript is informative for characterizing wheat chlorophyll content at different growth stages.

References are ok but missing citations of some of the pioneer research papers.

The figures, tables and plots are ok.

Do the authors have validated/compared these spectra based estimations with conventional spectro photometric methods?  

Few of Examples are:-

The meaning of the sentence is generalized kindly Re-write the sentence "Red light and near-infrared are sensitive ...... carried out a lot of research" factually.

 

Comments on the Quality of English Language

Manuscript needs extensive reading by authors for improvement of language.

Author Response

Please refer to the attachment.  

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript by authors Guitang Gao et al. entitled "Estimation of chlorophyll content in wheat based on optimal spectral index" addresses an interesting topic with a high level of novelty. The introduction is developed sufficiently and explains the rationale for this study. However, the methodology is not described adequately. Given the number of results presented, I also think the discussion is poorly developed (with only two authors discussed). The conclusion lacks the development of the idea of linking the obtained results to agronomic practice. I have the following specific comments on the manuscript: i) the weather conditions of the season and the soil properties belong to the factors that can influence the chlorophyll content of wheat leaves, I kindly recommend the authors to add the soil properties of the experimental site and the weather course of the different months (temperature, precipitation) to better understand some of the differences in the results; ii) I regret to note that the study does not describe the wheat cultivation technology, the above needs to be added; iii) the description of the biological material - wheat variety is also lacking; iv) the discussion needs to be revised; v) the conclusion needs to be revised. I believe that after incorporating the suggestions, the manuscript will be suitable for publication in Applied Sciences.

Author Response

Please refer to the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

In this work, the authors present an interesting application of the reflectance spectroscopy for an accurate estimation of chlorophyll content in wheat leaves, which in turn could provide information on its growth, nutritional status, and possibly a systemic field management. In order to achieve this, they evaluate the potential of multi-spectral data to estimate the chlorophyll content of wheat leaves, covering different growth stages. By using an analysis based on the optimal spectral index, polynomial regression, random forest, decision tree and artificial neural network the finally establish and optimize an estimation model of chlorophyll value; as one would expect the results showed that the optimal spectral indexes at the four growth stages were mainly composed of red band, red edge band and NIR band, and the optimal spectral indexes at the grain filling stage had the highest correlation with the chlorophyll value. 

 

I consider that the main issue with this work is the lack of a connection between the analysis methods, its results and the actual significance of the measured magnitudes (amount of chlorophyll)

Please consider the following observations prior to the paper publication:

 

1.      In my opinion this paper deals with multispectral information instead of hyperspectral. Please perform a better check on your bibliographical sources in order to better clarify this.

2.      Provide a representative spectrum of your measurements, in which you clearly establish the relevant spectral features associated with your study. This is important for the interpretation of the results, and to provide a scientific basis on the study.

3.      In connection with the previous observation, how does the MVC or the chlorophyll prediction correlate with the spectral features? Prior to the mathematical analysis performed in this work.

4.      Provide a better description on how figure 2 is generated. Do the maps consider different measuring points? What is the meaning of the mirror-like image?

5.      In order to provide useful information, provide a better discussion on trying to describe if there is a relation between the representative parameters from your models and the growth stages.

6.      Provide the uncertainties in figure 3, and clearly discuss how do you calculate them.

7.      Discussion and conclusions. Improve the presented information by discussing your results, providing some spectral meaning. Check again observation #2. As it is now it lack of what the reader could consider the scientific basis of the problem.

Comments on the Quality of English Language

Please perform a general check for typos and syntax along the text

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed most of the issues in a proper way, however there are still two major issues:

 

1. As pointed out before, In my opinion this paper deals with multispectral instead of hyperspectral analysis. Please clearly establish in the text (providing bibliographical references) why this should be considered "hyper-".

 

2. (comment 6) You discuss the origin of possible uncertainties, but still fail to provide an actual value on the plot. Perhaps from your discussion, you could try to (at least) estimate a value for the uncertainties on each point presented on the plot (Fig.3)

 

Comments on the Quality of English Language

Perform a general check for minor typos and errors

Author Response

Please see the attachment.

Author Response File: Author Response.pdf