Assessing Drought Stress of Sugarcane Cultivars Using Unmanned Vehicle System (UAS)-Based Vegetation Indices and Physiological Parameters

Round 1

Reviewer 1 Report

The manuscript by Ittipon Khuimphukhieo et al. examines drought tolerance in sugarcane cultivars through unmanned aerial systems (UAS)-based vegetation indices (VIs) and physiological parameters. It aims to identify drought-tolerant cultivars using VIs in comparison with traditional methods and evaluate the accuracy of VIs-based prediction models in estimating stomatal conductance (Gs) and chlorophyll content (Chl). Conducted in a randomized complete block design with seven cultivars, the study finds consistent identification of drought-tolerant and sensitive cultivars between VIs and traditional methods. Additionally, the random forest model outperforms linear models in predicting cultivar performance in untested environments for both Gs and Chl.

Major Issues

1. The manuscript would benefit from a more detailed explanation and justification of the methodology, particularly the use of specific VIs and the selection criteria for the seven sugarcane cultivars. For instance, why were these specific VIs chosen, and what is the rationale behind selecting the particular cultivars for this study? Clarifications could improve the manuscript's comprehensibility and scientific rigor.

2. The manuscript presents an extensive data analysis and model validation section; however, it lacks a clear explanation of the statistical tests used for analyzing the variance among cultivars and the choice of cross-validation schemes for the prediction models. Expanding on the statistical rationale and the decision-making process behind the chosen methodologies would significantly enhance the reader's understanding and the manuscript's credibility.

3. While the study makes a comparison between traditional methods and VIs for identifying drought-tolerant cultivars, it falls short of a comprehensive literature review on this topic. A more thorough comparison with existing studies, including any conflicting results and how this study contributes to resolving those conflicts or adding new insights, would be beneficial.

The manuscript occasionally suffers from awkward phrasing and grammatical errors, which could be addressed by a thorough proofreading. Here are some examples:

Original: "For prediction model random forest outperformed linear models when it was used to predict the performance of cultivars grown in the untested crop/environment for both Gs and Chl."

Suggestion: "In the prediction model, the random forest outperformed linear models in predicting the performance of cultivars in untested crops/environments for both Gs and Chl."

Original: "Unlike yield and its components drought tolerance cannot be directly quantified but it is indirectly measured in the plant through drought-related traits such as yield crop height visual assessment physiological and biochemical traits."

Suggestion: "Unlike yield and its components, drought tolerance cannot be directly quantified; instead, it is indirectly measured in the plant through drought-related traits, including yield, crop height, visual assessment, and physiological and biochemical traits."

Original: "Identification of tolerant cultivars using prediction models showed that at least two out of three cultivars were common in measured and predicted ranking for both traits."

Suggestion: "The identification of tolerant cultivars through prediction models revealed that at least two out of three cultivars had consistent rankings in both measured and predicted outcomes for both traits."

Author Response

Responses Reviewer 1

We have revised the manuscript with the title “Assessing drought stress of sugarcane cultivars using un-manned vehicle system (UAS)-based vegetation indices and physiological parameters”. All reviewers’ comments have been addressed and your comments have been addressed as shown in the table below and in the revised manuscript.

Reviewer 2 Report

very well written. no comments

Author Response

Dear Reviewer 2,

Thank you for reviewing our manuscript.

Reviewer 3 Report

The authors identified drought-tolerant genotypes of sugarcane by using VIs compared to the physiological methods. The results showed that the same cultivars were identified as drought-tolerant cultivars when VIs and traditional method were used for identification.The methods are attractive and expected for application in other regions. I suggest the editor only minor revision is required before acceptance.

Specific comments

1. The title of the MS is “Assessing drought stress of sugarcane cultivars using un-manned vehicle system (UAS)-based vegetation indices and physiological parameters”. The logic of the conclusion is not well and it seems that description is not related with drought stress. I suggest authors rewrite this section.

2. In the 2.4.1 and 2.4.2 section, what are frequencies of Stomatal conductance (Gs), Chlorophyll content meter (Chl)? There is soil moisture profile in the figure 4. I suggest the authors may make the figures combined the soil moisture and Gs and Chl. 

Author Response

Responses Reviewer 3

We have revised the manuscript with the title “Assessing drought stress of sugarcane cultivars using un-manned vehicle system (UAS)-based vegetation indices and physiological parameters”. All reviewers’ comments have been addressed and your comments have been addressed as shown in the table below and in the revised manuscript.

Thank you for you review.

Reviewer 4 Report

The article is interesting, its subject is sugar cane, an important crop for global agriculture. The evaluated article addresses the use of vegetation indices derived from sensors mounted on UAS to identify drought-tolerant sugarcane cultivars. The purpose of the study is to investigate the effectiveness of these indices in predicting physiological parameters, such as stomatal conductance (Gs) and chlorophyll content (Chl), in different validation scenarios. The main contributions of the article include the comparison between linear and non-linear models in predicting Gs and Chl, as well as the discussion on the importance of choosing the appropriate model for different growing environments.

The strengths of the study in the proposal to improve high-throughput phenotyping in plants using high-resolution images from commercial drones as a methodology.

The manuscript is clear and the text presented is well structured. I highlight that the authors used commercial drones that are widely accessible in any country in the world, which facilitates the potential replication of the experiment.

The references cited are adequate, including recent publications in the last 5 years. No self-citations were mentioned in the document.

The experimental design proved to be appropriate for testing the hypothesis. It describes the use of machine learning algorithms to predict the performance of cultivars in different growing environments, highlighting the effectiveness of models such as random forest in specific scenarios

The results of the manuscript appear to be reproducible. I only indicate one consideration regarding a used sensor that has not been treated. See at the end of the opinion. Details about the iterations, data split, and models used are provided to enable replication of the experiments.

The figures and tables are adequate with good resolution, which helped to understand the data clearly. The images and tables are used to present the results of the Gs and Chl prediction models, highlighting the accuracy of the different algorithms in specific validation scenarios.

The conclusions are consistent with the data and arguments presented in the manuscript. The results of the prediction models are discussed in relation to drought tolerance in plants, highlighting the importance of choosing the correct model for different validation scenarios.

I make two recommendations:

1- Improve the methodology, information on data collection and information processing with SlantRange 4P, in order to guarantee the reproducibility of results. It seemed to me that there was a lack of considerations about this sensor. Was there any difference with the multispectral P4? Why was this sensor difference not evaluated?

2 - Furthermore, I suggest including the limitations of the study and possible future research directions to further enrich the work.

Author Response

Responses Reviewer 4

We have revised the manuscript with the title “Assessing drought stress of sugarcane cultivars using un-manned vehicle system (UAS)-based vegetation indices and physiological parameters”. All reviewers’ comments have been addressed and your comments have been addressed as shown in the table below and in the revised manuscript.

Thank you for your review.

Reviewer 5 Report

1.  In the abstract part, it can be simplified a little more, focusing on the new methods of drought resistance breeding of sugarcane.

2.  Picture 8 is a bit blurry, it is recommended to replace it with a clearer one.

3.  In the introduction part, the first sentence can be changed to "under the general environmental trend of global warming", in fact, drought resistance is a very important plant characteristic, under the trend of global water resources reduction, the development of water-saving irrigation technology also needs drought-resistant crops, just to say only under the conditions of global warming, a bit limited.

4.  The conclusion section of the 104-line “Moreover, even though .....” is suggested to be a separate paragraph for ease of reading.

5.  239 lines have formatting errors.

6.  Lines 541 to 552 conclude that are not detailed enough to provide a vision of the future and how the technology will perform in practical applications. Add the conclusions and data obtained in the previous section to this section as appropriate.

Minor editing

Author Response

Responses Reviewer 5

We have revised the manuscript with the title “Assessing drought stress of sugarcane cultivars using un-manned vehicle system (UAS)-based vegetation indices and physiological parameters”. All reviewers’ comments have been addressed and your comments have been addressed as shown in the table below and in the revised manuscript.

Thank you for your review.