Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a well-written manuscript. However, I was left with several comments that needed to be addressed while revising the manuscript.

1. The authors applied the SPAD-based method to explore chlorophyll absorption bands using a sensor limited to the SWIR region (1000-2500nm). However, strong chlorophyll absorption bands are normally obtained in the spectral region of 400-1000nm, particularly in the visible region. This aspect should be clarified. I strongly recommend revising this discrepancy in the methodology or providing supporting references for the chosen spectral range. Additionally, the SPAD measurements may also serve as a good indicator of overall crop health, so their application in this context may also be justified.
2. Provide sufficient references supporting the characteristic peak of chlorophyll content at 5000-7000 cm⁻1.
3. The chlorophyll measurements from the laboratory are suggested for precise growth stage-specific modelling of Chlorophyll Content rather than using SPAD measurements. The authors may explore their relationship with SPAD values at each growth stage.
4. The methodology section has not sufficiently explained the modelling approach using vegetation indices. Why were only six indices selected for chlorophyll analysis?
5. Include a detailed flowsheet illustrating the overall workflow.
6. A study area map with details of experimental plots and sampling design is suggested.
7. The authors should justify the bands selected through different FS techniques and discuss the relevance of each wavelength to the specific constituents being analysed.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The study aims to develop non-destructive, growth stage–specific models for predicting chlorophyll content (SPAD values) in Korla fragrant pear (Pyrus sinkiangensis) leaves by combining near-infrared/short-wave infrared spectroscopy with vegetation indices and machine learning methods. The authors demonstrate that integrating near-infrared spectroscopy with vegetation indices and machine learning yields accurate, stage-specific chlorophyll prediction models, advancing precision orchard management.

1. For Figure 1, you show clear differences in spectral reflectance at 5000–7000 cm⁻¹ and 7000–8000 cm⁻¹ across growth stages.
   Could you elaborate on whether these spectral shifts are more strongly influenced by chlorophyll changes or by water content variation during each stage?
2. The discussion nicely links spectral changes with chlorophyll/water content at different stages, but the physiological insights could be expanded. For example, why do stage-specific models outperform full-season models from a plant development perspective? Adding mechanistic explanations (chlorophyll synthesis vs. degradation, water redistribution) would enhance the biological relevance.
3. For Figure 8. In the combined spectral + vegetation index models, performance improved notably.
   Which vegetation indices contributed most strongly to accuracy improvement, and do they directly correlate with chlorophyll-related physiology (e.g., NDVI, GCI)?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

1. No line numbers are provided in the manuscript.

2.In the Introduction, the research objective is stated twice (e.g., “The aim was to …” and “This study aims to …”), resulting in redundancy.

3. The Introduction also includes result-oriented statements (e.g., descriptions of Fig. 7, such as “The results demonstrate …”), which are inappropriate for this section.

4. The Introduction is overly long and verbose.

5. Experimental design concerns:
The use of spectral imaging for SPAD value analysis is generally intended for real-time, non-destructive prediction. However, in this study, leaves were detached, freeze-dried, and then subjected to spectral imaging, which diminishes the practical significance of the approach.
If the freeze-drying was conducted to avoid issues such as water-related peak overlap, it would still be necessary to also acquire spectral images from fresh (non–freeze-dried) leaves. This would allow for a comparative analysis between the two datasets and facilitate the development of a model capable of real-time, non-destructive analysis.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

I confirm that the requested revisions have been fully addressed in the revised manuscript.

The revised manuscript is acceptable for publication in this journal.