Estimation of the Leaf Area Index of Winter Rapeseed Based on Hyperspectral and Machine Learning
, Junyu Fu 1,2,3, Wenjie Xu 1,2,3, Youzhen Xiang 1,2,3 and Fucang Zhang 1,2,3
Round 1
Reviewer 1 Report
Please see my comments attached.
Comments for author File: 
Comments.pdf
Moderate editing of English language required
Author Response
Responses to Reviewer #1 (Manuscript Number: sustainability-2506620)
General comments:
Research study focused on the Estimation of LAI based on hyperspectral and machine learning looks pretty interesting. In modern agriculture and digital technology era， conserving time and energy is utmost important. Measuring LAI using hyperspectral and machine learning techniques would bring feasible options in the research field.
Hypothesis and the key objectives are well established with the accurate methodology.
In the introduction part, some more references and related studies need to be highlighted about the Back Propagation Neural Network (BPNN), Support Vector Machine (SVM), and Random Forest (RF)...Specifically in different crops...some statistics regarding the outcomes of other researchers also need to be incorporated to enrich the introduction part.
Response: Thank you for your careful review and positive comments. We have revised and added relevant content to this section as suggested, and we have also revised the corresponding references.
Line 79–91: However, these studies have mainly focused on soil and environmental disciplines [14] and other crops, less on winter oilseed rape. Xiang et al [15] studied the LAI of soybean at flowering stage under different nitrogen application levels and film mulching treatments. The original hyperspectral reflectance data of soybean at flowering stage were processed by 0 - 2 order differential transformation, and the index with the highest correlation with LAI of soybean at flowering stage was selected as the optimal spectral index for input. Three machine learning methods: SVM, RF, and BP neural network optimized by genetic algorithm (GA-BP), were used to construct soybean LAI prediction models. Conclusions are as follows: 1.5-order differential and RF methods are the optimal differential order and the optimal model construction method in this study, respectively. The R2 of the optimal soybean LAI estimation model modeling set and validation set are: 0.890 and 0.880, RMSE are: 0.348 cm2 / cm2 and 0.320 cm2 / cm2, NRMSE are: 11.278 % and 10.354 %, MRE are: 9.795 % and 9.572 %, respectively.
Methodological section contains enough information about the models, processing and evaluation. Study area image (Fig 1.) need to be enlarged with high resolution picture. Why this area was chosen for the study need to be highlighted in the Material section.
Response: Thank you for your careful review and positive comments. We have made revisions to the section, we have replaced the higher resolution images, and we have revised the content of the material to include an explanation of why this area was selected for the study.
Line 103–112: The experiment was carried out at the Water-saving Irrigation Experimental Station of Water-saving Agriculture Research Institute of Northwest A & F University (34 ° 14 ′ N, 108 ° 10 ′ E, altitude 521 m), the location of the experimental area is shown in Figure 1. Shaanxi Province is an important production base of rapeseed, the experimental area is located in the Guanzhong Plain irrigation area, it is a warm temperate monsoon semi-humid climate. This specific climatic condition has an important impact on the growth and yield of winter rapeseed. The experimental station is equipped with irrigation system and meteorological monitor-ing station, which provides the necessary conditions for the experiment. The winter oilseed rape variety ' Shanyou 18 ' cultivated by Northwest A & F University was sown on October 3,2021
Result part need to be improved by focusing on the significant outcomes. Accordingly discussed with supporting references. Seems there are some more outcomes need to be drawn in the results. Focus on this section.
Response: Thank you for your careful review and positive comments. We have modified the Results section to add some new results as suggested
Line 245–264: Table 4 presents the prediction results of different models for winter oilseed rape LAI estimation on the modeling and validation sets under 0th and 1st order differential changes. It can be observed that under the 1st order differential, the R2 values of all three models are higher than those under the 0th order, and the validation set R2 values of all three models are above 0.6, which indicates that all three models exhibit good linear fitting results. Among them, under the 1st order differential, the RF model achieves a high validation set R2 of 0.810. This is significantly higher compared to the BPNN and SVM models, which have an increment of 0.171 and 0.180 in R2, respectively. The RF model under the 1st order differential also demonstrates a lower validation set RMSE of 0.455 cm2/cm2, reducing by 24.0% and 31.3% compared to BPNN and SVM, respectively. Additionally, the MRE of the RF model under the 1st order differential is 10.465%, which is lower by 30.5% and 33.9% compared to BPNN and SVM, respectively. When considering the same modeling method, the order of accuracy for winter oilseed rape LAI estimation among the three models is as follows: RF > BPNN > SVM. Therefore, the 1st order differential processing combined with the RF method are identified as the optimal fractional differentiation and modeling approach in this study. The LAI estimation model for winter oilseed rape constructed using this approach achieves a training set R2 of 0.705, RMSE of 0.725 cm2/cm2, and MRE of 14.562%. Moreover, the validation set R2 is 0.810, with an RMSE of 0.455 cm2/cm2 and an MRE of 10.465%. The evaluation results of the model are shown in Figure 3.
Conclusion section is fine.
Additional comments:
- L324:716&724
Response: Thank you for your careful review and positive comments. We've made changes here.
- L324：check repetition of number
Response: Thank you for your careful review and positive comments. We have removed duplicate numbers.
- Use the symbol “-" to show the ranges 670-760. Replace throughout the draft.
Response: Thank you for your careful review and positive comments. We have checked the entire manuscript and made corrections for similar expressions.
- Table 4. Check the title. Authors did not check the corrections adopted before
the submission.
Response: Thank you for your careful review and positive comments. We have made changes to the title of Table 4.
- Table 2. Why there is Chinese language in table.? Replace with English.
Response: Thank you for your careful review and positive comments. We have modified the Chinese that appeared in the text and replaced it with English.
- Provide a footnote under every table to abbreviate the things for quick understanding.
Response: Thank you for your careful review and positive comments. We have added footnotes to Table 2, Table 3 and Table 4, but we feel that Table 1 is already very clear and there is no need to add footnotes.
- Check the spacing properly throughout the draft.
Response: Thank you for your careful review and positive comments. We checked the entire manuscript and unified the spacing.
- Units superscripts need to be checked.cm2 use either cm-2 cm-2 or cm2/cm2 follow similarity throughout the manuscript.
Response: Thank you for your careful review and positive comments. We decided to use cm2/cm2 and modified it throughout the manuscript.
- High quality resolution pictures may be replaced.
Response: Thank you for your careful review and positive comments. We have replaced low resolution images with high resolution images.
- There are several language and grammatical mistakes need to be rectified before accepting.
Response: Thank you for your careful review and positive comments. We have checked the grammar and corrected it.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear Authors,
I have thoroughly reviewed your article titled "Estimation of leaf area index of winter rapeseed based on hy-perspectral and machine learning" and must commend you on the well-structured and well-written piece. The research you conducted on winter oilseed rape and leaf area index (LAI) using hyperspectral data is commendable and will undoubtedly have a significant impact on the scientific community.
After careful examination, I have identified a few minor adjustments that could further enhance the clarity and precision of your work. Please find the suggested revisions outlined in the attached PDF document. These changes are aimed at strengthening the already valuable contributions you have made in your study.
Comments for author File: Comments.pdf
Author Response
Responses to Reviewer #2 (Manuscript Number: sustainability-2506620)
Specific comments:
Materials and Methods
- line 101 Figure 1: Improve the quality of image and write a self-explanatory caption
response: Thank you very much for pointing out. We have replaced it with a higher resolution image and we also revised the title.
Line 114：Figure 1. Geographical location of the test area
- line 146-147 table 2: Translate in English
response: Thank you very much for pointing out. We have translated the Chinese into English.
- Line146-147: Harmonize the text (Font, dimension etc..)
response: Thank you very much for pointing out. We have made changes to this section, harmonizing the text and fonts in the tables, etc.
Results
- Line 251: Improve the quality of image and write a self-explanatory caption
response: Thank you very much for pointing out. We have replaced with a higher quality image and also revised the title.
Line 269 –270: Figure 4. Correlation between the predicted and measured values of winter rapeseed LAI of 0-order and 1-order based on BPNN, RF and SVM.
Conclusion
- Line 308: text deletion
response: Thank you very much for pointing out. We have changed the ‘Conclusions’ into ‘Conclusion’.
- Line 324: text deletion
response: Thank you very much for pointing out. We have removed the duplicate (2).
Author Response File: Author Response.docx
