Assessing Suitability of Sentinel-2 Bands for Monitoring of Nutrient Concentration of Pastures with a Range of Species Compositions
, David J. Humphries 3 and Christopher K. Reynolds 3
Round 1
Reviewer 1 Report
The manuscript aiming at assessing the potential of Sentinel-2 satellite data on monitoring pasture quantity and quality is generally well written. The topic is of interest to the journal. My major comments include, (1) the Abstract, Result, and Discussion sections, especially the Discussion section highlighted the influence of pasture parameters on canopy reflectance. The current analysis only explored the correction between parameters and different two-band-based NDI. This analysis cannot directly assess the sensitivity of each pasture parameter on reflectance. A study employing a method to assess the sensitivity of parameters on the reflectance of Sentinel-2 bands is required in the revision. In addition, based on the field collection, the canopy radiative transfer model such as PROSAIL may be useful to explore and discuss the contribution of different pasture parameters on reflectance and vegetation indices. (2) explanations of using NDVI-like vegetation index in the analysis are needed. How about the other vegetation indices, such as RVI-like or EVI-like? 3) Total number of samples in each parameter was not clearly present.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
Submitted article falls into the scope of Agronomy journal and I found it as interesting and useful. The article is focus on testing whether Sentinel-2 MS bands can be used to determine pasture quality and quantity parameters and understanding the influence of these parameters on pasture quality estimation.
Introduction is well written with 25 relatively new references describing sufficiently the issue solved in this study. Aims of study are clearly defined.
Page 2 L 65 – please add the numbering of references.
Materials and Methods chapter describes the data and methods used appropriately and clearly. However – Page 4 L165 – please, if it is common practice, the standard must be stated (eg European or American).
Results and Discussion chapter are also well written.
Conclusions are clearly defined and in line with the objectives in the Introduction chapter.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 3 Report
The manuscript is well written, the methodology is sound and the results are clearly presented and discussed.
My main concern is related to the use of 2nd and 3rd order polynomial regressions to link the traits of interest with the Sentinel 2 bands, especially when looking at figure 4: in all cases, except maybe for g), the trend does look linear, although the points can be scattered. Without any strong evidence that suggest to use a 2nd or 3rd or nth order polynomial regression, I would avoid to use such an approach as the risk of overfitting is very important. However, as I understood it, the aim of the work presented in this article is not to provide a user ready solution for quality estimation, but rather to explore the potential of a simple, robust method based on adjusted VI, which has been clearly demonstrated.
As a specific comment, units on figure 2 (NDF) need to be updated.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
