A Transformed Triangular Vegetation Index for Estimating Winter Wheat Leaf Area Index

Round 1

Reviewer 1 Report

The authors present a comparison between vegetation indices for the estimation of winter wheat LAI. The novelty of the study is the development of a new vegetation index called TTVI aimed of reduce the well-known saturation effect that affect most of the commonly used vegetation indices (e.g. NDVI).

The results seem promising and is good to see the development of new indices based on strong theorical basis. In my opinion, the authors, appropriately developed the new index using hyperspectral bands and, in order to make it more widely applicable, correctly converted it for a widely used multispectral sensor (i.e. Sentinel-2).

I only have a comment to make the paper stronger. Winter wheat, despite being the subject of the study, is only nominated briefly in the paper. I would suggest integrating the manuscript (introduction, and discussion) with more background on the remote sensing of the plant.

Author Response

Dear reviewer,

We are truly grateful for your kindly comments and suggestions on our manuscript. 

Point 1:I only have a comment to make the paper stronger. Winter wheat, despite being the subject of the study, is only nominated briefly in the paper. I would suggest integrating the manuscript (introduction, and discussion) with more background on the remote sensing of the plant.

Response 1: Remote sensing has become a widely-accepted technology for large-scale monitoring in agriculture for last several decades due to its fast and non-destructive characteristics. Researchers have used remote sensing for different crop types. Winter wheat, as one of the most important cereal crops in China has been studied in the manuscript. According to your suggestion, we have added more descriptions in introduction and methods as below:

line 51:LAI remote sensing retrieval methods have been widely investigated for several decades. During the past years, researchers have conducted studies on different vegetation types like broadleaf forest, coniferous forest and crop including soybean, maize and winter wheat[4-6]. LAI values vary with different vegetation types for different phenology. 

line 115:Winter wheat (Triticum aestivum L.) is a vital crop in the world and also one of the most important cereal crops in northern and southern China.It is planted widely and regularly in China every year which makes China a main wheat producing country. Knowing the growth condition and yield of winter wheat timely can provide technological suggestions for food security. 

 

Reviewer 2 Report

This manuscript introduces a new index for LAI estimation using field and satellite datasets.  I have seen some issues that need to be addressed, which I detailed below.  

 

Abstract

Line 29: Do you really need those many significant numbers for RMSE? 

 

Intro

Lines 60 to 63: I do not think this is true.  While these are the common groups of VIs, there is another VI group that is based on the shape of the reflectance curve.  The red edge of the vegetation spectrum is one of them.

 

Darvishzadeh, R.; Atzberger, C.; Skidmore, A.K.; Abkar, A.A. Leaf area index derivation from hyperspectral vegetation indices and the red edge position. Int. J. Remote Sens. 2009, 30, 6199-6218.

 

I suggest to create a new group and then add the corresponding references.  Also add a new section 2.4.5. Here are some more references on shapes of curves.

 

Kouadio, L. et al., 2012. Estimating regional wheat yield from the shape of decreasing curves of green area index temporal profiles retrieved from MODIS data. International Journal of Applied Earth Observation and Geoinformation 18, 111–118

 

Salas EAL, Henebry GM (2013) A New Approach for the Analysis of Hyperspectral Data: Theory and Sensitivity Analysis of the Moment Distance Method. Remote Sensing 6 (1): 20-41.

 

On section 2.4, you need to have a reference to Table 1.

 

The choice of these 11 VIs on Table 1 must be justified.  While there has been a sentence that says that they are widely-used VIs for LAI, this is not sufficient.   Also, there are other VIs for LAI.  Check this link:

 

https://www.indexdatabase.de/

 

Line 244: The definition/concept of TTVI is very simple but isn’t well described here.  Provide a clear diagram that would show the vertices of the triangle used to demonstrate how equation 1 is derived.  I tried to follow the equation but it won’t result to an area of a triangle. 

 

Also, how would TTVI be defined if used in other sensors that do not have those exact wavelengths?

 

In Figure 2, it is assumed that the red-edge position does not change and so when there’s an increase in NIR, the triangle (area) increases….TTVI increases too!  However, the REP varies as well with regards to the variations of R and NIR.  The possibility of having the same values of TTVI for two different reflectance curves is very likely!  How would your new index deal with this issue?

 

Line 270: Check spelling.

 

Line 273: Starting a paragraph with a numeric character?

 

In the method section, there has been an interchange of active and passive voices all throughout.  Be consistent.

 

Results

Again, check the number of significant numbers you need to use.

 

In table 2, TTVI is ranked 1 and MTVI2 is second.  Although the R^2 values of the two are different, they may not actually be significantly different.  Have you tested whether or not these R^2 values are different significantly?  Performing this test is crucial in your attempt to introduce a new index.

 

Do the same test for the other tables too.

 

Line 324 to 327: There are only two sentences that explain the results of the TTVI against LAI.  Looking at Figure 3, TTVI obviously does NOT do well at higher LAI.  There is clustering evident at lower LAI and dispersion at higher LAI.  If you divide these points into two groups according to LAI, surely, TTVI could come out with poor relationship against LAI.  Is there an explanation for this behavior of TTVI based on your results?

 

Again in Figure 4, the clustering of TTVI at lower LAI is evident for measured vs estimated.   It looks like that EVI and TVI are the best indices in this Figure.  Please discuss this behavior of TTVI.

 

I would recommend a major revision and edit the manuscript according to the suggestions.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Intersting work.

Why did you choose winter wheat? Is there an application of the index proposed on winter wheat? Justify your choice better.

You did not discuss the relationship between LAI measures and the observed surface uniformity.

The notion of the saturation effect, I find that is not sufficiently explained. What is its importance?

More precision on the notions of hight vegetation, low vegetation, etc.

Present in a clearer way the strengths of the proposed index.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

This study created a new vegetation index and evaluated the performance by comparing existed 11 indices. The topic is interesting and the manuscript is well written and easy to read. However,  I feel some parts should be improved because there are unclear parts.

1. Method

How did you create the model estimating LAI? Linear regression? non-linear regression?

2. Discussion

Could you please compare the results with previous studies?Especially, you used 11 vegetation indices. Were the performances comparable to previous studies?

 

Detailed comments

L133. Please add information of measurement height of LAI-2200.

Fig. 2. (a) The figure forms the basis of your hypothesis but I don't understand how to calculate it. Did you calculated the correlation from your field measurements, as written in L265?

Fig 3. I think you tried to sort the indices according to the order of indices in Table 1 (i.e. DVI, SR, ...). However, Fig 1 started from SR (Top left should be DVI).

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear authors,

Kudos to you for quickly revising the paper according to my comments.  This new version of the manuscript has improved tremendously compared to the previous one. One specific improvement is the clear definition of the derivation of the TTVI in section 2.5.  Another improvement is the addition of shape indices in your tests and checking their significant differences.

I have a couple of minor suggestions:

Line 351: change "form Figure 3" to "from Figure 3"

In your response to my question you said "As TTVI was calculated by NIR and red-edge bands, it might not be possible to use for sensors without these bands, but it can be used for other satellites with red-edge bands of similar wavelengths."  I think this line is important and must be included somewhere in the manuscript as a limitation of the TTVI.

Once these edits are done, I would recommend acceptance for publication. 

Cheers!

 

 

 

 

 

Author Response

Dear Reviewer,

Thank you very much for the comments and suggestions which help us improve the manuscript. We have made some adjustments according to your comments and highlighted them in yellow in the manuscript to be distinguished from the revisions of Round 1. Here are our responses point-by-point:

Point 1:Line 351: change "form Figure 3" to "from Figure 3"

Response: Sorry for the mistake. We have fixed the spelling error.

Point 2: In your response to my question you said "As TTVI was calculated by NIR and red-edge bands, it might not be possible to use for sensors without these bands, but it can be used for other satellites with red-edge bands of similar wavelengths."  I think this line is important and must be included somewhere in the manuscript as a limitation of the TTVI.

Response 2: Thank you for the suggestion. We have added the limitation of TTVI from line 485 to line 487 as "To be noted that TTVI was calculated by NIR and red-edge bands, it might not be possible to use for sensors without these bands, but it can be utilized by those with red-edge bands of similar wavelengths."

Thank you again for the comments and suggestions. We hope that the manuscript is ready for publication now after our revision. 

Best,

Naichen

On behalf of all authors