Hyperspectral Data Can Classify Plant Functional Groups Within New Zealand Hill Farm Pasture

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The authors have meticulously revised the article according to the feedback. Their efforts have significantly improved its quality, so I recommend publishing it in its current form.

Author Response

Comment 1: The authors have meticulously revised the article according to the feedback. Their efforts have significantly improved its quality, so I recommend publishing it in its current form.

 

Response 1: We thank the reviewer for their feedback that highlighted the improvements.

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

This manuscript uses hyperspectral data to classify plant functional groups within heterogenous New Zealand hill farm pasture. I do not recommend accepting this manuscript as it has the following issues.
1. This manuscript is more like an experiment report. The logic of the manuscript is confused, which does not match the quality of a research article.
2. There is no innovation in this manuscript. Lines 104-108 of the manuscript state “Hyperspectral data has been used to describe pasture or grassland quality [26, 35] but generating a classification or map of species distribution [8] is less common. It is the latter that is needed for critical on-farm and off-farm decision making. The goal of this research was to develop and validate a simple technique to map pasture quality using hyperspectral data and Support Vector Machines (SVM) at farm scale.”. 
There have been many studies on hyperspectral data classification based on support vector machine (SVM) method. And to my knowledge, there have been multiple papers using hyperspectral data to classify pasture or grassland, such as:
(1) Yokoya N, Yairi T, Iwasaki A. Coupled non-negative matrix factorization (CNMF) for hyperspectral and multispectral data fusion: Application to pasture classification[C]//2011 IEEE international geoscience and remote sensing symposium. IEEE, 2011: 1779-1782.
(2) Jain D K, Dubey S B, Choubey R K, et al. An approach for hyperspectral image classification by optimizing SVM using self organizing map[J]. Journal of Computational Science, 2018, 25: 252-259.
(3) Zhao X, Pan X, Yan W, et al. Visible‑NIR hyperspectral classification of grass based on multivariate smooth mapping and extreme active learning approach[J]. Scientific Reports, 2022, 12(1): 9017.
3. The Abstract should begin with the importance of monitoring species composition or habitat distribution for pasture management, rather than introducing the work done in this manuscript. The same issue exists in the Introduction, which should first introduce the significance and necessity of conducting this experiment, and then introduce the characteristics of hyperspectral remote sensing.
4. The Introduction lacks sufficient information on the relevant references of hyperspectral data classification.
5. Supplement the process of hyperspectral data processing.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Our response to this reviewer is in the attached word document.

Author Response File: Author Response.docx

Reviewer 3 Report (New Reviewer)

Comments and Suggestions for Authors

The hyperspectral data used has to be properly described. The  amount of dimension reduction before classification should be include.

Author Response

Comment 1: The hyperspectral data used has to be properly described. The amount of dimension reduction before classification should be included.

 

Response 1: We thank the reviewer for their valuable comments to improve this manuscript.

We have added information to the methods to address the comments.

Line 174 "The hyperspectral data covered the range from 380nm to 2500nm in 448 spectral bands."

Line 222 "10 transformed components from the MNF were included in the analysis"

 

All changes are tracked for clarity.

Reviewer 4 Report (New Reviewer)

Comments and Suggestions for Authors

see attached

Comments for author File: Comments.pdf

Comments on the Quality of English Language

The grammar and sentence structure are poor. The whole manuscript should be rewritten to improve readability

Author Response

Our response to this reviewer is in the attached word document.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

The English could be improved to more clearly express the research.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Reviewer 4 Report (New Reviewer)

Comments and Suggestions for Authors

Being a native English speaker does not necessarily ensure one would write a good scientific journal.

Comments on the Quality of English Language

This manuscript still needs to be improved the writing, eg-

Line 425-'The 1st derivative full spectrum data was deemed most accurate overall so was used 
in the subsequent site validation visit and owner discussions.' rewrite

The authors should revise the whole draft again to polish the writing.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Please check the attachment, thank you!

Comments for author File: Comments.docx

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper attempted to classify the plant function groups using hyperspectral data. The theme is within the journal's scope and is meaningful. However, it was poorly written and lacked innovation. So, I recommend rejection. Here are my suggestions:

(1) The abstract needs to be rewritten. Some key information is missing. For instance, the difficulty of classifying plant functional groups. Meanwhile, there is no introduction to the hyperspectral data.

(2) All the maps are unqualified, lacking important information such as scale and north arrow.

(3) This article only uses SVM for classification and lacks innovative.

(4) The Conclusions Section should be rewritten to elaborate on new discoveries and conclusions.

(5) English should be improved.

Comments on the Quality of English Language

The English could be improved to more clearly express the research.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf