Review Reports - UAV-Based Spectral and Thermal Indices in Precision Viticulture: A Review of NDVI, NDRE, SAVI, GNDVI, and CWSI

Round 1

Reviewer 1 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The article titled “UAV-based spectral and thermal vegetation indices for precision viticulture: A Review on NDVI, NDRE, SAVI, GNDVI, and CWSI” is excellently aligned with the scope of the Journal Agronomy. It is a comprehensive, current, and relevant literature review focused on the use of images obtained by UAVs (Unmanned Aerial Vehicles) in precision viticulture. The manuscript is rich in information, well-structured, and pleasant to read. The authors demonstrate a strong command of the content, deepening the discussions with high technical and analytical quality. The work effectively synthesizes and highlights several impactful studies in the field, resulting in an engaging text that is valuable to the scientific community.

Specific comments: (?) = doubt

11... as NDVI (?), NDRE (?), GNDVI (?), and SAVI (?) a... Give meaning of the acronyms.

18... across varieties and seasons (Seasons of the year or phenological cycles?)

256... GNDVI, CWSI, ExGR), and their practical appli... What is ExGR?

326-327 ...research include [insert model, spectral bands, resolution, and field of view] (?).... Please review the excerpt!

328... as Pix4Dmapper (version and reference?), Agisoft Metashape (Version and reference?)....

357-358... SegNet (?), U-Net (?), and ModSegNet (?)... Meaning of the abbreviations?

425...ple, Matese A (?) et al,. [81] reported... Is A a typo?

519-526... In practical applications, NDRE values in this study (?)... Is paragraph 519-526 not a continuation of the work of Marty et al. [101]? If so, please join the paragraphs to avoid ambiguity!

530 ... tral and physiological indicators (?). Add a period after indicators.

676...(?)[105] applied artificial neural networks (ANN) and generalized linear... (?) Cite the authors of the study.

772... cultivate interactions. [13,60]. Remove a period.

806... while (?) [85] showed that... Cite the author!

900 ...have been recommended [20,151]. [33,150] (?)

Author Response

Specific comments: (?) = doubt

Comment 1#

11... as NDVI (?), NDRE (?), GNDVI (?), and SAVI (?) a... Give meaning of the acronyms.

Response 1#

Thank you for your comment. Due to the 200-word limit set by Agronomy for the Abstract (our version contains 191 words), we were unable to expand the full definitions of the indices (NDVI, NDRE, GNDVI, and SAVI) within this section. However, all acronyms are fully defined and described in the Introduction, where they are presented with their complete names and context.

Comment 2#

18... across varieties and seasons (Seasons of the year or phenological cycles?)

Response 2#

Thank you for the clarification request. We have modified the sentence to specify “across varieties and phenological cycles” to better reflect the intended meaning. This change is now reflected in line 18 of the revised manuscript.

Comment 3#

256... GNDVI, CWSI, ExGR), and their practical appli... What is ExGR?

Response 3#

Thank you for noticing this. The term “ExGR” was inadvertently included and has now been removed from the manuscript, as it is not addressed in the review. The list now only includes the vegetation indices analyzed in depth: NDVI, NDRE, SAVI, GNDVI, and CWSI. See line 262

Comment 4#

326-327 ...research include [insert model, spectral bands, resolution, and field of view (?).... Please review the excerpt!

Response 4#

Thank you for pointing this out. The placeholder has been replaced with specific UAV sensor details, including typical models (e.g., MicaSense RedEdge-M, Parrot Sequoia, DJI Phantom 4 Pro), their spectral bands, resolution, and field of view. These updates are now included in lines 336–340.

Comment 5#

328... as Pix4Dmapper (version and reference?), Agisoft Metashape (Version and reference?)

Response 5#

Thank you for your suggestion. We have added both the software versions and appropriate references. The updated text now includes: Pix4Dmapper (v.4.5.6, Pix4D S.A., Lausanne, Switzerland) and Agisoft Metashape Professional (v.2.0.4, Agisoft LLC, St. Petersburg, Russia), along with their official websites. Please see the updated text in Line 342 of the revised manuscript.

Comment 6#

357-358... SegNet (?), U-Net (?), and ModSegNet (?)... Meaning of the abbreviations

Response 6#

Thank you for your comment. These terms have now been defined in the manuscript as follows: Segmentation Network (SegNet), U-shaped Convolutional Network (U-Net), and Modified Segmentation Network (ModSegNet). These changes appear in lines 373–375.

Comment 7#

425...ple, Matese A (?) et al,. [81] reported... Is A a typo?

Response 7#

Yes, thank you for catching that. The initial “A.” was a typographical mistake and has been removed. The correct citation now reads “Matese et al. [91]”. Please see the revised version in Line 442

Comment 8#

519-526... In practical applications, NDRE values in this study (?)... Is paragraph 519-526 not a continuation of the work of Marty et al. [101]? If so, please join the paragraphs to avoid ambiguity!

Response 8#

Thank you for your valuable observation. The paragraph indeed refers to the same study by Marty et al. [107] (updated reference number). To avoid ambiguity, we have merged the paragraphs and clearly indicated that the NDRE values and UAV specifications correspond to this study. Please see the revised version in Lines 531–539

Comment 9#

530 ... tral and physiological indicators (?). Add a period after indicators.

Response 9#

Thank you for the correction. We have added the missing period after the word “indicators.” This update is reflected in line 543.

Comment 10#

676...(?)[105] applied artificial neural networks (ANN) and generalized linear... (?) Cite the authors of the study.

Response 10#

Thank you for your observation. The appropriate citation has now been added as “Romero et al. [84]” in line 689 of the revised manuscript.

Comment 11#

772... cultivate interactions. [13,60]. Remove a period.

Response 11#

Thank you. The extra period has been removed to correct the punctuation. Please see line 786.

Comment 12#

806... while (?) [85] showed that... Cite the author!

Response 12#

Thank you for your comment. We have added the corresponding author, and the text now reads: “while Dorin et al. [85] showed that…”. Please refer to line 819.

Comment 13#

900 ...have been recommended [20,151]. [33,150] (?)

Response 13#

Thank you for pointing this out. There was indeed a mistake in the reference numbering. It has now been corrected to: “… have been recommended [33,152].” as reflected in line 912.

Reviewer 2 Report (Previous Reviewer 4)

Comments and Suggestions for Authors

In this work, we demonstrate that unmanned aerial vehicles (UAVs) with multispectral sensors are transforming precision viticulture, enabling detailed monitoring of vineyard variability. Fundamental to this objective is the use of vegetation indices such as NDVI, NDRE, GNDVI, and SAVI, which are widely applied to estimate vine vigor, canopy structure, and water status. Additionally, UAV-derived indices can provide information on grape composition, including sugar content (°Brix), total phenolics, anthocyanins, titratable acidity, berry weight, and yield variables measurable in the field or laboratory to validate spectral predictions. The research progress in this area of research is notable, and this review work in the case of grapevines is necessary. However, proper publication of this paper requires some adjustments:
Introduction
The paragraphs are very long, making it difficult to read, understand, and interpret. It is recommended that it be rewritten with greater clarity.
It is suggested to establish a clearer structure in the introduction. It would be advisable to use a text based on vegetation indices for plant vigor and then on the derivatives that allow for the composition of the grape as a fruit.
Methodology of Literature Selection
It is suggested to incorporate a methodological outline; this will allow for a more precise visualization of the methodology used.
Overall, I consider this review work to be a good contribution to the study of precision agriculture for grapevines.

Author Response

Comments 1#

Introduction
The paragraphs are very long, making it difficult to read, understand, and interpret. It is recommended that it be rewritten with greater clarity.

It is suggested to establish a clearer structure in the introduction. It would be advisable to use a text based on vegetation indices for plant vigor and then on the derivatives that allow for the composition of the grape as a fruit.

Response 1#

Thank you very much for your constructive suggestion. The Introduction has been thoroughly revised to improve clarity, structure, and readability. Long paragraphs have been divided into shorter, thematic sections. The new structure introduces vegetation indices in relation to vine vigor and subsequently addresses their link to grape and wine composition. Additionally, we have refined the manuscript’s overall structure into the following sections: (1) Introduction, (2) Methodology of Literature Selection, (3) UAV Data Acquisition and Processing, (4) Vegetation and Thermal Indices in Precision Viticulture, (5) Applications of UAV-Based Indices, (6) Limitations, and (7) Future Projections. All changes have been highlighted in the revised version.

Comments 2#

Methodology of Literature Selection.

It is suggested to incorporate a methodological outline; this will allow for a more precise visualization of the methodology used.

Overall, I consider this review work to be a good contribution to the study of precision agriculture for grapevines.

Response 2#

Thank you for the excellent suggestion. We have included a new Figure 3 titled “Methodological workflow for literature selection and analysis in this review,” located at line 239 of the revised manuscript. This figure summarizes the main steps of the methodology to improve transparency and reproducibility.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript is more like a popular science reading material. It lacks in-depth analysis of the application of drone remote sensing technology in precision grape cultivation. The following issues needs to improve farther:
1. Establishing statistical models based on spectral indices to derive vegetation parameters is a common approach. The authors merely provide a brief introduction to these indices without explaining the underlying mechanisms for estimating vegetation parameters or outlining future development directions.
2. There is no description of using mechanistic models or hybrid models to estimate grape parameters in mansucript.
3. The discussion is overly simplistic and lacks distinctive insights.
4. The overall organization of the article is rather disorganized.

Author Response

Reviewer # 1

Comments # 1

Establishing statistical models based on spectral indices to derive vegetation parameters is a common approach. The authors merely provide a brief introduction to these indices without explaining the underlying mechanisms for estimating vegetation parameters or outlining future development directions.

Response #1

Dear Reviewer, thank you for your comment. We would like to clarify that the manuscript now explicitly addresses both the underlying mechanisms of spectral indices and future development directions. Specifically, lines 125–130 discuss statistical models based on vegetation indices and how they enable the estimation of key vine parameters, including vigor, pigment content, water relations, and yield potential. Furthermore, lines 133–136 outline future directions, highlighting the integration of spectral indices with advanced statistical and machine learning methods, as well as hybrid frameworks combining multispectral, thermal, and structural data to improve predictive accuracy and support dynamic vineyard monitoring.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comments # 2

There is no description of using mechanistic models or hybrid models to estimate grape parameters in manuscript.

Response #2

Dear Reviewer, thank you for your comment. We have added a paragraph in section 3.3.3 (lines 863–876) that explicitly addresses the use of mechanistic and hybrid models to estimate grapevine physiological parameters and ripening dynamics. This paragraph explains how mechanistic models simulate underlying biophysical and biochemical processes, while hybrid models combine UAV-derived spectral, thermal, and structural data with mechanistic or machine learning frameworks. This addition provides more detail on modeling approaches beyond empirical vegetation indices and clarifies their potential for dynamic, spatially explicit vineyard monitoring.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comments # 3

3.The discussion is overly simplistic and lacks distinctive insights.

Response #3

Dear Reviewer, thank you for this comment regarding the distinctiveness of our discussion. We appreciate the opportunity to clarify the novel contribution of our review. While the vegetation indices we focus on (NDVI, NDRE, SAVI, GNDVI, CWSI) are indeed well-established, our review provides a distinctive insight by systematically synthesizing and critically evaluating their application specifically for predicting grape composition and wine quality parameters a critical link that has not been thoroughly explored in previous review articles. Many existing reviews (e.g., Sassu et al., 2021; Giovos et al., 2021; Tardaguila et al., 2021, as cited in our introduction) concentrate on general technological applications, vigor zoning, or stress detection. The novelty of our work lies in bridging the gap between remote sensing technology and practical enological outcomes. We critically assess how these common indices can serve as non-destructive proxies for key quality markers like sugar content (°Brix), anthocyanins, phenolics, and acidity, thereby moving directly into the realm of precision enology. This focused synthesis provides a pragmatic and actionable framework for researchers and viticulturists, demonstrating how accessible technology can be leveraged to make informed decisions that ultimately impact wine quality.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comments # 4

The overall organization of the article is rather disorganized.

Response #4

Dear Reviewer, thank you for this valuable feedback regarding the article's organization. We have carefully reviewed the manuscript's structure in response to your comment and those from the other reviewers.

Significant efforts have been made to improve the logical flow and coherence throughout the manuscript. These changes include restructuring several sections for better clarity, ensuring a more seamless transition between ideas, and enhancing the overall narrative to guide the reader more effectively from the introduction to the conclusions.

We believe these revisions have substantially improved the organization and readability of the article, and we thank you for your suggestion which has strengthened the manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

The article entitled “UAV-based spectral and thermal vegetation indices for precision viticulture: A Review on NDVI, NDRE, SAVI, GNDVI, and CWSI” is well aligned with the scope of the Journal Agronomy. It presents a current and relevant literature review on the use of drone imagery in precision viticulture. The manuscript is comprehensive and rich in information; however, some aspects require improvement or further elaboration. I recommend including a dedicated section on grapevine phenology, which would help to more accurately contextualize the application of vegetation indices throughout the crop cycle. I also suggest improving Figure 1, as the y-axis represents the reflectance factor of a healthy green leaf rather than a canopy, and the unit of measurement is incorrectly indicated as a percentage (%), which may confuse readers. In the case of canopy analysis, it would be necessary to specify the number of layers, as exemplified by Myers (1970) (reference below). Additionally, it would be important to indicate whether the coefficient of determination (R²) reported in each cited article is statistically significant, specifying the confidence level used. I recommend a careful review of the citations to ensure they are correctly matched to the corresponding references. The formula for the SAVI index, as currently presented, contains errors and should be corrected to reflect its proper definition. I also suggest consulting and including classic and foundational references in the field to strengthen the theoretical basis of the article. Further specific comments follow below to assist in improving the manuscript.

Specific comments: L = Line; ( ?) = doubt

L 15. ... indices can also inform grape composition (?)... Give examples of types of grape composition that can be detected by the indices?

L 18. ... attributes such as soluble solids (Which ones?), phenolic content (Which ones?), acidity (Which ones?), and other quality-related parameters (Which ones?). Can these variables be measured to validate the indices? Explain.

L 20. ... and discuss their strengths (Which ones?), limitations (Which ones?), and integration (Which ones?) with... Please give examples of the strengths, limitations, and types of integrations.

L 21-22. ... abiotic stress (Which ones?) from biotic factors (Which ones?), the saturation of indices in dense canopies (Which one is the crop?), and the need for robust calibration across varieties (Which ones?) and seasons (?)... Please improve the text with examples! L 23. Finally, we outline future directions (?) ... Please give an example.

L 27. Do not repeat words from the title. Use words related to your study. Replace NDVI, UAV, precision agriculture. Suggestions: Multispectral images, wine quality, spectral indices, etc.

L 30. ... most economically significant (?) perennial... The word 'significant' has a double meaning, as it conveys the idea of a level of significance. Please replace it with another word (e.g., crucial or imperative).

L 34. ... due to their long-standing viticultural traditions (?) ... Sorry, I don't know! Give an example of viticultural traditions.

L 35. ... the second-largest wine exporter by volume (?) and the... ? Write the value of the volume of wine exported.

L 37. Viticulture is practiced across a wide range of climatic regions (?) and soil types (?),... What are these regions and soil types?

L 38. ...yet grapevines are highly sensitive to biotic (?) and (?) abiotic stress factors... Specify the biotic and abiotic factors.

L 41-42. ...excessive humidity (How much?) and inadequate irrigation management [5]. Give examples of inadequate management.

L 43. ...accelerate phenology (What stage? All phenological cycles?), alter berry composition (How?)...

L 44. ...which in turn amplifies pest pressure (?)[6]... What type of pests? List them.

L 45-46. ...equipped with multispectral sensors (Only multispectral? Thermal ones? Give examples).

...as they enable precise monitoring (What is the value of this precision?) ...

L 47.... optimization of water resources (Give an example?), and early detection of stress symptoms (How?) ...

L 48... viticulture[7]... remove one point, there are two.

L 49. Precision agriculture technologies (Which ones?) have become valuable...

L 51. ...multispectral data... And don't thermal sensor data deserve to be highlighted as well? L 52-53 ...capture ultra-high-resolution imagery (I don't see the need to use the word ultra-high. Just high-resolution would suffice, wouldn't it? What is the spatial resolution? mm? cm?) provides a high-throughput alternative to traditional monitoring methods (How? List them.)

L 54. Compared to satellite platforms (Which satellites?)

63-64 ...by pigment concentration (Which ones?), water content, and cell anatomy (What anoyomia is this?), as well as by soil properties (Which ones?) and background effects (Which ones? Explain.).

L 65. ...vegetation typically displays low... What vegetation? Under what conditions?

L 69 ...design (Which design?) and interpretation of vegetation indices, and requires careful consideration (Which ones?) of soil...

L 71. Improve Figure 1. The y-axis title is Reflectance Factor with no unit of measurement. If the reflectance factor is in percentage, change the scale to 0 to 80%. The spectral signatures are for a healthy, green leaf and for a stressed leaf (What is the percentage of water in the leaf?). For vegetation or a vegetative canopy, which can be homogeneous or heterogeneous, you must specify: The name of the vegetation? The number of leaf layers?

L 77. ...which capture reflectance (?) in selected regions of the spectrum from visible to near-infrared... Wouldn't this capture be spectral radiance? Spectral radiance is the amount of electromagnetic energy reflected or emitted by the object in different spectral bands that reaches the sensor per unit area and per solid angle. Spectral reflectance is an intrinsic property of the target that represents the fraction of incident radiation that is reflected by the target. It is obtained after processing the radiance data.

L 78. Sensor performance is defined by spatial and spectral resolution. And isn't radiometric resolution important?

L 80-81 ... such as individual leaves, disease symptoms (Which diseases and which symptoms? List them.)

L 100 ... facilitating site-specific management and harvest strategies (The reference is missing, isn't it?)

L 102-103 ... with distinct physicochemical properties (Which ones? List them) at harvest.

L 109. ... related indices such as RDVI (?), OSAVI (?), and NDRE (?) have... The acronym for these indices appears for the third time without the meaning. Please write the meaning.

L 113 ... Unlike many other fruits (Which ones?),

L 116. Grape maturity at harvest plays a decisive role in determining wine quality (?). Explain why grape maturity at harvest determines wine quality.

L 123 ... and aromatic precursors (Which ones?), thereby linking...

L 130 ... spectral indices with wine quality attributes (?), What are these attributes?

L 143 ... using major databases including Scopus (?), Web of Science (?), and Google Scholar (?). Include the references or websites consulted. There are packages in R and Python that facilitate the automated search for specific scientific articles and the generation of graphs based on predefined criteria. Did you use any libraries in these languages for this purpose or did you select the articles manually, analyzing them one by one?

L 148 ... (e.g., R² (significance?), RMSE (significance?), MAE (significance?), significance tests (Which ones?)),

L 149 ... exclusively on other crops (Which ones?),

L 156 ... pigment and phenolic trait (Which ones?) estimation.

L 161. Increase the size and font of Figure 2. ... NDRE (?), SAVI (?), and GNDVI (?), while indices such as TCARI (?) / OSAVI (?) remain unexplored. Why are the TCARI and OSAVI indices, despite their potential for estimating pigments and phenolic characteristics, little explored in practice?

L 167. The use of drones to obtain detailed data at both individual plant and vineyard-block scales (?) has significantly increased (?) in viticulture [47]. How large is the plot scale? What was the significant increase? Cite it.

L 169-170 ... especially under climate change scenarios (What scenarios are these?), requires accurate (How accurate is the monitoring of vine health and condition?) [48].

L 175 ... more efficiently than ground-based methods (?) [49] What ground-based methods are these? L 176 ... with a wide range of RGB sensors (?)...

L 177 ... adapted to specific monitoring objectives (?) [50].

L 183 ... physiological imbalances (Examples?),

L 188 ... of vegetation indices and spatial models (Which ones?), thereby facilitating site-specific decision-making (For example?)...

L 191. It would be interesting if you could address the main characteristics of drone flights, as well as the adjustment parameters used in capture by multispectral and thermal cameras. Furthermore, it is worth describing the technical specifications of the multispectral cameras used in the research, as well as the main software used for image processing and analysis. Finally, I suggest including a photo of the calibration plate or reflectance panel in the flowchart.

I couldn't find a figure specifying the phenological stages of grapevines. This information is important.

L 194-196 ... ... two-dimensional (2D) workflows are sufficient ... Explain the sentence better! It is not clear.

... are processed into orthomosaic (?), enabling precise assessment of vineyard health, vigor, and spatial variability... Is orthomosaic a program?

L 209. Give the meaning of DSM, DEM. List the steps of the flowchart, and cite them in the caption of Figure 3. Increase the size of Figure 3. The scales of the figures are not visible.

L 213 ... accounting for solar inclination (?) Wouldn't it be the zenith angle of the sun or solar elevation?

L 220 ... such as k-means and CLARA (Meaning?)

L 225 ... These methodologies significantly enhance the precision of... How much does it increase precision?

L 229. Figure 4. It shows this segmentation process in three key steps:... It is not a paragraph. Improve the text in the presentation of Figure 4. Increase the size of Figure 4. You cannot see the scale of the NDVI, nor of Figure 4A.

L 260... It is calculated from the difference between near-infrared (NIR) and red reflectance (Eq. 1),... Complete the explanation of the formula! Why does NDVI have to be normalized?

L 269. Include a reference to Table 1.

L 275... NDVI reached a maximum 𝑅2= 0.61 (?) in predicting total soluble solids (TSS) during... At what confidence level was the R² value significant? Please report it!

L 281... NDVI remains a powerful tool (?) for vineyard zoning and... NDVI is not a tool, but a biophysical variable! Review this!

L 282... NDVI-based vigor classes correlate (?) with grape and wine quality... What is the value of this correlation? Is it significant, at what confidence level? L 283... low-vigor (?) vines produced lower yields but grapes with higher sugar (?) and anthocyanin levels (?). What is the NDVI value of this low vigor? List the values.

L 284... resulting in wines with superior sensory properties (?). Give an example of sensory properties.

L 290... complementary indices (which ones?) to improve...

L 295... specific limitations (?). GNDVI is... Which ones? List them. Complete the explanation of the formula! Why does GNDVI have to be normalized?

L 301. Ferro et al. [73] reported a strong correlation (R² = 0.839) with grape... Is the R² significant, at what confidence level?

L 315... SAVI improves the accuracy (?) of vigor estimations in vineyard inter-rows and arid zones [82,83]. By how much does SAVI improve the accuracy of the estimates? Explain.

L 318. Review formula 3. It's not correct!

Huete, A.R. A soil-adjusted vegetation index (SAVI), Remote Sensing of Environment, V.25, Issue 3, 1988, p. 295-309.

L 323 ...combined with crop coefficients (Kcb), improves evapotranspiration modeling (?), highlight... By how much does Kc combined with SAVI improve evapotranspiration? Explain. In grapevines? Sorry, I couldn't find this comparison in the article by Campos et al. [87]. Review.

L 365-366 ...differences relative to reference conditions (?) (Eq. 5) [93]. What reference conditions are these?

L 407 ...area and ground-projected vine area (Eq. 6) (?): Enter the reference. L 415-420...reported a strong correlation (R² = 0.76)...

... Ilniyaz et al.[103] achieved high accuracy (R² = 0.899)...

... VNIR data into a PLSR (?) model (R² ≈ 0.79), an approach particularly useful in...

R² is significant at what confidence level?

What are PLSR and VNIR?

L 452. Review the SAVI formula. It's not correct.

L 492... (?) [108] showed that UAV-based thermal and multispectral integration improved (?) CWSI map... Cite the authors. What is the value of the improvement in the CWSI estimate? Cite them.

L 502... (?) [111] applied UAV thermal imagery...

L 729-730... an insufficient distribution of ground control points (GCPs). How many control points would be enough?

L 732 ...the adoption of RTK (?) or PPK (?) systems,... Give the meaning?

Author Response

Reviewer #2:

General Response

Dear Reviewer,

Thank you for your constructive feedback. In response to your suggestion regarding the inclusion of grapevine phenology, we have incorporated a dedicated paragraph in the introduction outlining the phenological stages of the crop. This addition provides essential context for the application of vegetation indices across the crop cycle. We chose not to create a separate section on phenology, as it would have diverted attention from the primary focus of the article, which is to review UAV-based spectral and thermal indices in precision viticulture.

Furthermore, we have revised and clarified Figure 1, specifying the canopy structure, correcting the units of measurement, and providing a clearer description of the number of leaf layers. All reported coefficients of determination (R²) have been updated to reflect their statistical significance and confidence intervals where applicable. The formula for the SAVI index has been corrected to its accurate definition, and we have meticulously reviewed all citations to ensure proper referencing. To reinforce the theoretical foundation of the article, we have also included key classical and foundational references. These revisions collectively enhance the clarity, accuracy, and relevance of the manuscript while maintaining the focus on UAV-based vegetation indices for precision viticulture.

Comment (1)

L 15. ... indices can also inform grape composition (?)... Give examples of types of grape composition that can be detected by the indices?

Response (1)

Dear Reviewer, Thank you for this suggestion. We have revised the abstract to explicitly include examples of grape composition. New line: 13-16.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (2)

Response (2)

Thank you for the comment. The abstract now specifies that field or lab measurements of key grape traits validate UAV-based predictions, highlighting the importance of ground-truth data for precision viticulture. New line: 13-16

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (3)

L 20. ... and discuss their strengths (Which ones?), limitations (Which ones?), and integration (Which ones?) with... Please give examples of the strengths, limitations, and types of integrations.

Response (3)

Thank you for the suggestion. We now highlight the strengths (e.g., high resolution, fast acquisition), limitations (e.g., saturation in dense canopies, environmental sensitivity), and integration of UAV indices with ground data and other sensors. New line: 16-19

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (4)

Response (4)

Dear Reviewer, thank you for this suggestion. In the manuscript, we now specify abiotic stresses (water deficit, nutrient deficiency) and biotic factors (pest and fungal infections), highlight index saturation in dense canopies (Merlot, Cabernet Sauvignon), and emphasize calibration across varieties (Riesling, Cabernet Sauvignon, Merlot) and multiple seasons. New line: 22-27.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (5)

L 27. Do not repeat words from the title. Use words related to your study. Replace NDVI, UAV, precision agriculture. Suggestions: Multispectral images, wine quality, spectral indices, etc.

Response (5)

Dear reviewer, thank you very much for your suggestion to modify my keywords. New line: 28

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (6)

Response (6)

Dear reviewer, thank you very much for your suggestion. I have modified part of the introduction of the manuscript, so as not to confuse the reader New line: 32

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (7)

L 34. ... due to their long-standing viticultural traditions (?) ... Sorry, I don't know! Give an example of viticultural traditions.

Response (7)

Dear reviewer, thank you very much for your suggestion. I have modified part of the introduction of the manuscript. New line: 37-39.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (8)

L 35. ... the second-largest wine exporter by volume (?) and the...? Write the value of the volume of wine exported.

Response (8): Dear Reviewer, thank you for your valuable suggestion. We have revised the manuscript to include the exact volume of Spain’s wine exports in 2023. The updated sentence now reads: “In 2023, Spain ranked as the second-largest wine exporter by volume, shipping approximately 20.9 million hectoliters. New line: 38-39.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (9)

L 37. Viticulture is practiced across a wide range of climatic regions (?) and soil types (?), ... What are these regions and soil types?

Response (9):

Dear Reviewer, we appreciate your suggestion regarding the need to clarify the climatic regions and soil types where viticulture is practiced. To address this, we have revised the sentence to include representative climatic regions (Mediterranean, continental, oceanic, and semi-arid) and diverse soil types (sandy, clay, loam, limestone, and volcanic) to provide greater specificity and context. New line: 40-42

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (10)

L 38. ...yet grapevines are highly sensitive to biotic (?) and (?) abiotic stress factors... Specify the biotic and abiotic factors.

Response (10)

Dear Reviewer, we appreciate your suggestion to specify the biotic and abiotic stress factors affecting grapevines. In response, we have expanded the paragraph to include examples of biotic factors and abiotic factors to enhance precision and clarity. New line: 43-48

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (11)

L 41-42. ...excessive humidity (How much?) and inadequate irrigation management [5]. Give examples of inadequate management.

Response (11)

Dear reviewer, we appreciate your feedback regarding the need for greater specificity in our description of environmental factors. New line: 50-51.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (12)

L 43. ...accelerate phenology (What stage? All phenological cycles?), alter berry composition (How?) ...

Response (12)

Dear Reviewer, we appreciate your suggestion regarding the need to clarify which phenological stages are accelerated and how berry composition is altered under rising temperatures. In response, we have specified that climate change accelerates budburst, flowering, and véraison and alters berry composition by increasing sugar accumulation and decreasing acidity, which can impact wine quality and ripening dynamics. New line: 76-79.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (13)

L 44. ...which in turn amplifies pest pressure (?) [6] ... What type of pests? List them.

Response (13)

Dear Reviewer, we appreciate your suggestion to specify the types of pests affected by climate change. We have revised the sentence to include grapevine moths and spider mites as representative insect pests whose population pressures are amplified under irregular water availability and warmer conditions. New line: 80-81

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (14)

L 45-46. ...equipped with multispectral sensors (Only multispectral? Thermal ones? Give examples).

...as they enable precise monitoring (What is the value of this precision?) ...

Response (14)

Dear reviewer, thank you for your insightful comment. We have clarified in the manuscript that UAVs are equipped with both multispectral and thermal sensors and emphasized the precision they provide in vineyard monitoring. New line: 82-83

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (15)

L 47.... optimization of water resources (Give an example?), and early detection of stress symptoms (How?) ...

Response (15)

Dear Reviewer: We revised the sentence to clarify how drone technologies optimize water resources and detect stress symptoms. The text now specifies that targeted irrigation scheduling, based on the spatial variability of vine water demand, can improve water use efficiency. We also explained that early detection of water stress, nutrient deficiencies, and disease onset is achieved through the analysis of spectral indices (NDVI, NDRE) and thermal images (canopy temperature maps). New line: 84-86

Comment (16)

L 48... viticulture [7] ... remove one point, there are two.

Response (16)

Ref 14 New Line 88

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (17)

L 49. Precision agriculture technologies (Which ones?) have become valuable...

Comment (18)

L 51. ...multispectral data... And don't thermal sensor data deserve to be highlighted as well?

Response (17-18)

Dear Reviewer, we appreciate your suggestion. Precision agriculture technologies, including unmanned aerial vehicles (UAVs), proximal sensors, soil moisture probes, and satellite-based monitoring, have become valuable tools for vineyard management, with UAVs emerging as particularly effective for acquiring multispectral and thermal sensor data over large areas in a non-invasive manner. New line: 90-94.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (19)

L52-53 ...capture ultra-high-resolution imagery (I don't see the need to use the word ultra-high. Just high-resolution would suffice, wouldn't it? What is the spatial resolution? mm? cm?) provides a high-throughput alternative to traditional monitoring methods (How? List them.)

Response (19)

Dear Reviewer, we appreciate your suggestion. Their ability to capture high-resolution imagery with spatial resolutions ranging from 1-5 centimeters provides a high-throughput alternative to traditional monitoring methods, including manual leaf sampling, visual crop scouting, ground-based phenological assessments, conventional meteorological station measurements, and laboratory-based plant tissue analysis, enabling detailed assessments of canopy vigor, disease incidence, pest infestations, and vine water status. New line: 94-97.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (20)

L 54. Compared to satellite platforms (Which satellites?)

Response (20)

Dear reviewer, we appreciate your suggestion to modify the paragraph in the article as follows. These are commonly used in precision viticulture, such as Landsat 8/9 (native resolution of 30 m, panoramic focus up to 15 m) and Sentinel-2 (10-20 m). New line: 99-101

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (21)

L63-64 ...by pigment concentration (Which ones?), water content, and cell anatomy (What anoyomia is this?), as well as by soil properties (Which ones?) and background effects (Which ones? Explain.).

Response (21)

Dear reviewer, we appreciate your suggestion to modify the paragraph in the article as follows Spectral responses are also influenced by pigment concentrations such as chlorophyll a/b and carotenoids, leaf cell anatomy including mesophyll structure and air space distribution, and by soil properties (moisture content, organic matter, texture, and surface roughness) as well as background effects like understory vegetation, vine row shadows, or residues from previous crops. New line: 110-111.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (22)

L 65. ...vegetation typically displays low... What vegetation? Under what conditions?

Comment (23)

L 69 ...design (Which design?) and interpretation of vegetation indices, and requires careful consideration (Which ones?) of soil...

Response (22-23)

Dear reviewer, we appreciate your suggestion to modify the paragraph in the article as follows

Spectral responses are also influenced by pigment concentrations such as chlorophyll a/b and carotenoids, leaf cell anatomy including mesophyll structure and air space distribution, and by soil properties (moisture content, organic matter, texture, and surface roughness) as well as background effects like understory vegetation, vine row shadows, or residues from previous crops. New line: 109-112

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (24)

Response (24)

Dear Reviewer, we appreciate your comment. We have clarified that what was previously referred to as Figure 1 is now Figure 2 in the revised manuscript. The Y-axis in Figure 2 now reads “Reflectance Factor (dimensionless),” with the scale kept from 0 to 1, as it represents a dimensionless quantity (the ratio of reflected to incident radiation). According to the literature, the water content of a healthy leaf typically ranges from 70–80%, whereas stressed leaves may have water content around 50–60%. The spectral signature shown corresponds to a single-layer leaf canopy under normal conditions and illustrates general vegetation patterns, without specifying a particular species or the exact number of leaf layers in the canopy, as indicated in the updated figure legend (New Lines 114–124).

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (25)

L 78. Sensor performance is defined by spatial and spectral resolution. And isn't radiometric resolution important?

L 80-81 ... such as individual leaves, disease symptoms (Which diseases and which symptoms? List them.)

Response (25)

Dear reviewer, we appreciate your technical corrections and suggestions. We have modified the manuscript to address your three specific comments:

L77: We corrected the terminology to clarify that sensors initially capture "spectral radiance" rather than "spectral reflectance.". New lines: 134.

L78: We acknowledge the importance of radiometric resolution and have updated the sentence to: "Sensor performance is defined by spatial, spectral, and radiometric resolution." New line: 137.

L80-81: We have specified the disease symptoms by listing: "such as individual leaves and disease symptoms, including powdery mildew (leaf spots and coating), downy mildew (yellow lesions and sporulation), and leaf necrosis." New line: 141-142.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (26)

L 100 ... facilitating site-specific management and harvest strategies (The reference is missing, isn't it?)

Response (26)

Dear reviewer, thank you for your viewing. I just added the missing reference. New line: 162.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (27)

L 102-103 ... with distinct physicochemical properties (Which ones? List them) at harvest.

Response (27)

Dear reviewer, we appreciate your suggestion. To clarify, the physicochemical properties that differentiate grape lots at harvest, as reported by Urretavizcaya et al. [32], include berry weight (BW), total soluble solids (TSS), titratable acidity (TA), malic acid (MA), tartaric acid (TA), yeast assimilable nitrogen (YAN), total anthocyanins (TA), extractable anthocyanins (ETA), and total phenolics (TP). These variables were significantly associated with spatial variability assessed using NDVI combined with soil and yield data. We have explicitly included these parameters in the manuscript on lines 165-168.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (28)

L 109. ... Related indices such as RDVI (?), OSAVI (?), and NDRE (?) have... The acronyms for these indices appear for the third time without their meanings. Please write them out.

Response (28)

Dear Reviewer, thank you for pointing this out. We have now expanded the acronyms for the related vegetation indices the first time they are mentioned in the manuscript: RDVI (Renormalized Difference Vegetation Index), OSAVI (Optimized Soil-Adjusted Vegetation Index), and NDRE (Normalized Difference Red-Edge Index). This clarification has been added to improve readability and precision. New line: 174-176.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (29)

L113 – “Unlike many other fruits (Which ones?)”

Response (29)

Dear Reviewer, we appreciate your observation. We have specified examples of climacteric fruits such as apples, bananas, mangoes, and tomatoes that continue ripening after harvest, in contrast to grapes. New line: 180

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (30)

L 116. Grape maturity at harvest plays a decisive role in determining wine quality (?). Explain why grape maturity at harvest determines wine quality.

Response (30)

L116 – “Explain why grape maturity at harvest determines wine quality.”

Dear Reviewer, we have expanded the explanation to clarify that grape maturity influences sugar concentration, acidity, phenolic composition, tannin structure, and aromatic precursors, all of which affect the sensory attributes, stability, and aging potential of the resulting wine. New line: 184-188.

Comment (31)

L 123 ... and aromatic precursors (Which ones?), thereby linking...

Response (31)

Dear Reviewer, we have detailed examples of aromatic precursors, including monoterpenes (e.g., linalool, geraniol), norisoprenoids (e.g., β-damascenone), and methoxypyrazines, which are key contributors to varietal aroma. New line: 194-195.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (32)

L 130 ... spectral indices with wine quality attributes (?), What are these attributes?

Response (32)

L130 – “What are these wine quality attributes?”

Dear Reviewer, we have specified wine quality attributes such as alcohol content, acidity, color intensity, phenolic composition, tannin structure, aromatic complexity, and overall sensory quality. New line: 209-210

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (33)

Response (33)

Dear Reviewer, we added the URLs for Scopus (https://www.scopus.com/), Web of Science (https://www.webofscience.com/), and Google Scholar (https://scholar.google.com/) in the methodology and clarified that the search and selection were performed manually, without using automated bibliometric packages in R or Python. Articles were analyzed individually according to predefined inclusion/exclusion criteria. New lines: 227-231

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (34)

L 148 ... (e.g., R² (significance?), RMSE (significance?), MAE (significance?), significance tests (Which ones?)),

Response (34)

Dear Reviewer, we specified that R², RMSE, and MAE values were considered only when their statistical significance was provided in the original studies (e.g., R² with p < 0.05, ANOVA, Tukey’s HSD, or Pearson correlation). This ensures the robustness of the findings summarized in the review. New line: 236-239.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (35)

L 149 ... exclusively on other crops (Which ones?),

Response (35)

Dear Reviewer, we clarified that excluded studies focused on non-vineyard crops, including wheat, maize, and rice, as these were outside the scope of precision viticulture. New line: 239-239-241

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (36)

L 156 ... pigment and phenolic trait (Which ones?) estimation.

Response (36)

Dear Reviewer, thank you for your comment. In the revised manuscript, we no longer refer to pigment or phenolic traits. Instead, we highlight that TCARI and OSAVI remain underexplored mainly due to their need for complex calibration and sensitivity to canopy and soil conditions, which limits their use in commercial vineyard operations. New Line 247 -250

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (37)

Response (37)

Dear Reviewer, we enlarged the font and markers in Figure 2 to enhance readability. Additionally, we explained that TCARI and OSAVI are less frequently applied because they require more complex calibration or are sensitive to canopy and soil conditions, limiting their routine use in commercial viticulture. New line: 256-259

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (38)

Response (38)

Dear Reviewer, we appreciate your comment. We have clarified the spatial scale in the manuscript. The revised sentence now reads: The use of drones to obtain detailed data at both individual plant and vineyard-block scales, typically ranging from single vines up to plots of 1–5 hectares, has increasingly been applied in viticulture. New lines: 276-278

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (39)

L 169-170 ... especially under climate change scenarios (What scenarios are these?), requires accurate (How accurate is the monitoring of vine health and condition?)

Response (39)

Dear Reviewer, we have specified the climate change scenarios and the level of monitoring accuracy. Ensuring optimal grape production and consistent wine quality, especially under climate change scenarios such as increased frequency of heat waves, irregular precipitation patterns, and drought events, requires highly accurate monitoring of vine health and condition, with typical sensor spatial resolutions ranging from 1 to 5 cm per pixel for canopy-level observations [54]. New lines: 278-281

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (40)

L 175 ... more efficiently than ground-based methods (?) [49] What ground-based methods are these?

Response (40)

Dear Reviewer, we appreciate your comment. We have clarified the types of ground-based methods used in vineyard monitoring, specifying manual canopy measurements, leaf sampling, destructive biochemical assays, and proximal sensors mounted on tractors. New line: 285-287.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (41)

L 176 ... with a wide range of RGB sensors (?)...

Response (41)

Dear Reviewer, we appreciate your comment. We have clarified that the UAVs are equipped with a variety of sensors, including RGB, multispectral, hyperspectral, thermal, and LiDAR, to indicate that the diversity applies to all sensor types and not only to RGB sensors. New lines: 288-290.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (42)

L 177 ... adapted to specific monitoring objectives (?) [50]

Response (42)

Dear Reviewer, we appreciate your comment. We have clarified the manuscript by specifying that UAVs can be adapted to specific monitoring objectives, including estimating canopy vigor, detecting disease outbreaks, assessing water stress, and mapping nutrient variability. New lines: 288-290

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (43)

L183 – physiological imbalances (Examples?)

Response (43)

Dear reviewer, we clarified that physiological imbalances such as nitrogen deficiency, water stress, and early disease can be detected through variations in red and NIR reflectance.". New line: 296

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (44)

L188 – of vegetation indices and spatial models (Which ones?), thereby facilitating site-specific decision-making (For example?)

Response (44)

Dear Reviewer, we have specified the vegetation indices and spatial models used, and provided an example of site-specific decision-making. Vegetation indices such as NDVI, GNDVI, and PRI, and spatial models including kriging and regression-based yield prediction, were applied. For example, areas showing low NDVI values or irregular canopy structure can be targeted for differential fertilization, irrigation adjustments, or selective pruning to optimize vineyard performance. New line: 301-305

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (45)

I couldn't find a figure specifying the phenological stages of grapevines. This information is important.

Response (45)

Dear reviewer, we have added detailed UAV flight parameters (altitude 30–50 m, speed 3–5 m/s, frontal overlap 75–85%, lateral overlap 65–75%), camera settings and specifications, software used for image processing (Pix4Dmapper, Agisoft Metashape, QGIS), and a calibration panel in the workflow. Additionally, a phenological diagram of the vines was included in the introduction, showing key stages from bud break to harvest, to provide context for interpreting UAV-derived spectral indices throughout the growing season. New line: 311-313

Comment (46)

L 194-196 ... ... two-dimensional (2D) workflows are sufficient ... Explain the sentence better! It is not clear.

... are processed into orthomosaic (?), enabling precise assessment of vineyard health, vigor, and spatial variability... Is orthomosaic a program?

Response (46)

Dear Reviewer, we have clarified the sentence regarding two-dimensional (2D) workflows. It now specifies that 2D workflows involve aligning and mosaicking overlapping images to generate a continuous canopy representation without reconstructing 3D structures, which is suitable for standard vegetation index analyses. New line: 322-324

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (47)

L 209. Give the meaning of DSM, DEM. List the steps of the flowchart, and cite them in the caption of Figure 4. Increase the size of Figure 4. The scales of the figures are not visible.

Response (47)

Dear Reviewer, we have defined DSM (Digital Surface Model) and DEM (Digital Elevation Model) in the manuscript. The steps of the flowchart are now listed in the caption of Figure 5, and the figure size has been increased to improve the visibility of the scales. New line: 334–335 and Fig. 5.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (48)

L 213 ... accounting for solar inclination (?) Wouldn't it be the zenith angle of the sun or solar elevation?

Response (48)

Dear Reviewer, we have corrected the terminology. The manuscript now refers to solar zenith angle and slope, aspect corrections rather than “solar inclination,” to clarify the angular correction applied to spectral data. New line: 338-340

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (49)

L 220 ... such as k-means and CLARA (Meaning?)

Response (49)

Dear Reviewer, we have included definitions for k-means (partitional clustering) and CLARA (Clustering Large Applications), explaining that these algorithms are used to discriminate vegetation, soil, and shadows based on spectral traits. New line: 344-347

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (50)

L 225 ... These methodologies significantly enhance the precision of... How much does it increase precision?

Response (50)

Dear Reviewer, we specified the effect on precision. These methodologies significantly enhance the precision of multispectral analyses, improving classification accuracy and vineyard monitoring by approximately 10–15% compared to basic thresholding methods, as reported in the cited studies. New line: 350-352

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (51)

L 229. Figure 4. It shows this segmentation process in three key steps: ... It is not a paragraph. Improve the text in the presentation of Figure 4. Increase the size of Figure 4. You cannot see the scale of the NDVI, nor of Figure 4A.

Response (51)

Dear Reviewer, we improved the text describing Figure 6. The figure now presents the segmentation workflow in a paragraph format: (A) a binary mask distinguishes vines from soil and other non-plant elements; (B) NDVI is applied to the segmented image to analyze crop vigor and detect spatial variations; and (C) the segmentation is validated with field observations to ensure accuracy. Figure 6 has been enlarged, and the scales for NDVI and Figure 4A are now clearly visible. New line: 353-354

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (52)

L 260... It is calculated from the difference between near-infrared (NIR) and red reflectance (Eq. 1) ... Complete the explanation of the formula! Why does NDVI have to be normalized?

Response (52)

Dear Reviewer, thank you for your valuable comment. As detailed in Line 385, we expanded the explanation of the NDVI formula, clarifying that NDVI is calculated as the difference between NIR and RED reflectance divided by their sum. This normalization minimizes the effects of illumination variability, sensor sensitivity, and atmospheric conditions, ensuring consistent comparisons of vegetation vigor across different flights, dates, and sensors. New line: 385-388

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (53)

L 269. Include a reference to Table 1.

Response (53)

Dear reviewer, thank you for pointing this out. We have now included a direct reference to Table 1, summarizing typical NDVI vigor classifications based on UAV-derived multispectral imagery, as indicated in Line 269 of the revised manuscript. New line: 398

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (54)

L 275... NDVI reached a maximum ?2= 0.61 (?) in predicting total soluble solids (TSS) during...

Response (54)

Dear reviewer, we have clarified the statistical significance. As now specified in Line 351, the R² value of 0.61 for predicting total soluble solids (TSS) using NDVI was statistically significant at p < 0.01, indicating over 99% confidence that the relationship is not due to random chance. New line: 403

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (55)

L 281... NDVI remains a powerful tool (?) for vineyard zoning and... NDVI is not a tool, but a biophysical variable! Review this!

Response (55)

Dear Reviewer, thank you for your comment. As clarified in Line 411, we have revised the text to indicate that NDVI is a biophysical variable rather than a tool. It functions as an indicator integrated into tools and methods for vineyard zoning, vigor assessment, water status evaluation, and other ecophysiological analyses, making it a key parameter in precision viticulture.. New line: 411-412

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (56)

L 282... NDVI-based vigor classes correlate (?) with grape and wine quality... What is the value of this correlation? Is it significant, at what confidence level?

L 283... low-vigor (?) vines produced lower yields but grapes with higher sugar (?) and anthocyanin levels (?). What is the NDVI value of this low vigor? List the values.

Response (56)

Dear Reviewer, Thank you for your comments. As clarified in new Line 414, the correlation between NDVI-based vigor classes and grape wine quality was significant at p < 0.05 according to Filippetti et al. [77]. The study did not report exact numerical correlation coefficients or specific NDVI thresholds for low-vigor (LV) vines. However, LV zones were characterized by lower NDVI values relative to high-vigor (HV) zones, reflecting reduced vegetation activity. This supports the use of NDVI maps for vineyard zoning and site-specific management. New line: 413-416

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (57)

L 284... resulting in wines with superior sensory properties (?). Give an example of sensory properties.

Response (57)

Dear Reviewer, as detailed in Line 418, wines from low-vigor vines exhibited superior sensory properties, including more intense fruity aroma, refined astringency, and concentrated flavor, enhancing the overall perception of wine quality compared to wines from high-vigor zones. New line: 418

Comment (58)

L 290... complementary indices (which ones?) to improve...

Response (58)

Dear Reviewer, thank you for your comment. As clarified in Line 373, we have specified the complementary indices commonly used alongside NDVI in precision viticulture to improve interpretation accuracy. These include NDRE, EVI, GNDVI, VARI, SAVI, PCDI, and PRI, each contributing specific information such as dense-canopy discrimination, early-season vigor assessment, soil-background correction, canopy structure evaluation, or stress detection. New line: 425-429

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (59)

L 295... specific limitations (?). GNDVI is... Which ones? List them. Complete the explanation of the formula! Why does GNDVI have to be normalized?

Response (59)

Dear Reviewer, thank you for your comment. We have clarified the specific limitations of GNDVI in the manuscript. These include the phenological stage (lower correlation before ripening), soil-background effects on scattered canopies, and environmental conditions such as cloud cover or atmospheric scattering. In addition, we have completed the explanation of the GNDVI formula and why it needs to be normalized (see lines 390-392). New line: 442-444

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (60)

L 301. Ferro et al. reported a strong correlation (R² = 0.839) with grape... Is the R² significant, at what confidence level?

Response (60)

Dear reviewer: Thank you for your comment. As clarified in line 301, the strong correlation reported by Ferro et al. (R² = 0.839) for grape yield during ripening is highly significant, exceeding the 99% confidence threshold and meeting the stricter criterion of p ≤ 0.001. This confirms the robustness of the GNDVI as an indicator of nutritional status and yield potential during ripening. See lines 396 and 397. New line: 448-449

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (61)

L 315... SAVI improves the accuracy (?) of vigor estimations in vineyard inter-rows and arid zones [82,83]. By how much does SAVI improve the accuracy of the estimates? Explain.

Response (61)

Dear Reviewer, we have clarified the role of SAVI in improving precision in vineyard vigor estimations. SAVI incorporates a soil-adjustment factor that reduces the influence of soil reflectance, particularly in inter-row zones, early growth stages, or arid vineyards. This correction improves the correlation between vegetation indices and biophysical parameters such as leaf area, crop coefficient (Kcb), and yield, thereby enhancing the reliability of vigor assessments under heterogeneous canopy and soil conditions. New line: 461-464

Comment (62)

L 318. Review formula 3. It's not correct!

Huete, A.R. A soil-adjusted vegetation index (SAVI), Remote Sensing of Environment, V.25, Issue 3, 1988, p. 295-309.

Response (62)

Dear Reviewer, thank you for your observation. The formula reference has been updated. New line: 446

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (63)

Response (63)

Dear Reviewer, we apologize, the document originally cited was not correct and has now been updated. Please see the revised explanation in lines 418–420 of the manuscript. Campos et al. [89] showed that adding a SAVI-derived basal crop coefficient (K cb) to the traditional K c model for grapevines improves evapotranspiration estimation only marginally, reducing the root-mean-square error by approximately 2–5 %, with limited practical advantage under the studied vineyard conditions. New line: 470-476

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (64)

L 365-366 ...differences relative to reference conditions (?) (Eq. 5) [93]. What reference conditions are these?

Response (64)

Dear Reviewer, thank you for your observation on line 518-519. We have revised the text to clarify that "reference conditions" specifically refer to "well-watered and fully-stressed reference baselines" as established in CWSI methodology. This modification addresses your concern and improves the technical precision of the statement. New line: 519

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (65)

L 407 ...area and ground-projected vine area (Eq. 6) (?): Enter the reference.

Response (65)

Dear reviewer, thank you for your observation. I have updated the reference in the manuscript. Line 561. New line: 561

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (66)

L 415-420...reported a strong correlation (R² = 0.76) ...

... Ilniyaz et al. [103] achieved high accuracy (R² = 0.899) ...

... VNIR data into a PLSR (?) model (R² ≈ 0.79), an approach particularly useful in...

R² is significant at what confidence level?

What are PLSR and VNIR?

Response (66)

Dear Reviewer, we have clarified the statistical significance of the reported R² values in our manuscript.

Vélez et al. [104] reported R² = 0.76, which is significant at p < 0.01 (≥ 99 % confidence).
Ilniyaz et al. [105] reported R² ≈ 0.899, significant at p < 0.05 (95 % confidence).
Stobbelaar et al. [106] reported R² ≈ 0.79 for the PLSR model, significant at p < 0.001 (> 99.9 % confidence).

Partial Least Squares Regression (PLSR) and VNIR (Visible Near Infrared) have been briefly explained in the manuscript. Please see the revised paragraph in lines 415–420 for these clarifications. New line: 569-575

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (67)

L 452. Review the SAVI formula. It's not correct.

Response (67)

Dear Reviewer, thank you for your observation. The formula reference has been updated. New line: 601- Table 2

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (68)

L 492... (?) [108] showed that UAV-based thermal and multispectral integration improved (?) CWSI map... Cite the authors. What is the value of the improvement in the CWSI estimate? Cite them.

Response (68)

The cited study Alchannatis et al [6] reported that the integration of UAV-based thermal imagery with multispectral data improved the accuracy of remotely estimating the Crop Water Stress Index (CWSI). However, the authors did not provide a specific quantitative value (e.g., percentage reduction in error or absolute change in CWSI) for this improvement. New line: 640-644

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (69)

L 502... (?) [111] applied UAV thermal imagery...

Response (69)

Dear Reviewer, thank you for your comment; the suggested improvement has been incorporated into the manuscript. New line: 649

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (70)

L 729-730... an insufficient distribution of ground control points (GCPs). How many control points would be enough?

Response (70)

Dear Reviewer, the optimal number of ground control points (GCPs) depends on field size and topography; typically, 5–10 well-distributed GCPs are sufficient for vineyard-scale UAV surveys. See the extended discussion in lines 729–730 of the manuscript. New line: 878-879

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment (71)

L 732 ...the adoption of RTK (?) or PPK (?) systems, Give the meaning?

Response (71)

Dear Reviewer, RTK (Real-Time Kinematic) and PPK (Post-Processed Kinematic) are high-precision GNSS techniques that provide centimeter-level positioning corrections during or after UAV flights, respectively. Their use, combined with careful flight planning and photogrammetric processing, improves spatial accuracy and reduces geometric distortions. See lines 881–884 in the manuscript.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

I have reviewed the manuscript titled "UAV-based spectral and thermal vegetation indices for precision viticulture: A Review on NDVI, NDRE, SAVI, GNDVI, and CWSI". The authors have presented an informative and thorough review paper, and I congratulate them on their work. I have only a few minor comments that I believe will help them further improve their manuscript.

I recommend revising the paper’s title to be more general and removing any specific indices currently named in the title, because the manuscript presents additional specific indices.
Lines 127-131: While I appreciate the content, the authors should provide more detailed information about previously published review papers and include specific comparisons to better highlight the novelty of their work.
Lines 145–146: Please include a list of the reviewed papers (e.g., title, DOI, year, etc.) in a table in the Appendix, a link to such a table.
It would also be useful to present publication trends from 2015 to 2024, in which for example assessing whether the application of NDVI has received less attention in recent years.

Author Response

Reviewer #3:

Comment 1

I recommend revising the paper’s title to be more general and removing any specific indices currently named in the title, because the manuscript presents additional specific indices.

Response 1

Dear Reviewer, thank you for your helpful suggestion. We have revised the title to provide a broader perspective and to better reflect the overall scope of the review. After careful consideration, we decided to retain the primary indices (NDVI, NDRE, SAVI, GNDVI, and CWSI) in the title because these indices represent the core focus and the majority of the studies analyzed, making them central to the paper’s contribution. While we briefly mention other indices (e.g., TCARI/OSAVI, ExGR), they were only discussed as supplementary or underexplored alternatives; therefore, including them in the title would unnecessarily lengthen it and dilute the emphasis on the indices most relevant to precision viticulture.

The revised title now reads:

UAV-based multispectral and thermal indices for precision agriculture applications: A review focused on NDVI, NDRE, SAVI, GNDVI, and CWSI. Line 3

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment 2

Lines 127-131: While I appreciate the content, the authors should provide more detailed information about previously published review papers and include specific comparisons to better highlight the novelty of their work.

Response 2

Dear Reviewer, thank you for your valuable suggestion. We have expanded the relevant paragraph (lines) to include detailed information and explicit comparisons with previously published review papers (Sassu et al., 2021; Giovos et al., 2021; Ammoniaci et al., 2021; Ferro & Catania, 2023; Singh et al., 2022; Tardaguila et al., 2021). These comparisons now clearly highlight how our work differs from prior reviews by systematically synthesizing UAV-derived NDVI, NDRE, SAVI, GNDVI, and CWSI in relation to grape maturity and wine quality attributes an aspect not comprehensively addressed in earlier studies. Line 202-207

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Comment 3

Lines 145–146: Please include a list of the reviewed papers (e.g., title, DOI, year, etc.) in a table in the Appendix, a link to such a table.

Response 3

Dear Reviewer,

Thank you very much for your thoughtful suggestion. All the articles selected and analyzed in this review are already fully cited in the reference list, which includes the complete bibliographic details (title, DOI, year, etc.). Because adding a table with the same information could be redundant, we did not initially include it as an appendix. However, if you believe a dedicated table would enhance clarity or accessibility for readers, we are happy to prepare and include such a table in the Appendix according to your preference. We appreciate your guidance on whether this addition would meaningfully improve the manuscript.

Title	Doi	Year
Response of plants to water stress	doi.org/10.3389/fpls.2014.00086	2014
Use of very high-resolution airborne images to analyse 3D canopy architecture of a vineyard	doi.org/10.5194/isprsarchives-XL-3-W3-399-2015	2015
Dynamics and simulation of the effects of wind on UAVs and airborne wind measurement	doi.org/10.2322/tjsass.58.187	2015
Vineyard detection from unmanned aerial systems images	doi.org/10.1016/j.compag.2015.03.011	2015
Unmanned Aerial Vehicle (UAV)-based remote sensing to monitor grapevine leaf stripe disease within a vineyard affected by esca complex	doi.org/10.14601/Phytopathol	2015
Using active canopy sensors and chlorophyll meters to estimate grapevine nitrogen status and productivity	doi.org/10.1007/s11119-014-9363-8	2015
Multispectral band selection for imaging sensor design for vineyard disease detection: case of Flavescence Dorée	doi.org/10.1017/s2040470017000802	2017
Detection of Flavescence dorée grapevine disease using Unmanned Aerial Vehicle (UAV) multispectral imagery	doi.org/10.3390/rs9040308	2017
Using multispectral imaging to improve berry harvest for wine making grapes	doi.org/10.1051/ctv/20173201033	2017
Assessment of a canopy height model (CHM) in a vineyard using UAV-based multispectral imaging	doi.org/10.1080/01431161.2016.1226002	2017
Vine vigour modulates bunch microclimate and affects the composition of grape and wine flavonoids: an unmanned aerial vehicle approach in a Sangiovese vineyard in Tuscany	doi.org/10.1111/ajgw.12293	2017
Monitoring vineyards with UAV and multi-sensors for the assessment ofwater stress and grape maturity	doi.org/.doi.org/10.1139/juvs-2016-0024	2017
Daily Mapping of 30 m LAI and NDVI for Grape Yield Prediction in California Vineyards	doi.org/10.3390/rs9040317	2017
Relevance of sink-size estimation for within-field zone delineation in vineyards	doi.org/10.1007/s11119-016-9450-0	2017
Evaluating NIR-Red and NIR-Red edge external filters with digital cameras for assessing vegetation indices under different illumination	doi.org/10.1016/j.infrared.2017.01.007	2017
High Resolution Multispectral and Thermal Remote Sensing-Based Water Stress Assessment in Subsurface Irrigated Grapevines	doi.org/10.3390/rs9090961	2017
Behavior of vegetation/soil indices in shaded and sunlit pixels and evaluation of different shadow compensation methods using UAV high-resolution imagery over vineyards	doi.org/10.1117/12.2305883	2018
Chlorophyll concentration estimation using non-destructive methods in grapes (Vitis vinifera L.) cv	doi.org/10.17584/rcch.2018v12i2.7566	2018
UAV-based multispectral remote sensing for precision agriculture: A comparison between different cameras	doi.org/10.1016/j.isprsjprs.2018.09.008	2018
Using clustering algorithms to segment UAV-based RGB images	doi.org/10.1109/ICA-ACCA.2018.8609822	2018
Estimation of Water Stress in grapevines using proximal and remote sensing methods	doi.org/10.3390/rs10010114	2018
Practical applications of a multisensor UAV platform based on multispectral, thermal and RGB high resolution images in precision viticulture	doi.org/10.3390/agriculture8070116	2018
Canopy Temperature-Based Water Stress Indices: Potential and Limitations	doi.org/10.1007/978-981-13-1861-0_14	2018
Sentinel-2 data analysis and comparison with uav multispectral images for precision viticulture	doi.org/10.1553/GISCIENCE2018_01_S105	2018
Vineyard properties extraction combining UAS-based RGB imagery with elevation data	doi.org/10.1080/01431161.2018.1471548	2018
Multi-temporal vineyard monitoring through UAV-based RGB imagery	doi.org/10.3390/rs10121907	2018
Mapping Cabernet Franc vineyards by unmanned aerial vehicles (UAVs) for variability in vegetation indices, water status, and virus titer	doi.org/10.1051/e3sconf/20185002010	2018
Vineyard water status estimation using multispectral imagery from an UAV platform and machine learning algorithms for irrigation scheduling management	doi.org/10.1016/j.compag.2018.02.013	2018
Multi and hyperspectral UAV remote sensing: Grapevine phylloxera detection in vineyards	doi.org/10.1109/AERO.2018.8396450	2018
Vegetation Indices Combining the Red and Red-Edge Spectral Information for Leaf Area Index Retrieval	doi.org/10.1109/JSTARS.2018.2813281	2018
Assessment of different methods for shadow detection in high-resolution optical imagery and evaluation of shadow impact on calculation of NDVI, and evapotranspiration	doi.org/10.1007/s00271-018-0613-9	2019
Applications of Remote Sensing in Agriculture - A Review	doi.org/10.20546/ijcmas.2019.801.238	2019
On the potentiality of UAV multispectral imagery to detect Flavescence dorée and Grapevine Trunk Diseases	doi.org/10.3390/rs11010023	2019
Quantifying the effect of Jacobiasca lybica pest on vineyards with UAVs by combining geometric and computer vision techniques	doi.org/10.1371/journal.pone.0215521	2019
Comparison of NDVI and NDRE Indices to Detect Differences in Vegetation and Chlorophyll Content	doi.org/10.26782/jmcms.spl.4/2019.11.00003	2019
Radiometric calibration for multispectral camera of different imaging conditions mounted on a UAV platform	doi.org/10.3390/su11040978	2019
Automatic UAV-based detection of Cynodon dactylon for site-specific vineyard management	doi.org/10.1371/journal.pone.0218132	2019
Vineyard variability analysis through UAV-based vigour maps to assess climate change impacts	doi.org/10.3390/agronomy9100581	2019
What relevant information can be identified by experts on unmanned aerial vehicles’ visible images for precision viticulture? Precision Agriculture, 20(2), 278–294	doi.org/10.1007/s11119-019-09634-0	2019
Influence of drone altitude, image overlap, and optical sensor resolution on multi-view reconstruction of forest images	doi.org/10.3390/rs11101252	2019
Use of multispectral and thermal imagery in precision viticulture	doi.org/10.1088/1742-6596/1224/1/012034	2019
Use of a Multirotor-UAV Equipped with a Multispectral Camera to Detect Vineyard Diseases: A Case Study on Barbera and Dolcetto Cultivars	doi.org/10.1007/978-3-030-39299-4_86	2020
Mapping vineyard vigor using airborne remote sensing : relations with yield , berry composition and sanitary status under humid climate conditions	doi.org/10.1007/s11119-019-09663-9	2020
Water Stress Estimation in Vineyards from Aerial SWIR and multispectral UAV data	doi.org/10.3390/RS12152499	2020
Remote sensing in agriculture—accomplishments, limitations, and opportunities	doi.org/10.3390/rs12223783	2020
Individual Grapevine Analysis in a Multi-Temporal Context Using UAV-Based Multi-Sensor Imagery	doi.org/10.3390/rs12010139	2020
State of the art of monitoring technologies and data processing for precision viticulture	doi.org/10.3390/agriculture11030201	2021
Remote and proximal sensing-derived spectral indices and biophysical variables for spatial variation determination in vineyards	doi.org/10.3390/agronomy11040741	2021
Vineyard pruning weight prediction using 3D point clouds generated from UAV imagery and structure from motion photogrammetry	doi.org/10.3390/agronomy11122489	2021
High-resolution drone-acquired RGB imagery to estimate spatial grape quality variability	doi.org/10.3390/agronomy11040655	2021
Estimation of grapevine crop coefficient using a multispectral camera on an unmanned aerial vehicle	doi.org/10.3390/rs13132639	2021
Remote sensing vegetation indices in viticulture: A critical review	doi.org/10.3390/agriculture11050457	2021
Investigating a selection of methods for the prediction of total soluble solids among wine grape quality characteristics using normalized difference vegetation index data from proximal and remote sensing	doi.org/10.3389/fpls.2021.683078	2021
Assessment of vineyard water status by multispectral and RGB imagery obtained from an unmanned aerial vehicle	doi.org/10.5344/ajev.2021.20063	2021
Assessing the Effects of Vineyard Soil Management on Downy and and Powdery Mildew Development	doi.org/10.3390/horticulturae7080209	2021
Advances in unmanned aerial system remote sensing for precision viticulture	doi.org/10.3390/s21030956	2021
Estimation of leaf area index in vineyards by analysing projected shadows using uav imagery	doi.org/10.20870/oeno-one.2021.55.4.4639	2021
Using Aerial Thermal Imagery to Evaluate Water Status in Vitis vinifera cv	doi.org/10.3390/s22208056	2022
Multispectral vineyard segmentation: A deep learning comparison study	doi.org/10.1016/j.compag.2022.106782	2022
Effects of Variable Rate Mechanical Pruning under distinct NDVI ( Normalised Differential Vegetation Index ) levels on the Wine Chemical Composition using the grapevine variety Trincadeira ( Vitis vinifera L .)	(Master's Thesis)	2022
Water Stress Impacts on Grapevines (Vitis vinifera L.) in Hot Environments: Physiological and Spectral Responses	doi.org/10.3390/agronomy12081819	2022
Spectral Comparison of UAV-Based Hyper and Multispectral Cameras for Precision Viticulture	doi.org/10.3390/rs14030449	2022
Utilization of unmanned aerial vehicles for zonal winemaking in cool-climate Riesling vineyards	doi.org/10.20870/oeno-one.2022.56.3.5352	2022
Yield estimations in a vineyard based on high-resolution spatial imagery acquired by a UAV	doi.org/10.1016/j.biosystemseng.2022.10.015	2022
Imagenes multiespectrales en la detección del oídio (Erysiphe necator) en vid ( Vitis vinífera) variedad Thompson Seedless, Titor- Arequipa	(Thesis)	2022
Prediction of the Nitrogen, Phosphorus and Potassium Contents in Grape Leaves at Different Growth Stages Based on UAV Multispectral Remote Sensing	doi.org/10.3390/rs14112659	2022
UAV-based chlorophyll content estimation by evaluating vegetation index responses under different crop coverages	doi.org/10.1016/j.compag.2022.106775	2022
Evaluation of the Use of UAV-Derived Vegetation Indices and Environmental Variables for Grapevine Water Status Monitoring Based on Machine Learning Algorithms and SHAP Analysis	doi.org/10.3390/rs14235918	2022
Effects of Grape Downy Mildew on Photosynthesis of ‘Red Globe’ Grape Leaves under High Temperature Stress	doi.org/10.1080/15538362.2022.2084805	2022
UAV-based individual plant detection and geometric parameter extraction in vineyards	doi.org/10.3389/fpls.2023.1244384	2023
Estimation of crop water stress index and leaf area index based on remote sensing data	doi.org/10.2166/ws.2023.051	2023
Identifying the Optimal Radiometric Calibration Method for UAV-Based Multispectral Imaging	doi.org/10.3390/rs15112909	2023
Assessment of vineyard vigour and yield spatio-temporal variability based on UAV high resolution multispectral images	doi.org/10.1016/j.biosystemseng.2023.06.001	2023
Leaf area index estimation of pergola-trained vineyards in arid regions using classical and deep learning methods based on UAV-based RGB images	doi.org/10.1016/j.compag.2023.107723	2023
Evaluation of the Effect of Water Stress on Clonal Variations of Wine-Quality Responses	doi.org/10.3390/agronomy13020433	2023
Mapping the spatial variability of Botrytis bunch rot risk in vineyards using UAV multispectral imagery	doi.org/10.1016/j.eja.2022.126691	2023
Evaluation of canopy fraction-based vegetation indices, derived from multispectral UAV imagery, to map water status variability in a commercial vineyard	doi.org/10.1007/s00271-023-00907-1	2024
Evaluating the precise grapevine water stress detection using unmanned aerial vehicles and evapotranspiration-based metrics	doi.org/10.1007/s00271-024-00931-9	2024
Application of Unmanned Aerial Vehicle (UAV) Sensing for Water Status Estimation in Vineyards under Different Pruning Strategies	doi.org/10.3390/plants13101350	2024
Potential of a Remotely Piloted Aircraft System with Multispectral and Thermal Sensors to Monitor Vineyard Characteristics for Precision Viticulture	doi.org/10.3390/plants14010137	2025

Comment 4

It would also be useful to present publication trends from 2015 to 2024, in which for example assessing whether the application of NDVI has received less attention in recent years.

Response

Dear Reviewer, thank you for the suggestion. We have added a description of the temporal trend of NDVI-related publications among the articles selected for this review (2015–2024). As shown in the new Figure 5, the number of studies increased from 5 in 2015 to a peak of 14 in 2018, declined slightly in 2019–2020, and resurged in 2021–2022 before gradually decreasing through 2024. This trend confirms that NDVI remains the most widely used index in UAV-based viticulture, while recent research increasingly incorporates complementary indices for specific physiological or management applications. Line 260-270

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

In today's world, collecting high-quality, high-volume, and low-cost information has become a crucial task for researchers. In this regard, unmanned aerial vehicles (UAVs) equipped with multispectral sensors have become powerful tools for monitoring crop variability and supporting precision agriculture, a field of viticulture. Spectral vegetation indices such as NDVI, NDRE, GNDVI, and SAVI are commonly used, and they are widely applied to estimate plant vigor—specifically, grapevines—water status, and canopy structure. Many works are available in the bibliography, but it is important to clearly and reliably establish the structure, importance, and impact of the variables taken into account in a review.
In this regard, I will detail some reviews that I consider important to consider:
2. Methodology of Literature Selection: This should be described more clearly and reliably. In this case, methodological aspects play an important role in establishing a systematic review. It is important to mention the computational tools used, the statistics applied, and other factors.
Is there a reference to Figure 3. Multispectral Image Processing Workflow?
On line 274, there is a typo and citation error: For example, [72] reported that UAV-based …..
It would be advisable to include a conclusion section for this work.

Author Response

Reviewer #4:

In this regard, I will detail some reviews that I consider important to consider:

Methodology of Literature Selection: This should be described more clearly and reliably. In this case, methodological aspects play an important role in establishing a systematic review. It is important to mention the computational tools used, the statistics applied, and other factors.

Is there a reference to Figure 3. Multispectral Image Processing Workflow?

On line 274, there is a typo and citation error: For example, [72] reported that UAV-based …..

It would be advisable to include a conclusion section for this work.

Response

Dear Reviewer, Thank you for your valuable comments and suggestions. The manuscript has been revised to improve clarity and structure. Specifically, the methodology of literature selection has been clarified, detailing the computational tools and statistical approaches used

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The quality of the manuscript has not significantly improved. For detailed comment, please refer to my previous comments. The revised manuscript does not incorporate sufficient changes as stated in the provided revision instructions.

Author Response

Dear Editor and Reviewers,

We would like to express our sincere gratitude for your valuable comments and suggestions, which have significantly contributed to enhancing the clarity, structure, and overall quality of our manuscript. All modifications have been implemented in the revised version of the manuscript and are clearly highlighted in yellow for ease of reference.

Reviewer # 1

Comments # 1

Establishing statistical models based on spectral indices to derive vegetation parameters is a common approach. The authors merely provide a brief introduction to these indices without explaining the underlying mechanisms for estimating vegetation parameters or outlining future development directions.

Response #1

Dear Reviewer,

Thank you for your constructive comments. We have revised the manuscript to better address the mechanistic basis of spectral indices and future development directions, explicitly linking UAV-derived vegetation indices to vine phenology, physiology, and stress responses.

The manuscript now includes a more detailed discussion of how grapevine phenology affects spectral indices across the crop cycle (Lines 58–71), clarifies how statistical models based on vegetation indices estimate key vine parameters such as vigor, pigment content, and water status (Lines 125–130), and expands the description of future approaches integrating advanced statistical and machine learning methods, as well as hybrid frameworks combining multispectral, thermal, and structural data (Lines 133–136).

These revisions explicitly link spectral indices to vine physiological processes, clarify the underlying mechanisms, and highlight practical applications and future developments. We believe these changes fully address your comments and improve the clarity and scientific depth of the manuscript.

Comments # 2

There is no description of using mechanistic models or hybrid models to estimate grape parameters in manuscript.

Response 2

Dear Reviewer, thank you for your insightful comment. We fully agree that UAV-based plant phenotyping represents an important application of remote sensing data, particularly for the calibration and validation of mechanistic models. In the revised manuscript, this aspect has been explicitly incorporated into Section 3.3.3 (Lines 871–876), where we now emphasize that UAV-derived phenotyping provides the high-resolution, spatially explicit information required to parameterize and calibrate process-based models.

Comments # 3

The discussion is overly simplistic and lacks distinctive insights.

Response 3

Dear Reviewer, thank you for your valuable comment regarding the depth and distinctiveness of our discussion. We have carefully revised the manuscript to better highlight the novel contribution of our review.

Although numerous reviews have examined UAV applications and vegetation indices in viticulture, most have focused on vigor zoning, irrigation scheduling, or disease detection without systematically linking spectral information to grape or wine quality. In contrast, our review specifically integrates and critically evaluates the application of UAV-based vegetation indices NDVI, NDRE, SAVI, GNDVI, and CWSI for monitoring grapevine water status, vigor, and maturity, with a particular emphasis on their implications for grape composition and wine quality. By focusing on the connection between widely used multispectral indices and key enological parameters (e.g., sugar content, anthocyanins, phenolics, acidity), our work provides a practical framework for precision viticulture that has not been thoroughly addressed in previous literature (e.g., Sassu et al., 2021; Giovos et al., 2021; Tardaguila et al., 2021).

We believe this revision clarifies the distinctive insights of our review and strengthens the manuscript by explicitly highlighting how accessible remote sensing technologies can be applied to improve grape and wine quality. Please see lines 213–226 for the revised discussion.

Comments # 4

The overall organization of the article is rather disorganized.

Response 4

Dear Reviewer, thank you for this valuable feedback regarding the article's organization. In response to your comment and those from other reviewers, we have carefully revised the structure of the manuscript to improve its logical flow and coherence.

Specifically, we reorganized several sections to provide a clearer progression of ideas. The current structure now begins with an overview of the study context and objectives in the Introduction, followed by sections that address the methodological approach and key aspects related to UAV-based monitoring in viticulture. This reorganization allows the reader to better understand the background before engaging with the technical and analytical content presented in Sections

Finally, the Discussion and Conclusions have been refined to more effectively synthesize the findings and highlight their relevance within the broader framework of precision viticulture.

We believe the changes made in the manuscript have contributed to substantially improve the clarity and readability of this manuscript. We hope this will help the readers to better contextualize the importance of UAV-based technologies. We sincerely appreciate your suggestions, which have help us to strength the organization and overall quality of this review article.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have satisfactorily addressed the suggestions provided in the previous review of the article titled “UAV-based spectral and thermal vegetation indices for precision viticulture: A Review on NDVI, NDRE, SAVI, GNDVI, and CWSI”. The structure, content, and clarity of the text have been consistently improved, demonstrating an appropriate response to the comments.

However, a point-by-point revision is recommended for lines 352-354, where there is an omission of essential information about the drone used (for example, model, technical specifications, or flight configurations). This inclusion is crucial to contextualize the practical application of vegetation indices in the field of precision viticulture and to strengthen the reproducibility of the study.

Author Response

Reviewer # 2

Comments # 1

Response # 1

Dear Reviewer, thank you for your valuable suggestion. Following your comment, we have included a detailed description of the UAV platform and sensor specifications to enhance methodological transparency and reproducibility. The revised manuscript (Lines 526–532) now specifies the drone model (DJI Matrice 210), camera type (MicaSense RedEdge-M), flight altitude (120 m), ground sampling distance (8.2 cm/pixel), image capture rate (1 frame/s), overlap settings (90% front and side), and field of view (47.2°). These parameters contextualize the acquisition conditions and strengthen the interpretation of NDRE-derived vegetation indices within the framework of precision viticulture.

Reviewer 4 Report

Comments and Suggestions for Authors

We thank the authors for the work carried out based on the revisions submitted. I consider that it is a publishable version, subject to previous format adjustments of the journal.

Author Response

Reviewer # 4

Comments #1

We thank the authors for the work carried out based on the revisions submitted. I consider that it is a publishable version, subject to previous format adjustments of the journal.

Response #1

Dear Reviewer, thank you very much for your kind words and for taking the time to evaluate our revised manuscript. We truly appreciate your positive feedback and are delighted that you consider our work suitable for publication. We will make the necessary formatting adjustments according to the journal guidelines and look forward to the final publication.