Maize Canopy and Leaf Chlorophyll Content Assessment from Leaf Spectral Reflectance: Estimation and Uncertainty Analysis across Growth Stages and Vertical Distribution

Round 1

Reviewer 1 Report

1. The Non-uniformed vertical distribution of LCC in crop canopy has great influence on canopy spectral reflectance, and the subsequence estimation of chlorophyll content at canopy scale (CCC). The topic of this study is very interesting. The author tried to estimate CCC of maize using spectral data measured at leaf scale, this is a relative new methodology, and in deed needs the theory to support the conclusions. However, the innovation of the study is not enough to be published at present. For example, the authors only used early published VIs to establish the linear model for LCC estimation, simple multivariable stepwise regression to establish the LCC-CCC model, etc.
2. There is something wrong with Equation (2): the unit in the left is ug/cm², whereas the unit in the right is mg/cm².
3. In the Leaf sampling and chlorophyll measurement part, as the authors stated that “the canopy chlorophyll content (CCC, μg/cm2) was used to represent the chlorophyll status of the canopy”, however, why LAI was not be considered when calculating CCC? From the description of the authors mentioned in Line 180-182, “Measurements at several (three to four) leaf positions were halted during the vegetative period because the leaves grew rolled together at the top of the canopy, which hindered chlorophyll extraction and spectral measurements.” The LAI can not be simply added by the total leaf area. What’s more, the methodology will directly influence the following empirical model of LCC and CCC.
4. Line 465-468: “Although the NIR band did not show sensitivity to chlorophyll (Figure 3, d), models using non-NIR bands (such as PPR) showed negative relationships with chlorophyll (R2 = 0.18). Previous studies have shown that the NIR band has the ability to reflect cell structure and biochemical content 468 (such as proteins or starch) [50-53].” The illustration doesn’t make sense, because the authors didn’t explain the reason why NIR bands is needed.
5. Line 524-525: “This implies that UAV has the capacity to overcome the saturation phenomenon by monitoring the sensitive leaf position.” I do not think this statement is correct. UAV is a platform that used to mount sensors and acquire remote sensing data, the factors that would induce the saturation phenomenon are the sensor, the spectral reflectance or the condition of crop canopy, etc.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The study assessed leaf chlorophyll content of Maize crop at various leaf vertical positions and determined the optimal positions for the retrieval of canopy chlorophyll content based on leaf reflectance. The study developed calibration equations for multiple vegetation indices based on multiple stepwise regression. The study is relevant and will be interesting to various readers. 

The manuscript contains a lot of confusing sentence constructions which make it hard to read and follow. Therefore, it should be edited by a professional English editor or English speaking colleague.

The manuscript has a limited review of previous related studies in the introduction. I have a reservation about the choice to use 20% for modelling and 80% for validation. The 2020 data is used for modelling and 2019 data is used to validate the models. It would be great if the data were combined and divided into 70%/30% calibration and validation data. Also, the field spectral data have to be resampled to the spectral settings of the target sensor, instead, the authors used the band centres only which do not represent the width of the band. Leaf positions (L1-L18) need to be defined in detail. There is inconsistent use of terms throughout the manuscript. 

My detailed comments are as follows:

The title is a bit confusing in its current form. Please rephrase it. 

Please capitalize the "Leaf Chlorophyll Content" and "Canopy Chlorophyll Content" throughout the manuscript. 

L15: Please consider rephrasing the sentence as follows: "It has a non-uniform vertical distribution in Leaf Chlorophyll Content (LCC), which limits remote sensing of Canopy Chlorophyll Content (CCC)."

L23: "late growing stage" - Please be specific here about the growing stage.

L23: Please indicate relative to what? Correlation can be better described as strong / moderate / weak positive or negative correlations. Please address this throughout the manuscript.

L23: "mREP" - Please abbreviations in full on the first mention both in the abstract and main text. 

L25: "Combining.." please add "By combining..."

L24-25: "spatio-temporal conditions" - Please be specific here. Which spatio-temporal conditions are the authors referring to? An abstract should be able to stand alone and it is an important part of the manuscript, as readers will decide to give your manuscript attention based on it.

L26: "L14 and L15.." - Please describe these positions first.

L26: "...in the..." - Consider deleting

L27: "to have" - Consider replacing with: "resulted in"

L27: "stability" - Consider replacing with a more specific term here and throughout the manuscript.

L29: "by" - Consider replacing with: "using"

L30: "years" - Please specify the years. 

L56: "has within" - Please insert "has been found to have..."

L56-57: Please provide a brief review of these different methods and identify their limitations relative to VIs. Provide motivation for using VIs over other methods.

L58: "" - Please consider inserting "a range of"

L63: Please a more specific term such as "heights" or "positions"

L57: Please cite a study where the R-squared value is reported.

L66:"is not stable"- please replace with "not consistent"

L64 "plant analyzer development" - please capitalize

L71: "divided into upper, middle, and lower ranks" - This should precede sentences where rank is first mentioned. Also, define what these positions are in heights or other quantitative terms to allow replication in other areas

L78-L83: Please review other methods for determining CCC, such as LAI*LCC

L105: Give a scientific name for maize as well.

Table 1: please remove the first column, the table is hard to read in present form.

Table 2: Please consider removing the units for N, and write it in the caption

L130: "nitrogen" - please include "(N)"

L132-135: What was the impact of different stages of using 2019 and 2020 data on modelling?

L170: Please capitalize "field of view"

L173: Please mention the resampling method.

L178: Why was it necessary to calibrate for every measurement?

Section 2.4. The spectral reflectance data has to be resampled to the spectral settings of the target sensor (RedEdge-MX).

L186: Please replace "spectrum" with "spectral"

L199:  why is the model built with only 20% of the data? How much was the 20% of the data?

L217: Replace with "Multivariate..."

L220: Please capitalize "Variance inflation factor"

L235-236: Can the authors explain why 2019 is used to validate the models build with 2020 data.

Section 2.5.4. Are the previous steps not considered "statistical analysis"? Why is this section titled "statistical analysis" and not others? This section should be titled "Validation metrics"

L241: "drafting" what does this mean in the current context?

Figure 3. Some colours in the legend should be changed as they are currently not visible. 

L307: Please call it what it is. Coefficient of determination.

L246-L247: What kind of difference was tested here?

L309: Please indicate where the results are presented in brackets.

Figure 5 caption "Differences between observed and predicted"- please replace with simply "rRMSE (%) for the six LCC-VI models (a) ..." L1-L18 need to be further defined somewhere in the Methods section.

Section 3.6 Please consider deleting "Establishment and verification of the"

L396: "selected sensitive..." It is not clear what the authors are referring to here.

L399: Please indicate Beta scores and R2 values near where you report leaf positions and their relationships with LCC/CCC.

L411: "serious" Please choose a more specific term.

L422, L436 and L439:"sixteen leaf stage" -This is confusing. please be consistent in the use of terms. Is this the same as L16?

L426: Are the rRMSE significantly different to support the claim of stability? Also, choose a more specific term instead of "stability".

L446-447: replace the last part of the sentence with: "remote sensing of crops"

L456: "And" - It is unconventional to start a sentence with a conjunction. Please consider removing it.

L474: There is inconsistent use of terms throughout the manuscript. Sometimes "leaf spectrum" other times "leaf reflectance"

L482: What do the authors regard as the "middle growth stage". please distinguish between growth stages and leaf positions.

L486-487: What was the reason?

L497: "stable" what does stable mean in the context of correlations?

L517: Authors make recommendations about UAVs that they did not use in the current study. I suggest the entire paragraph be removed.

 

 

 

 

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Overall, this is a clear, concise and well-written manuscript. The introduction is pertinent and based on interesting papers.

The procedure is described in details and gives sufficient information on the study logic.

In addition, the results are clear.

Kind Regards

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 4 Report

The study investigated the vertical distribution of LCC, its relationship with N supply, and its relationship with CCC. This topic is a very interesting. The findings are thought-provoking. The manuscript is well written. I only have several minor suggestions.

1. It might be intuitive to add scatter plots showing relationships between LCC and CCC with different leaf location, N and stages.
2. The implications for remote sensing can be further extended. For example, remote sensing signals, in particular satellite RS, mainly come from top of canopy. On the other hand, canopy radiative transfer models assume a uniform LCC within canopy. How do you think of this mismatch in LCC/CCC retrieval? Many field work do not account for vertical variations, then do those validation make sense? UAV can see leaves down into canopy, but how to use that observation (e.g., to get leaf reflectance), given the complex radiation environment within canopy? Such discussion could potentially make a bigger impact of this study. 

L74. Missed a ".".
Fig. 3. Y-axis is blocked in (f).

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

My comments have been solved. The manuscript can be considered to be accepted.

Reviewer 2 Report

The authors have addressed my concerns. Thank you.