Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study evaluates and compares three remote sensing products for monitoring white clover and perennial ryegrass ratios, and for predicting pasture productivity using spectral indices such as NDVI, SAVI, BSI, CI, and AVI, within a highland in Ecuador. The Materials section reflects the considerable effort the authors have devoted to collecting and analyzing data from various sources. However, the manuscript contains several significant flaws that need substantial revisions. Below are the shortcomings that must be addressed for the manuscript to be publishable:

* The Introduction Section lacks a literature review. It does not explicitly state the research gap this study intends to address compared to earlier research. Moreover, research objectives are absent. The structure of the introduction is also problematic, with abrupt transitions between topics, and the last paragraph reads more like a methodology description.
* As long as the research gap and study objectives remain unclear, the proposed methodology, regardless of how detailed or advanced, lacks sufficient justification.  
* Although multiple indices (NDVI, SAVI, BSI, CI, AVI) are utilized, the relationships between them and any intercorrelation analysis are not thoroughly explored to determine which index is most appropriate in this context.
* The study was conducted at a single site (the Santa Catalina Station in Ecuador) which limits the generalizability of the findings to other regions with different climatic, altitudinal, and soil conditions.
* Many paragraphs focus primarily on detailing devices and procedural steps,rather than on justifying methodological decisions. Also, the absence of a methodological flowchart further impairs clarity.
* The headings of 2.4 and 2.5. are duplicated.
* The Results section focuses heavilyon numerical differences without providing deeper analytical insight. 
* the results lack spatial variability analysis or and heatmaps. Including GIS-based maps showing the distribution of spectral indices across the site would significantly enhance the interpretation and practical value of the findings.
* results were not validated against conventional measurement methods known for their accuracy. 
* the study does not address the generalizability of its results, limiting its relevance beyond the specific study site.
* The results are not adequately interpreted or compared to those from previous studies to assess consistency or divergence. 
* The Conclusion section reads more like a summary of the results rather than a synthesis of the study’s conclusion. It should be rewritten to include a concise interpretation of the main findings and how they relate to the study’s objectives. It should also highlight the practical implications of the findings and their relevance for real-world applications. A well-developed conclusion should acknowledge the study’s limitations and provide recommendations for future research, both of which are missing here

Author Response

Dear expert reviewer,

 

As authors of the manuscript entitled “Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction”, we appreciated a lot your suggestions and comments on the document, as we are certain and convinced, that they have been useful to enrich the fluency and clarity of the entire article. We also confirm that the writing in English has been thoroughly reviewed and accordingly improved. Below, we will detail the changes made and you will be able to find them all exposed and answered here or directly within the manuscript in the edited (colored) format.

 

 

Your comments and recommendations:

This study evaluates and compares three remote sensing products for monitoring white clover and perennial ryegrass ratios, and for predicting pasture productivity using spectral indices such as NDVI, SAVI, BSI, CI, and AVI, within a highland in Ecuador. The Materials section reflects the considerable effort the authors have devoted to collecting and analyzing data from various sources. However, the manuscript contains several significant flaws that need substantial revisions. Below are the shortcomings that must be addressed for the manuscript to be publishable:

* The Introduction Section lacks a literature review. It does not explicitly state the research gap this study intends to address compared to earlier research. Moreover, research objectives are absent. The structure of the introduction is also problematic, with abrupt transitions between topics, and the last paragraph reads more like a methodology description.

 

Response:

Thank you for your valuable feedback. In response to your comments regarding the Introduction section, we have revised it to include a more comprehensive literature review on the use of remote sensing, UAVs, and vegetation indices in pasture monitoring, which provides a clearer scientific context. We have also explicitly stated the research gap, emphasizing the limited availability of studies applying these technologies in the Ecuadorian highlands, and the lack of comparative analysis between different types of spectral sensors. Furthermore, we clarified the main objective of the study in the final paragraph, which now focuses on evaluating the performance of three spectral sensors in estimating the botanical composition and yield of a grass-legume mixture. Lastly, we improved the structure and flow of the introduction by refining transitions between paragraphs to ensure a more coherent and logical progression from general context to specific research aims.

 

 

Your comments and recommendations:

* As long as the research gap and study objectives remain unclear, the proposed methodology, regardless of how detailed or advanced, lacks sufficient justification.

 

Response:

Thank you for your observation, very appreciated. We have clarified the research gap by emphasizing the limited availability of studies applying remote sensing technologies to forage monitoring in the Ecuadorian highlands, and the lack of comparative analysis between different types of spectral sensors. This gap justifies the need for our study, which aims to evaluate the performance of three sensors in estimating botanical composition and yield. We strongly believe and are very convinced that this actual revision strengthens the rationale for the proposed methodology.

 

 

Your comments and recommendations:

* Although multiple indices (NDVI, SAVI, BSI, CI, AVI) are utilized, the relationships between them and any intercorrelation analysis are not thoroughly explored to determine which index is most appropriate in this context.

 

Response:

See here from us, the almost entire discussion section, based on your suggestion:

We performed a detailed intercorrelation analysis in order to explore the relationships between vegetation indices (NDVI, SAVI, BSI and CI) calculated from the three sensors evaluated in our study (PSR-1100 spectroradiometer, MAPIR and Parrot), which revealed significant associations between them. Since most of the index data lacked to indicate a normal distribution according to the Shapiro-Wilk test (p < 0.05), we performed a Spearman correlation analysis, the results of which are presented further below.

To assess the internal consistency and reliability of vegetation indices obtained using the modified MAPIR camera versus the spectroradiometer pseudobands, a Spearman correlation analysis was performed (Figure 15, 16). The strongest association was observed for SAVI (ρ = 0.67, p < 0.001) followed by NDVI (ρ = 0.59, p < 0.001), indicating moderate to strong agreement between these indices derived from both sensors. On the other hand, BSI yielded a moderate correlation (ρ = 0.46, p = 0.009), while CI presented a negative correlation of similar magnitude (ρ = 0.43, p = 0.014). These results suggest that SAVI and NDVI are the most robust indices when comparing MAPIR camera measurements with the spectroradiometer.

 

Figure 15. Relationships between vegetation indices obtained with the MAPIR camera and pseudo-bands of the spectroradiometer using Spearman correlation.

 

Figure 16. Relationships between vegetation indices obtained with a multispectral camera (PARROT) and simulated bands of the spectroradiometer using Spearman correlation

To evaluate the intercorrelation between vegetation indices obtained with the PARROT multispectral camera and the spectroradiometer indices, a Spearman correlation analysis was performed. The resulting graphs (Figure 15, 16) indicate strong positive associations for SAVI (ρ = 0.83, p < 0.001), NDVI (ρ = 0.79, p < 0.001), and CI (ρ = 0.78, p < 0.001), while the BSI yielded a moderate correlation (ρ = 0.46, p = 0.009). These results suggest that SAVI and NDVI derived from the PARROT camera are the indices that present the greatest consistency with the reference spectroradiometer measurements, supporting their use as robust indices for this type of assessment. This comparative analysis reinforces the reliability of the indices in monitoring forage cover vigor.

 

Figure 17. Correlation (Spearman's ρ) of indices between MAPIR and PARROT sensors. The graph compares the Spearman correlation coefficients (ρ) between the vegetation indices calculated with the MAPIR and PARROt cameras, versus the same indices generated from the spectroradiometer (pseudo-bands).

 

Parrot consistently demonstrated higher correlations than MAPIR in NDVI (ρ=0.79 vs 0.59), SAVI (ρ=0.83 vs 0.67), and CI (ρ=0.78 vs -0.43). For BSI, both sensors yielded similar correlations (ρ≈0.46), with no clear advantage. For CI, the behavior was opposite between sensors, with PARROT indicating a strong positive correlation (ρ=0.78), while MAPIR had a moderate negative correlation (ρ=-0.43), suggesting differences in spectral sensitivity or in the way they respond to reflectance (Figure 17).

 

 

Your comments and recommendations:

 The study was conducted at a single site (the Santa Catalina Station in Ecuador) which limits the generalizability of the findings to other regions with different climatic, altitudinal, and soil conditions.

 

Response:

Thank you for your thought and suggestion. We acknowledge that conducting the study at a single site (Santa Catalina Station) may limit the generalizability of the findings to other regions with different environmental conditions. However, due to logistical and resource constraints, it was not feasible to replicate the study across multiple locations. Despite this limitation, we believe the results provide valuable insights into the use of spectral sensors for pasture monitoring in high-altitude tropical environments and serve as a useful reference for future studies aiming to validate or expand upon these findings in other agroecological contexts.

 

 

Your comments and recommendations:

* Many paragraphs focus primarily on detailing devices and procedural steps,rather than on justifying methodological decisions. Also, the absence of a methodological flowchart further impairs clarity.

 

Response:

We may add some details for a more detailed response, based on your comment

Experimental design (2.2 Plant material and sample design)

To evaluate the effect of progressively increasing legumes on forage yield and spectral indices, four treatments were established with increasing proportions of white clover in association with perennial ryegrass: 100:0, 90:10, 80:20, and 70:30. This gradient allows for simulating different botanical compositions commonly found in managed grasslands, analyzing their impact on productivity and plant vigor estimated using vegetation indices. The experimental design was structured as randomized complete blocks (RCBs) with three replicates, controlling for spatial variability and ensuring statistical robustness.

 

Botanical Composition and Yield (2.3 Evaluation of Variables)

Botanical composition and dry matter yield (DM ha-1) were measured before each grazing operation in four cuts (four different seasons or times), as these parameters are essential for understanding the competitive dynamics between grasses and legumes throughout the production cycle. Multi-cut evaluation captures seasonal variations (rainfall/drought) that directly influence available biomass and the spectral expression of the crop, providing a comprehensive view of the forage system.

 

Use of Spectroradiometer (2.5 Spectral Data Collection)

The use of the PSR-1100 spectroradiometer allowed detailed spectral signatures to be obtained directly from the canopy, providing high-resolution data for estimating NDVI, SAVI, BSI, and CI indices. These in situ measurements are essential as a radiometric reference for validating and comparing information obtained from UAV-mounted sensors, ensuring the reliability of the spectral models used.

 

Selection of UAVs and Cameras (2.6 capturing aerial images)

Two UAVs equipped with complementary multispectral cameras were used: the DJI Phantom with Parrot SEquioia (G, R, RE, NIR bands) and the DJI Mavic with Mapir 3w RGNIR (G, R, NIR). This selection allowed for a comparison of plant vigor and ground cover performance, which is key to selecting appropriate monitoring technologies in pastoral systems.

 

Image Processing and Index Extraction

Photogrammetric processing was performed in Pix4Dmapper to generate radiometrically calibrated orthomosaics, ensuring temporal and spatial comparability of the multispectral data. Vegetation and soil indices (NDVI, SAVI, BSI, CI) were subsequently calculated in QGIs, selected for their proven sensitivity to biophysical parameters such as chlorophyll, ground cover, and water status, critical aspects for evaluating the yield and quality of mixed grasslands.

 

Statistical Analysis

Analysis of variance (ANOVA) and Tukey's multiple comparison test at 5% were used to identify significant differences between treatments and cuts, with tests for normality and homogeneity of variance. Likewise, existing relationships were analyzed by calculating correlations between the indices obtained with the different sensors. The functional relationship between NDVI and forage yield was also explored using a linear regression model, given this index's recognized ability to estimate biomass in crops and pastures, which provides a predictive model applicable to precision agriculture and livestock practices.

 

Check also the corresponding figure of ours related to our commented response:

 

 

Your comments and recommendations:

* The headings of 2.4 and 2.5. are duplicated.

 

Response:

Thank you for bringing this error to our attention, it has been fixed.

 

 

Your comments and recommendations:

* The Results section focuses heavily on numerical differences without providing deeper analytical insight. 

 

Response:

 

We appreciate the expert reviewer’s observation regarding the need for deeper analytical insight in the Results section. In response, we have revised both the Results and Discussion sections to include a more thorough analytical interpretation of the numerical findings. Specifically, we now provide contextual explanations for the observed differences between sensors and indices, supported by relevant literature and technical considerations (e.g., spectral range, calibration procedures, and sensor configuration). These additions aim to enhance the interpretative depth of the results and clarify the implications of the numerical differences reported.

We strongly believe and are very convinced these improvements address the concern and contribute to a more comprehensive understanding of the data.

 

 

Your comments and recommendations:

* the results lack spatial variability analysis or and heatmaps. Including GIS-based maps showing the distribution of spectral indices across the site would significantly enhance the interpretation and practical value of the findings.

 

Response:

We appreciate the insightful comment. In response to the identified limitation regarding the absence of spatial variability analysis, we have incorporated GIS-based thematic maps depicting the spatial distribution of vegetation and soil indices across the study area. These additions, presented in Sections 3.5.1 and 3.5.2, provide a spatially explicit representation of the spectral indices, thereby enhancing the analytical depth and practical applicability of the results. This spatial visualization facilitates a more comprehensive understanding of intra-site heterogeneity and supports more robust interpretation and decision-making.

 

Your comments and recommendations:

* results were not validated against conventional measurement methods known for their accuracy.

 

Response:

Thank you for your observation. In this study, the validation of the UAV-based sensors was performed using a spectroradiometer, which is a well-established and highly accurate ground-based instrument for capturing spectral reflectance. This device served as a reference due to its higher spectral resolution and precision compared to UAV-mounted sensors. Therefore, while we did not use traditional destructive sampling methods as the primary validation tool, the spectroradiometer provided a robust and reliable benchmark for evaluating the performance of the UAV sensors in estimating botanical composition and yield.

 

Your comments and recommendations:

* the study does not address the generalizability of its results, limiting its relevance beyond the specific study site.

 

Response:

We acknowledge the limitation regarding generalizability. However, the study provides valuable site-specific insights that can serve as a reference for similar agroecological contexts and guide future multi-site research. Please consider this.

 

Your comments and recommendations:

* The results are not adequately interpreted or compared to those from previous studies to assess consistency or divergence. 

 

Response:

Thank you for this valuable observation. In response, we have revised the manuscript to include new references and comparisons with previous studies that have evaluated pasture yield estimation using UAV-derived vegetation indices. Specifically, we now discuss how our findings—particularly the strong correlation between NDVI and forage yield (R² = 0.8948)—are consistent with earlier research that demonstrated the effectiveness of NDVI in estimating biomass and discriminating vegetation from soil reflectance [95, 46, 48, 97]. These additions help contextualize our results within the broader body of literature and assess both the consistency and divergence of our findings relative to existing work. We believe this enhancement strengthens the interpretation of our results and addresses the reviewer’s concern.

 

Your comments and recommendations:

* The Conclusion section reads more like a summary of the results rather than a synthesis of the study’s conclusion. It should be rewritten to include a concise interpretation of the main findings and how they relate to the study’s objectives. It should also highlight the practical implications of the findings and their relevance for real-world applications. A well-developed conclusion should acknowledge the study’s limitations and provide recommendations for future research, both of which are missing here.

 

Response:

Thank you for your valuable feedback regarding the Conclusion section. In response, we have rewritten this section to go beyond a summary of results and instead provide a clear synthesis of the study’s main findings in relation to its objectives. The revised conclusion now includes a concise interpretation of the results, highlights the practical implications for pasture monitoring using remote sensing tools, and discusses the relevance of sensor selection in different agricultural contexts. We have also included a reflection on the study’s limitations and proposed directions for future research to strengthen the applicability and generalizability of the findings.

 

Once again and with all due respect, we are very thankful for your comments and corrections, which helped us to see a few unclear parts and or even faults of our side within our manuscript. With your comments we were able to smooth the text, clarify missing parts or inadequate interpretations, which resulted to a much better manuscript, than the initial version of this current study.

 

Thanks a lot on behalf of all authors (images in the attached file)

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The current study aims to compare the performance of three remote sensing products in terms of white clover and perennial ryegrass ratios. Ohau perennial ryegrass and giant Ladino white clover were tested in various quantities to measure yield and connection using vegetation and soil indices. This study lacks rigor in areas such as experimental design, data processing, and results presentation. Important revisions and tweaks are still required.

 

Principal changes

1. How were legumes and grass combined in the test plot? Are all the seeds proportionately blended and then planted in the community?
2. Do differing amounts of grass and legumes exhibit any variations in spectral information during their growth processes?
3. The paper's Sections 2.4.1 and 2.4.2, respectively, gathered data on plant and soil nutrients. However, the research did not analyze the relationship between soil nutrients, plant nutrients, and plant production; instead, it just performed a statistical analysis of the data. The author should add more pertinent information.
4. It is recommended that the author combine the information from 3.5.1 and 3.5.2 in the discussion and findings section. Additionally, the author should combine the analytical results of several indexes into a single chart for comparison analysis.

 

Additional modifications:

1. It is recommended that Figure 2 be combined with Figure 1 and that Figure 2 be eliminated.
2. Title 2.3.2 is identical to 2.3.3.
3. The value threshold range of Page12 line 321, NDVI, and SAVI is 0-1 instead of -1 to +1.

Comments on the Quality of English Language

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

Author Response

Dear expert reviewer,

 

As authors of the manuscript entitled “Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction”, we appreciated a lot your suggestions and comments on the document, as we are certain and convinced, that they have been useful to enrich the fluency and clarity of the entire article. We also confirm that the writing in English has been thoroughly reviewed and accordingly improved. Below, we will detail the changes made and you will be able to find them all exposed and answered here or directly within the manuscript in the edited (colored) format.

 

 

Your comments and recommendations:

The current study aims to compare the performance of three remote sensing products in terms of white clover and perennial ryegrass ratios. Ohau perennial ryegrass and giant Ladino white clover were tested in various quantities to measure yield and connection using vegetation and soil indices. This study lacks rigor in areas such as experimental design, data processing, and results presentation. Important revisions and tweaks are still required.

 

Principal changes

1. How were legumes and grass combined in the test plot? Are all the seeds proportionately blended and then planted in the community?

Response:

The typical forage mix in the central highlands of Ecuador is a mixture of grasses and legumes. The study used a combination of perennial ryegrass (Lolium perenne) and white clover (Trifolium repens). Field treatments were established as follows:

100% ryegrass 0% white clover

90% ryegrass 10% white clover

80% ryegrass 20% white clover

70% ryegrass 30% white clover

Seeds were mixed homogeneously before sowing.

 

1. Do differing amounts of grass and legumes exhibit any variations in spectral information during their growth processes?

Response:

Figure 19. Scatter plot or relationship between NDVI indices obtained from MAPIR and PARROT sensors and the botanical composition of perennial ryegrass and white clover.

This set of scatter plots shows the relationship between NDVI indices obtained from MAPIR, Parrot, and spectroradiometer sensors and the botanical composition expressed as the percentage of perennial ryegrass and white clover in a forage mixture.

No strong linear trends are evident, given the low coefficients of determination (R2) (less than 0.02), indicating that variation in NDVI indices explains very little of the variation in botanical composition.

In the case of NDVI_MAPIR, a negative slope is observed for perennial ryegrass and a positive slope for white clover, suggesting that in this sensor, an increase in NDVI was marginally associated with a lower percentage of ryegrass and a higher percentage of white clover, although the R2 was only 0.017, which prevents this relationship from being considered relevant from a predictive point of view.

For NDVI_Parrot, the opposite is true: slightly positive slopes are observed for perennial ryegrass and negative slopes for white clover, but also with an R2 lower than 0.01.

The calculated indices (MAPIR-R and Parrot-R) show practically flat lines, confirming the absence of a linear relationship.

These results suggest that under the conditions of the present experiment and with the forage mixture evaluated, NDVI indices derived from multispectral and RGB sensors are not sufficient on their own to predict the botanical composition between grasses and legumes.

This part has been added to the end of the results.

 

 

Your comments and recommendations:

The paper's Sections 2.4.1 and 2.4.2, respectively, gathered data on plant and soil nutrients. However, the research did not analyze the relationship between soil nutrients, plant nutrients, and plant production; instead, it just performed a statistical analysis of the data. The author should add more pertinent information.

 

Response:

Soil analysis was used as a starting point for the agronomic management of the experiment, these values allowed adjusting the fertilization level, this factor is not part of the study therefore it was not used in correlations or statistical analysis, the factor corresponded to the forage mixture and the spectral responses, the fertilization level is identical in all treatments, thus ensuring the uniformity of the experiment.

 

 

Your comments and recommendations:

It is recommended that the author combine the information from 3.5.1 and 3.5.2 in the discussion and findings section. Additionally, the author should combine the analytical results of several indexes into a single chart for comparison analysis.

 

Response:

Done

 

 

 

 

Your minor comments and recommendations as well as responses:

Additional modifications:

1. It is recommended that Figure 2 be combined with Figure 1 and that Figure 2 be eliminated.

Thank you for the suggestion. As recommended, Figure 2 has been merged with Figure 1 to streamline the presentation and avoid redundancy. Figure 2 has therefore been removed from the manuscript.

 

1. Title 2.3.2 is identical to 2.3.3.

Thank you for bringing this error to our attention, it has been fixed.

 

1. The value threshold range of Page12 line 321, NDVI, and SAVI is 0-1 instead of -1 to +1.

Thank you for bringing this error to our attention, it has been fixed.

 

Once again and with all due respect, we are very thankful for your comments and corrections, which helped us to see a few unclear parts and or even faults of our side within our manuscript. With your comments we were able to smooth the text, clarify missing parts or inadequate interpretations, which resulted to a much better manuscript, than the initial version of this current study.

 

Thanks a lot on behalf of all authors (images in the attached file)

 

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

In the study, various grass and legume associations (perennial ryegrass, Lolium perenne and white clover, Trifolium repens) were evaluated in different proportions to determine their yield and relationship through vegetation and soil indices. The following suggestions should be addressed.

1. There are too many small charts in the paper. It is recommended to combine them for display. For example, it is suggested to combine Figure 1 and Figure 2; Figure 5 is too simple, and Figures 7-10 can be combined together, etc.
2. The entire paper focuses on Evaluating Remote Sensing Products, and the abstract should emphasize the comparative results of the products.
3. How were the GCP (ground control points) selected, and what is the principle behind their selection? It is recommended to describe this in detail.
4. It is suggested to list "Discussion" as a separate chapter.
5. The conclusions in the paper are too simple. It is recommended to elaborate them point by point.

Author Response

Dear expert reviewer,

 

As authors of the manuscript entitled “Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction”, we appreciated a lot your suggestions and comments on the document, as we are certain and convinced, that they have been useful to enrich the fluency and clarity of the entire article. We also confirm that the writing in English has been thoroughly reviewed and accordingly improved. Below, we will detail the changes made and you will be able to find them all exposed and answered here or directly within the manuscript in the edited (colored) format.

 

 

Your comments and recommendations:

In the study, various grass and legume associations (perennial ryegrass, Lolium perenne and white clover, Trifolium repens) were evaluated in different proportions to determine their yield and relationship through vegetation and soil indices. The following suggestions should be addressed.

1. There are too many small charts in the paper. It is recommended to combine them for display. For example, it is suggested to combine Figure 1 and Figure 2; Figure 5 is too simple, and Figures 7-10 can be combined together, etc.

 

Response:

Thank you for the suggestion regarding the consolidation of figures to improve clarity and reduce fragmentation. In response, we have combined several of the smaller charts as recommended—for instance, Figures 1 and 2 have been merged, and Figure 5 has been revised accordingly. However, Figures 7 through 10 have been retained as separate entities due to the volume and complexity of the data presented. Each figure corresponds to a distinct subsection and provides specific insights that would be difficult to convey effectively if combined into a single visual. We believe this approach maintains clarity and supports a more precise interpretation of the results.

 

 

Your comments and recommendations:

1. The entire paper focuses on Evaluating Remote Sensing Products, and the abstract should emphasize the comparative results of the products.

 

 

Response:

Thank you for your comment. In response, we have revised the abstract to better reflect the comparative nature of the study, emphasizing the evaluation of the three remote sensing products and highlighting the key differences in their performance for estimating botanical composition and yield. We believe this change aligns the abstract more closely with the predominant aim of the given research.

 

 

Your comments and recommendations:

1. How were the GCP (ground control points) selected, and what is the principle behind their selection? It is recommended to describe this in detail.

 

Response:

We appreciate the reviewer’s suggestion regarding the clarification of the Ground Control Point (GCP) selection process. In response, we have added a detailed explanation in the methodology section of the revised manuscript. Specifically, GCPs were randomly selected within each sampling unit, ensuring that each point corresponded to an area with grass cover, in line with the study’s objectives. Spectral measurements were conducted using a spectroradiometer positioned at a height of 1 meter, which matches the spatial resolution of the imagery used. This setup ensures that the ground measurement area corresponds to the pixel size of the satellite images, thereby improving the accuracy of the calibration and validation processes.

 

 

Your comments and recommendations:

1. It is suggested to list "Discussion" as a separate chapter.

 

Response:

A wonderful suggestion which is based on our error to have missed to put the title of the obligatory subsection named “discussion”. We appreciated to point out our missing cut / mistake. As you may appreciate in the edited version of the manuscript, the adding of a real discussion, based on our results is now present. Thanks again

 

 

Your comments and recommendations:

1. The conclusions in the paper are too simple. It is recommended to elaborate them point by point.

 

Response:

We appreciate your suggestion to elaborate the conclusions in a more structured and detailed manner. Accordingly, we have reformulated the Conclusion section into a point-by-point format, addressing key aspects such as yield performance, sensor behavior, correlation analysis, practical implications, and methodological limitations. This new structure provides a clearer and more comprehensive synthesis of the study’s contributions and supports a better understanding of its relevance for both scientific and practical applications.

 

 

Once again and with all due respect, we are very thankful for your comments and corrections, which helped us to see a few unclear parts and or even faults of our side within our manuscript. With your comments we were able to smooth the text, clarify missing parts or inadequate interpretations, which resulted to a much better manuscript, than the initial version of this current study.

 

Thanks a lot, on behalf of all authors

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript "Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction" aims to assess the performance of three remote sensing products, including a spectroradiometer and two UAVs. One UAV is equipped with a MAPIR 3W RGNIR camera, and the other is equipped with a Parrot Sequoia multispectral camera, to monitor the proportions of white clover and perennial ryegrass and predict pasture yield. The study utilizes spectral data collected before and after each mowing or grazing to derive vegetation and soil indices, such as the Normalized Difference Vegetation Index, Soil-Adjusted Vegetation Index, Bare Soil Index, and Coloration Index. It validates the results through statistical analyses of forage yield and botanical composition, including analysis of variance and Tukey's test. The study shows that NDVI is highly correlated with forage yield, with a correlation coefficient R² of 0.8948. The spectroradiometer and Parrot camera yield similar index values, while there are no significant differences in yield or vegetation indices among treatments. Although the research highlights the potential of remote sensing in pasture management, it also implies the need for further optimization of sensor selection and seasonal yield prediction models.

Major Concerns

1. Inadequate Sample Size and Low Statistical Power

Only 3 replicates were set for each treatment, and G*Power simulation shows that the statistical power to detect a 20% yield difference is less than 50%, whereas the ideal power should exceed 80%. For example, the yield difference between T1 and T4 reaches 26%, but it was misjudged as "no significant difference" (p=0.9882) due to insufficient samples. This creates a severe false negative risk, potentially masking yield differences of practical significance in real-world production.

2. Uncorrected Multiple Comparisons Leading to High False Positive Risk

When analyzing over 10 spectral indices, the study did not apply Bonferroni or FDR correction methods. If the significance level is set at 0.05, the false positive probability exceeds 40%, making conclusions like "significant differences in BSI index between Parrot and Mapir" likely result from random errors rather than real sensor performance disparities. This severely undermines the credibility of sensor recommendation conclusions.

3. Logical Contradictions in Sensor Comparison Design Questioning Rationality

In a study aimed at "evaluating accuracy differences among remote sensing products," the inclusion of a Mapir camera known to lack spectral bands contradicts the original intention of "screening high-quality products." This design not only risks confusing "accuracy differences due to hardware band defects" with "actual product performance differences," but also may lead to conclusion biases due to insufficient variable control. Additionally, it wastes research resources and misleads technical selection decisions in practical applications, arousing significant controversy about the rationality of the methodological design..

 

Minor Concerns

1. Figure

The issue of chart specification deficiency exists at line 394, where "Figure 10." only indicates the number without including descriptive text reflecting the specific content of the chart.

The text layout of Figure 6 at line 313 is irregular, showing problems such as unaligned labels, inconsistent spacing, or misplaced legends, which do not conform to academic chart formatting standards.

Author Response

Dear expert reviewer,

 

As authors of the manuscript entitled “Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction”, we appreciated a lot your suggestions and comments on the document, as we are certain and convinced, that they have been useful to enrich the fluency and clarity of the entire article. We also confirm that the writing in English has been thoroughly reviewed and accordingly improved. Below, we will detail the changes made and you will be able to find them all exposed and answered here or directly within the manuscript in the edited (colored) format.

 

 

Your comments and recommendations:

The manuscript "Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction" aims to assess the performance of three remote sensing products, including a spectroradiometer and two UAVs. One UAV is equipped with a MAPIR 3W RGNIR camera, and the other is equipped with a Parrot Sequoia multispectral camera, to monitor the proportions of white clover and perennial ryegrass and predict pasture yield. The study utilizes spectral data collected before and after each mowing or grazing to derive vegetation and soil indices, such as the Normalized Difference Vegetation Index, Soil-Adjusted Vegetation Index, Bare Soil Index, and Coloration Index. It validates the results through statistical analyses of forage yield and botanical composition, including analysis of variance and Tukey's test. The study shows that NDVI is highly correlated with forage yield, with a correlation coefficient R² of 0.8948. The spectroradiometer and Parrot camera yield similar index values, while there are no significant differences in yield or vegetation indices among treatments. Although the research highlights the potential of remote sensing in pasture management, it also implies the need for further optimization of sensor selection and seasonal yield prediction models.

Major Concerns

1. Inadequate Sample Size and Low Statistical Power

Only 3 replicates were set for each treatment, and G*Power simulation shows that the statistical power to detect a 20% yield difference is less than 50%, whereas the ideal power should exceed 80%. For example, the yield difference between T1 and T4 reaches 26%, but it was misjudged as "no significant difference" (p=0.9882) due to insufficient samples. This creates a severe false negative risk, potentially masking yield differences of practical significance in real-world production.

 

Response:

We appreciate the expert reviewer’s comment regarding the limited number of replicates and its impact on statistical power. We confirm that a G*Power analysis demonstrates that, with only three replicates per treatment, the power to detect a 20% yield difference is indeed below 50%, increasing the risk of Type II errors. However, using three replicates is an extremely common practice in agronomic field trials, especially under real-world production conditions where resources are often constrained. Additionally, prior to the experiment, we lacked sufficient information on population means and standard deviations to accurately estimate the expected effect size and determine the optimal sample size. We acknowledge this limitation and will consider increasing replication in future studies to enhance statistical robustness.

 

 

Your comments and recommendations:

1. Uncorrected Multiple Comparisons Leading to High False Positive Risk

When analyzing over 10 spectral indices, the study did not apply Bonferroni or FDR correction methods. If the significance level is set at 0.05, the false positive probability exceeds 40%, making conclusions like "significant differences in BSI index between Parrot and Mapir" likely result from random errors rather than real sensor performance disparities. This severely undermines the credibility of sensor recommendation conclusions.

 

Response:

We thank the expert reviewer for pointing out the issue of multiple comparisons. In our study, we compared three sensors across four vegetation indices, resulting in 12 pairwise tests. While we recognize that this number of comparisons increases the risk of Type I errors, we note that the p-values for the significant results were extremely low (e.g., < 0.000001). Even under a conservative Bonferroni correction, these values would remain well below the adjusted significance threshold. Therefore, although we did not apply formal correction methods, we believe the conclusions drawn from these results are statistically robust. We appreciate the reviewer’s suggestion and will take this into account in future studies involving a larger number of comparisons.

 

Your comments and recommendations:

1. Logical Contradictions in Sensor Comparison Design Questioning Rationality

In a study aimed at "evaluating accuracy differences among remote sensing products," the inclusion of a Mapir camera known to lack spectral bands contradicts the original intention of "screening high-quality products." This design not only risks confusing "accuracy differences due to hardware band defects" with "actual product performance differences," but also may lead to conclusion biases due to insufficient variable control. Additionally, it wastes research resources and misleads technical selection decisions in practical applications, arousing significant controversy about the rationality of the methodological design.

 

Response:

We appreciate the expert reviewer’s critical perspective and the opportunity to clarify the rationale behind the inclusion of the MAPIR camera in our study. The objective of this research was not solely to compare high-end remote sensing products, but rather to evaluate the potential of both low-cost and high-performance sensors (including UAV-mounted cameras and field spectroradiometers) for estimating pasture productivity through vegetation and soil indices.

The inclusion of the MAPIR RGN camera was intentional and aligned with a key aim of the study, being to assess whether affordable, accessible technologies can provide sufficiently accurate data for practical agricultural monitoring, especially in resource-limited contexts such as many areas of the Andean region. We acknowledge that the MAPIR camera has spectral limitations. However, this was precisely part of the evaluation , meaning to determine how such limitations affect index performance and whether they can still yield usable information for decision-making.

Rather than introducing bias, this design allowed us to highlight the trade-offs between sensor cost, spectral resolution, and data reliability. We agree that this distinction could have been more clearly articulated in the original manuscript, and we have now revised the Introduction and Discussion sections to better reflect this methodological rationale and to avoid any misinterpretation regarding the study’s scope and intent.

 

 

Minor Concerns

1. Figure

The issue of chart specification deficiency exists at line 394, where "Figure 10." only indicates the number without including descriptive text reflecting the specific content of the chart.

 

Solved

 

The text layout of Figure 6 at line 313 is irregular, showing problems such as unaligned labels, inconsistent spacing, or misplaced legends, which do not conform to academic chart formatting standards.

 

Solved

 

Once again and with all due respect, we are very thankful for your comments and corrections, which helped us to see a few unclear parts and or even faults of our side within our manuscript. With your comments we were able to smooth the text, clarify missing parts or inadequate interpretations, which resulted to a much better manuscript, than the initial version of this current study.

 

Thanks a lot on behalf of all authors

Author Response File: Author Response.docx

Reviewer 5 Report

Comments and Suggestions for Authors

This study investigates the relationship between vegetation and soil indices-calculated from data acquired using the Spectral Evolution PSR-1100 spectroradiometer, MAPIR 3W RGNIR camera, and Parrot Sequoia multispectral camera-and the yield of perennial ryegrass and white clover at various mixture ratios.
Significant differences in index values were observed between sensors, suggesting that the findings provide valuable information for selecting appropriate sensors in agricultural applications. However, several revisions appear necessary, as outlined below:

1. Introduction
The rationale for selecting perennial ryegrass and white clover is not explained.

Although representative vegetation indices are mentioned, there is no review of which wavelengths of reflectance have traditionally been used for yield prediction, nor is there any explanation of the mechanisms through which favorable results have been achieved.

2. Materials and Methods
Please provide information on the time of day when the UAV flights were conducted and the solar elevation at the time. Additionally, clarify what kind of corrections were applied to the acquired data.

3. Results and Discussion
In this journal, the Results and Discussion sections must be presented separately.

While differences in calculated values between sensors are shown for each index, a discussion of the potential causes of these differences should be added. Were similar discrepancies observed in previous studies?

Please improve the resolution of the figures.

Regarding the regression lines, coefficients, intercepts, and R2 values are reported to four decimal places-do the data truly support this level of precision?

Author Response

Dear expert reviewer,

 

As authors of the manuscript entitled “Evaluating Remote Sensing Products for Pasture Composition and Yield Prediction”, we appreciated a lot your suggestions and comments on the document, as we are certain and convinced, that they have been useful to enrich the fluency and clarity of the entire article. We also confirm that the writing in English has been thoroughly reviewed and accordingly improved. Below, we will detail the changes made and you will be able to find them all exposed and answered here or directly within the manuscript in the edited (colored) format.

 

Your comments and recommendations:

This study investigates the relationship between vegetation and soil indices-calculated from data acquired using the Spectral Evolution PSR-1100 spectroradiometer, MAPIR 3W RGNIR camera, and Parrot Sequoia multispectral camera-and the yield of perennial ryegrass and white clover at various mixture ratios.
Significant differences in index values were observed between sensors, suggesting that the findings provide valuable information for selecting appropriate sensors in agricultural applications. However, several revisions appear necessary, as outlined below:

1. Introduction

The rationale for selecting perennial ryegrass and white clover is not explained.

Response:

We thank the expert reviewer for this important observation. We have now included a justification for the selection of perennial ryegrass and white clover in the revised manuscript. These species were chosen because they represent the typical forage mixture used in the central highlands of Ecuador, where this study was conducted. Perennial ryegrass (Lolium perenne) is widely used due to its high adaptability to local soil and climatic conditions, making it a reliable and productive grass species in this region. White clover (Trifolium repens), on the other hand, was included in the mixture to ensure uniform ground coverage. Its rhizomatous growth habit allows it to spread evenly without acting as an invasive competitor to ryegrass. This balance supports sustainable pasture development and reflects real-world agronomic practices in the study area.

 

Although representative vegetation indices are mentioned, there is no review of which wavelengths of reflectance have traditionally been used for yield prediction, nor is there any explanation of the mechanisms through which favorable results have been achieved.

 

Response:

Thank you for these valuable thoughts and observation. While the manuscript already included a general description of the vegetation and soil indices used (NDVI, SAVI, BSI, and CI), we have now expanded the introduction to include a more detailed explanation of the specific reflectance wavelengths involved in each index and the physiological mechanisms that justify their use in yield prediction. The following paragraph has been added to the Introduction section: “These indices are derived from reflectance values in specific regions of the electromagnetic spectrum, primarily the red, near-infrared (NIR), and shortwave infrared (SWIR) bands. For example, NDVI exploits the strong absorption of red light by chlorophyll and the high reflectance of NIR by healthy vegetation, making it a reliable indicator of plant vigor. SAVI incorporates a soil brightness correction factor to improve sensitivity in areas with sparse vegetation. BSI combines visible and SWIR bands to detect bare soil exposure, while CI uses red-edge and NIR reflectance to estimate chlorophyll content, which is directly related to photosynthetic activity and yield potential. These spectral mechanisms explain why such indices are widely used in yield prediction and crop monitoring.”

 

Your comments and recommendations:

2. Materials and Methods

Please provide information on the time of day when the UAV flights were conducted and the solar elevation at the time. Additionally, clarify what kind of corrections were applied to the acquired data. Thanks for this obsevation.

 

Response:

Thank you for pointing out the omission of this important information. We have now included the following details in the Methodology section of the revised manuscript: "UAV flights and spectroradiometer measurements were conducted between 09:00 and 13:00 local time, under stable lighting conditions and clear skies, to minimize the effects of changing solar angles. All data were collected on the same day for each sampling campaign to ensure consistency.

Radiometric calibration of the imagery was performed using reflectance calibration targets for both the MAPIR and Parrot Sequoia cameras. Prior to each flight, images of the calibration panels were captured and later processed using the corresponding software tools. These tools apply radiometric corrections by adjusting the image brightness levels based on the known reflectance values of the targets, thereby standardizing the data and compensating for variations in ambient light conditions."

 

Your comments and recommendations:

3. Results and Discussion

In this journal, the Results and Discussion sections must be presented separately.

Response:

A wonderful suggestion which is based on our error to have missed to put the title of the obligatory subsection named “discussion”. We appreciated to point out our missing cut / mistake. As you may appreciate in the edited version of the manuscript, the adding of a real discussion, based on our results is now present. Thanks again

 

Your comments and recommendations:

While differences in calculated values between sensors are shown for each index, a discussion of the potential causes of these differences should be added. Were similar discrepancies observed in previous studies?

 

Response:

While differences in calculated values between sensors are indeed presented for each index, we acknowledge the need to elaborate on the potential causes of these discrepancies. These differences are primarily attributed to the technical limitations and configurations of the sensors. As discussed in the conclusions, the MAPIR sensor exhibited lower performance for soil indices, likely due to its limited spectral range, non-standard band configuration, and the low resolution of its modified lens. These factors can significantly affect the accuracy of reflectance measurements, particularly in indices that are sensitive to subtle spectral variations such as BSI and CI.

Additionally, differences in calibration targets and procedures used for each sensor may have contributed to the observed discrepancies. Each sensor followed slightly different protocols for reflectance calibration, which can introduce variability in the resulting index values. This issue has also been noted in previous studies, where sensor-specific characteristics and calibration methods led to inconsistencies in vegetation and soil index outputs, even under similar environmental conditions.

We have now clarified these points in the discussion to provide a more comprehensive interpretation of the sensor-related differences.

 

Your comments and recommendations:

Please improve the resolution of the figures.

Solved

 

Your comments and recommendations:

Regarding the regression lines, coefficients, intercepts, and R2 values are reported to four decimal places-do the data truly support this level of precision?

 

Response:

We appreciate the expert reviewer’s comment regarding the number of decimal places reported for the regression coefficients, intercepts, and R² values. We have revised these values to display two decimal places, which is more consistent with standard reporting practices in applied research. This level of precision more accurately reflects the resolution and reliability of the original data, and avoids the impression of overprecision, which could be misleading. It also enhances the clarity and readability of the results without compromising their statistical validity or interpretability.

 

Once again and with all due respect, we are very thankful for your comments and corrections, which helped us to see a few unclear parts and or even faults of our side within our manuscript. With your comments we were able to smooth the text, clarify missing parts or inadequate interpretations, which resulted to a much better manuscript, than the initial version of this current study.

 

Thanks a lot, on behalf of all authors

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I thank the authors for their successful handling of all the comments. I believe the manuscript is now ready for publication and expect it to have a positive impact

Author Response

We sincerely thank the reviewer for their positive feedback and thoughtful evaluation. We are pleased to know that the revisions have addressed the concerns effectively and that the manuscript is now considered ready for publication. We truly appreciate the encouraging words and the time dedicated to reviewing our work.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript has undergone careful revision and refinement by the author. The details have been meticulously scrutinized and considered, resulting in greatly enhanced scientific reliability. Moreover, the research findings are presented in a more intuitive and systematic manner, making it easier for readers to understand and accept the viewpoints and conclusions. In terms of formatting and expression, the manuscript has been optimized to ensure clarity and organization. I believe that this paper now meets the requirements for publication in the journal.

Author Response

We sincerely thank the reviewer for their positive feedback and thoughtful evaluation. We are pleased to know that the revisions have addressed the concerns effectively and that the manuscript is now considered ready for publication. We truly appreciate the encouraging words and the time dedicated to reviewing our work.

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript has been revised according to the previous suggestions, but there are still some minor issues that need further improvement:

1. Figure 19: The axes (both X and Y) should be bolded for better clarity.

2. Conclusion section: The current numbering (1, 2, 3) may be confused with first-level headings. It is recommended to use parentheses (e.g., 1), 2), 3)) instead.

3. Table 7 (Note): "table6" should be capitalized as "Table 6" for consistency.

4. Figure 9: The figure caption is missing and should be added.

5. Figure consolidation: The number of figures is somewhat excessive. Consider merging Figures 6–9 into a single composite figure for better readability and conciseness.

Author Response

Thank you for the valuable feedback. We have carefully addressed each of the suggested revisions as follows:

1. Figure 19: The labels for both the X and Y axes have been bolded to enhance clarity and visual readability.

2. Conclusion Section: The numbering format has been updated to use parentheses (i.e., 1), 2), 3)) to avoid confusion with first-level headings and improve structural consistency.

3. Table 7 (Note): The reference to "table6" has been corrected to "Table 6" to maintain consistency in capitalization throughout the manuscript. All table and figure references have been reviewed and standardized accordingly.

4. Figure 9: The caption for Figure 9 has been removed, as the figure has now been integrated into a newly created composite figure (Figure 6), following the suggestion to consolidate Figures 6–9.

5. Figure Consolidation: A new composite figure (Figure 6) has been created to replace the original Figures 6, 7, 8, and 9. The remaining figures have been renumbered accordingly to reflect this change and ensure coherence throughout the manuscript.

We appreciate the constructive comments and believe these revisions contribute to a clearer and more concise presentation of the work.

Reviewer 5 Report

Comments and Suggestions for Authors

There are no additional requests.

Author Response

We sincerely thank the reviewer for their positive feedback and thoughtful evaluation. We are pleased to know that the revisions have addressed the concerns effectively and that the manuscript is now considered ready for publication. We truly appreciate the encouraging words and the time dedicated to reviewing our work.