Analysis of Spatiotemporal Characteristics of Drought in Transboundary Watersheds of Northeast Asia Based on Comprehensive Indices

Round 1

Reviewer 1 Report (New Reviewer)

Comments and Suggestions for Authors

The manuscript titled "Analysis of Spatio-temporal Characteristics of Drought in Transboundary Watersheds of Northeast Asia Based on Comprehensive Indices" provides a detailed investigation into the spatiotemporal patterns of drought using the SPEI and VCI indices. The study combines meteorological and remote sensing data to analyze drought severity, frequency, and its relationship with land cover types in Northeast Asia. The research topic is timely and relevant, given the increasing intensity of droughts under changing climatic conditions. The manuscript demonstrates methodological rigor and provides comprehensive datasets and analyses. However, several areas need improvement to enhance the scientific clarity, coherence, and impact of the paper.

Introduction:

• Clearly specify the research gap being addressed and how this study contributes to advancing drought analysis in transboundary watersheds.
• Include recent studies that are relevant to the topic, such as those involving advanced remote sensing and integrated drought indices. To enhance the references section and update the citations in the manuscript, you can utilize the following studies, which are closely related to drought analysis using satellite data and advanced modeling techniques:

1.      Khosravi, Y., Homayouni, S., & Ouarda, T. B. (2024). Spatio-temporal evaluation of MODIS temperature vegetation dryness index in the Middle East. Ecological Informatics, 84, 102894.
This study evaluates the spatio-temporal patterns of the MODIS Temperature Vegetation Dryness Index (TVDI) in the Middle East and can be useful for comparing methodologies and results with your research.

2.      Khosravi, Y., Homayouni, S., & St-Hilaire, A. (2024). An integrated dryness index based on geographically weighted regression and satellite earth observations. Science of The Total Environment, 911, 168807.
This article introduces a combined dryness index using Geographically Weighted Regression (GWR) and satellite Earth observations, which can serve as a complementary approach to your analyses.

• Incorporate a brief comparison with drought studies in other transboundary basins to situate the research in a broader context.

Materials and Methods:

• Method of Combining SPEI and VCI Indices: The method for integrating the SPEI and VCI indices has not been fully explained. It is unclear whether weighting or another approach was used for the combination.
• Processing of Land Use Data: The explanations regarding the processing of land use/land cover data are limited, and it is not fully clear how the classifications were utilized in the analysis.
• Standardization of Colors and Symbols in Maps and Figures: The colors and symbols used in maps and figures are not standardized, which may confuse readers. Standardized colors and symbols should be applied to better distinguish data in the maps.
• Use of Reference for IDW Explanation: When explaining the IDW method, the following reference should be included for additional context:
  Ali, M. G., Younes, K., Esmaeil, A., & Fatemeh, T. (2011). Assessment of geostatistical methods for spatial analysis of SPI and EDI drought indices. World Applied Sciences Journal, 15(4), 474-482.

Results:

• The results section includes excessive numerical data and detailed descriptions (e.g., specific drought levels across land cover types and regions) that could be condensed into tables or summarized in key points.
• Redundant repetitions of findings, such as seasonal drought trends and land cover-specific drought intensities, reduce readability and impact.
• The figures, such as maps and trend graphs, lack uniformity in design and do not effectively convey key findings. For example, the legends are inconsistent, and the color schemes are not intuitive for distinguishing drought levels.
• Seasonal and spatial variations of drought levels are presented in multiple figures, but the overlap between these representations is not well justified or explained.
• The structure is not cohesive, with results scattered across subsections without a clear flow from one point to another. For instance, spatial distribution results and their significance are not integrated well with temporal trends.
• Some results, such as the integrated drought index maps, are discussed without sufficient detail about their implications or relevance.
• The analysis of spatial and temporal variations in SPEI and VCI indices does not explore the statistical significance of trends sufficiently, leaving some results open to interpretation.
• The reported percentages and areas (e.g., for drought severity in different land cover types) are descriptive but lack deeper interpretation or mechanistic explanations.

Discussion:

• The discussion section predominantly restates the results without engaging in a critical evaluation of the findings or their broader implications.
• Key discrepancies, such as the divergence between SPEI and VCI results in winter, are noted but not adequately explained. For example, the impact of vegetation dormancy on VCI and its potential misrepresentation of winter drought conditions should be discussed in depth.
• The discussion fails to position the findings within the broader scientific literature. Recent studies and comparable research from other regions are not referenced to validate or contrast the results.
• The implications for transboundary water management and drought mitigation are not sufficiently explored.
• Despite the study's focus on transboundary watersheds, the discussion does not elaborate on practical recommendations for policymakers or water resource managers.
• Suggestions for improving drought resilience (e.g., through ecosystem management or improved monitoring systems) are generic and lack specificity.

Author Response

Comments 1: 1. Khosravi, Y., Homayouni, S., & Ouarda, T. B. (2024).Spatio-temporal evaluation of MODIS temperature vegetation dryness index in the Middle East.Ecological Informatics, 84,102894.
This study evaluates the spatio-temporal patterns of the MODIS Temperature Vegetation Dryness Index (TVDI) in the Middle East and can be useful for comparing methodologies and results with your research.
Response 1: Done as suggested. Thank you for the reviewer's comments. For further details, please refer to lines 82 of the manuscript. 

Comments 2: Khosravi, Y., Homayouni, S., & St-Hilaire, A. (2024).An integrated dryness index based on geographically weighted regression and satellite earth observations.Science of The Total Environment, 911,168807.
This article introduces a combined dryness index using Geographically Weighted Regression (GWR) and satellite Earth observations, which can serve as a complementary approach to your analyses.
Response 2: Done as suggested. Thank you for your suggestion. From the beginning, our study focused on using SPEI and VCI as primary indicators to analyze the spatiotemporal characteristics of drought in transboundary watersheds. The recommended articles are very valuable, and we will consider incorporating IGWDI models or other multi-parameter approaches in future research.

Comments 3: Method of Combining SPEI and VCI Indices: The method for integrating the SPEI and VCI indices has not been fully explained. It is unclear whether weighting or another approach was used for the combination.
Response 3: Thank you for your valuable comments. In this study, an equal-weight overlay method was applied. For further details, please refer to lines 347–349 of the manuscript.

Comments 4: Processing of Land Use Data: The explanations regarding the processing of land use/land cover data are limited, and it is not fully clear how the classifications were utilized in the analysis.
Response 4: Done as suggested. Thank you for the reviewer's comments. The revisions have been made. Please refer to lines 212–217 of the manuscript for details.

Comments 5: Standardization of Colors and Symbols in Maps and Figures: The colors and symbols used in maps and figures are not standardized, which may confuse readers. Standardized colors and symbols should be applied to better distinguish data in the maps.
Response 5: Done as suggested. Thank you for the reviewer's comments.

Comments 6: Use of Reference for IDW Explanation: When explaining the IDW method, the following reference should be included for additional context:
Ali, M. G., Younes, K., Esmaeil, A., & Fatemeh, T. (2011). Assessment of geostatistical methods for spatial analysis of SPI and EDI drought indices. World Applied Sciences Journal, 15(4), 474-482.
Response 6: Done as suggested. Thank you for the reviewer's comments. The revisions have been made. Please refer to lines 224 of the manuscript for details.

Comments 7: The results section includes excessive numerical data and detailed descriptions (e.g., specific drought levels across land cover types and regions) that could be condensed into tables or summarized in key points.
Response 7: Done as suggested. Thank you for the reviewer's comments. We have made the necessary revisions accordingly. Please refer to lines 580 to 590 of the manuscript for details.

Comments 8: Redundant repetitions of findings, such as seasonal drought trends and land cover-specific drought intensities, reduce readability and impact.
Response 8: Done as suggested. Thank you for the reviewer's comments. We have made revisions based on their suggestions. Please refer to sections 632–645 of the manuscript for details.

Comments 9: The figures, such as maps and trend graphs, lack uniformity in design and do not effectively convey key findings. For example, the legends are inconsistent, and the color schemes are not intuitive for distinguishing drought levels.
Response 9: Done as suggested. Thank you for the reviewer's comments.

Comments 10: Seasonal and spatial variations of drought levels are presented in multiple figures, but the overlap between these representations is not well justified or explained.
Response 10: Done as suggested. Thank you for the reviewer's comments. We have made the necessary revisions accordingly.

Comments 11: The structure is not cohesive, with results scattered across subsections without a clear flow from one point to another. For instance, spatial distribution results and their significance are not integrated well with temporal trends.
Response 11: Done as suggested. Thank you for the reviewer's comments. The revisions have been made.

Comments 12: Some results, such as the integrated drought index maps, are discussed without sufficient detail about their implications or relevance.
Response 12: Done as suggested. Thank you for the reviewer's comments. We have made revisions according to your suggestions. Please refer to line 519–541 of the manuscript for details. 

Comments 13: The analysis of spatial and temporal variations in SPEI and VCI indices does not explore the statistical significance of trends sufficiently, leaving some results open to interpretation.
Response 13: Done as suggested. Thank you for the reviewer's comments. We have made the necessary revisions accordingly. Please refer to lines 441–457 of the manuscript for details.

Comments 14: The reported percentages and areas (e.g., for drought severity in different land cover types) are descriptive but lack deeper interpretation or mechanistic explanations.
Response 14: Done as suggested. Thank you for the reviewer's comments. We have made modifications to improve the quality of the images. We have made the necessary revisions, please refer to lines 556–576 of the manuscript for details.

Comments 15: The discussion section predominantly restates the results without engaging in a critical evaluation of the findings or their broader implications.
Response 15: Done as suggested. Thank you for the reviewer's comments.

Comments 16: Key discrepancies, such as the divergence between SPEI and VCI results in winter, are noted but not adequately explained. For example, the impact of vegetation dormancy on VCI and its potential misrepresentation of winter drought conditions should be discussed in depth.
Response 16: Done as suggested. Thank you for the reviewer's comments. We have made the necessary revisions, please refer to lines 692–702 of the manuscript for details.

Comments 17: The discussion fails to position the findings within the broader scientific literature. Recent studies and comparable research from other regions are not referenced to validate or contrast the results.
Response 17: Done as suggested. Thank you for the valuable feedback from the reviewer. We have made revisions to include previous research/limitations and future plans. Please refer to lines 655 and 675–676 of the manuscript for details.

Comments 18: The implications for transboundary water management and drought mitigation are not sufficiently explored.
Response 18: Done as suggested. Thank you for the helpful suggestions from the reviewer. We have made revisions to the conclusion section, please refer to lines 772–782 of the manuscript for details.

Comments 19: Despite the study's focus on transboundary watersheds, the discussion does not elaborate on practical recommendations for policymakers or water resource managers.
Response 19: Done as suggested. Thank you for the helpful suggestions from the reviewer. We have made revisions, please refer to lines 746–751 and 772–782 of the manuscript for detailed information.

Comments 20: Suggestions for improving drought resilience (e.g., through ecosystem management or improved monitoring systems) are generic and lack specificity.
Response 20: Done as suggested. Thank you for the helpful suggestions from the reviewer. We have made revisions, please refer to lines 724–729,746–751 and 772–782 of the manuscript for detailed information.

Author Response File: Author Response.pdf

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

Introduction

The authors present an interesting state-of-the-art considering drought indices and techniques, with a focus on recent papers. However, they don’t bring the novelty of their study. The authors should state the scientific question of the paper.

Minor comments

-            Line 96: remove ‘.’ between ‘surveillance’ and ‘[‘, the same typos in line 110

-            Lines 121-124: the authors mentioned ‘several studies’ however they included only one Zhang (2019) [54]; this reference is a master thesis, low visibility.

Materials and Methods

The authors described summary the study area, the collected data, and the will be applied method. However, some concerns raised:

-       The authors don’t indicate if the precipitation data distribution was checked or not? This should be well stated.

-       How the LU classes had been identified? What is the class of ‘water column’?

-       Some the meteorological stations are far from the study area: what is the motivation for their selection?

-       Some LU classes mentioned in the map (Figure 3) are different from those indicated in section 2.2.3 (lines 188-194). Why ‘Water bodies and artificial surfaces’ were not included in the study? The extension of these classes should be a proxy of drought.

-       Lines 163-175: the authors should indicate the threshold for sporadic missing values

Minor improvement

- Figure 2: avoid split word

- Line 225: this is not clear!

- Equation 15 is wrong!

Results

The results are presented as observations with very limited analysis and explanations, despite some analysis had been included in the discussion section. This should be deeply reviewed.

-       The application of SPEI-12 indicates different trends across the regions. This should be explained: what are the drivers of these differences? Since there are many gauges in each region, the results shown in Figure 4 correspond to average or something else.

-       What about other indices (SPEI less or higher than one year): These trends are the same or you have some different behaviors? These indices impact water resources management also.

-       Add the site in figure 5 (X axis)

-       Sub sections of 3.2.1. (Spatial Distribution of Drought Intensity) are weak and should be merged.

-       The legend of figures 6-7 is more extended than the figures themselves: there no drought higher than -0.5 and lower than -1.5.

-       Lines 386-401: how the drought frequency classes were defined?

-       Lines 402-410: The temporal variation of VCI must be discussed: why this kind of trend?

-       Figure 9: Trends between observations are meaningless: it would be better to be removed.

-       I suggest merging section ‘3.3. Temporal Variation of VCI Index’ with section ‘3.4.1. VCI Index Trend Analysis’.

-       Lines 436-444: It should be stated that severe drought occurs in winter (figure 12 and table 3). This needs to be explained.

-       Figure 15 and table 5 show the same data (in a different manner): one of them should be removed.

-       In section ‘3.5. Annual Integrated Drought Index Spatial Distribution’ I guess that drought – LU relationship must be in connection with spatial LU, of different regions of the basin (presented in lines 521-536 and in Figure 17). I mean the drought pattern could be mainly explained by regions more than LU.

Discussion

-        Subsections are not needed.

-        More explanations are needed to discuss winter drought: the study area showed an increase in temperature (Figure 1, line 156) which should impact vegetation growth. However, the authors explain winter VCI drought by low temperature (lines 600-608). This must be reviewed.

-        The field validation is a good idea but, all used VCI values are high (Figure 20): the shown VCI values are higher than 0.76 (except the first one which is not readable for me). This can be an interesting validation for high VCI and not for the low one.

-        Are land management practices (afforestation, soil conservation practices, and conservation tillage) applied in the study area? What is the extension of irrigated land? These aspects are mentioned in the discussion (lines 614-648) but what about the regions – locations where they are implemented? What is their contribution in drought explanation?

References

Some references must be checked:

-            Line 62: check the reference

-            Line 77: add the reference number

-            Line 87: add the reference number

-            Line 118: add the reference number

-            Line 204: check the reference

-            Line 233: check the reference

-            Line 240: check the reference

-            Line 272: check the reference

-            Line 579: check the reference

Author Response

Comments 1: Line 96: remove ‘.’ between ‘surveillance’ and ‘[‘, the same typos in line 110
Response 1: Done as suggested. Thank you for the reviewer's comments. For details, see lines 96 and 110 of the manuscript.

Comments 2: Lines 121-124: the authors mentioned ‘several studies’ however they included only one Zhang (2019) [54]; this reference is a master thesis, low visibility.
Response 2: Done as suggested. Thank you for the reviewer's comments. For further details, please refer to lines 121–132 of the manuscript.

Comments 3: The authors don’t indicate if the precipitation data distribution was checked or not? This should be well stated.
Response 3: Done as suggested. Thank you for the reviewer's comments. For further details, please refer to lines 185–190 of the manuscript.

Comments 4: How the LU classes had been identified? What is the class of ‘water column’? 
Response 4: Done as suggested. Thank you for the reviewer's comments. The land-use data and classification used in this study refer to the GlobeLand30 dataset, which was developed by the team of the National High Technology Research and Development Program of China (863 Program) for Global Land Cover Mapping and Key Technology Research. It contains more comprehensive and detailed information on the spatial distribution of global land cover, and can better depict the landscape patterns formed by most human land-use activities. Additionally, I apologize for the misstatement regarding "water column"; it should be "water bodies". This has been corrected. For further details, please refer to lines 213–217 of the manuscript.

Comments 5: Some the meteorological stations are far from the study area: what is the motivation for their selection?
Response 5: Done as suggested. Thank you for the valuable feedback from the reviewer. For further details, please refer to lines 178–180 of the manuscript. The rationale for choosing stations that are somewhat distant from the study area is based on the necessity to perform spatial interpolation of meteorological data. To ensure comprehensive coverage of the entire study region and to maintain the authenticity of the interpolated precipitation data, we included stations located outside the immediate boundaries of the study area in our interpolation process. This approach helps to enhance the accuracy and reliability of the meteorological data across the entire study domain.

Comments 6: Some LU classes mentioned in the map (Figure 3) are different from those indicated in section 2.2.3 (lines 188-194). Why ‘Water bodies and artificial surfaces’ were not included in the study? The extension of these classes should be a proxy of drought.
Response 6: Done as suggested. Thank you for the reviewer's comments. See line 211–217 of the manuscript for details. The main research object of this paper is vegetation, so we use remote sensing and meteorological data for analysis, without considering water and man-made surface. This is also the limitation of this study.

Comments 7: Lines 163-175: the authors should indicate the threshold for sporadic missing values.
Response 7: Thank you for the reviewer's comments. There is no threshold for the missing value. The missing value is a null value because the platform data is incomplete due to some force majeure factors.

Comments 8: Figure 2: avoid split word 
Response 8: Done as suggested. Thank you for the reviewer's comments. See Figure 2 of the manuscript for details.

Comments 9: Line 225: this is not clear!
Done as suggested. Thank you for the reviewer's comments. See line 250–251 of the manuscript for details.

Comments 10: Equation 15 is wrong!
Response 10: Done as suggested. Thank you for the reviewer's comments. See the modified formula 15 for details.

Comments 11: The application of SPEI-12 indicates different trends across the regions. This should be explained: what are the drivers of these differences? Since there are many gauges in each region, the results shown in Figure 4 correspond to average or something else.
Response 11: Done as suggested. Thank you for the reviewer's comments. See line 373–375 of the manuscript for details. We agree with you. In the follow-up study, we plan to explore the driving factors behind these trends.

Comments 12: What about other indices (SPEI less or higher than one year): These trends are the same or you have some different behaviors? These indices impact water resources management also.
Response 12: Thank you for the valuable feedback from the reviewer. See line 356–368 of the manuscript for details.

Comments 13: Add the site in figure 5 (X axis)
Response 13: Done as suggested. Thank you for the reviewer's comments. See Figure 6 of the manuscript for details.

Comments 14: Sub sections of 3.2.1. (Spatial Distribution of Drought Intensity) are weak and should be merged.
Response 14: Done as suggested. Thank you for the reviewer's comments. See line 396–422 of the manuscript for details.

Comments 15: The legend of figures 6-7 is more extended than the figures themselves: there no drought higher than -0.5 and lower than -1.5. The legend of figures 6-7 is more extended than the figures themselves: there no drought higher than -0.5 and lower than -1.5.
Response 15: Thank you for your valuable comments. As for the legend in Figure 6-7, our classification is based on the above criteria for Drought Grades (see Table 1). Since these grades are classified according to the above criteria, the legend cannot be easily modified.

Comments 16: Lines 386-401: how the drought frequency classes were defined?
Response 16: Done as suggested. Thank you for the reviewer's comments. See line 424–427 of the manuscript for details.

Comments 17: Lines 402-410: The temporal variation of VCI must be discussed: why this kind of trend? 
Response 17: Done as suggested. Thank you for the reviewer's comments. See line 674–688 of the manuscript for details.

Comments 18: Figure 9: Trends between observations are meaningless: it would be better to be removed.
Response 18: Thank you for your valuable comments. I believe that studying different seasons is necessary, as each season has a different contribution. Additionally, the seasonal and interannual variations of the VCI index have been discussed in detail in the manuscript, specifically in lines 678–688.

Comments 19: I suggest merging section ‘3.3. Temporal Variation of VCI Index’ with section ‘3.4.1. VCI Index Trend Analysis’.
Response 19: Thank you for your valuable comments. The two sections cannot be merged, as one describes temporal changes and the other describes spatial changes. To express this more clearly, the title of section 3.4.1 should be changed to "Spatial trend analysis of VCI index," and the title of section 3.3 should be changed to "Temporal Variation Trend of the VCI Index."

Comments 20: Lines 436-444: It should be stated that severe drought occurs in winter (figure 12 and table 3). This needs to be explained.
Response 20: Done as suggested. Thank you for the reviewer's comments. This section has been elaborated in detail during the discussion, please refer to lines 692–702 of the manuscript for details.

Comments 21: Figure 15 and table 5 show the same data (in a different manner): one of them should be removed.
Response 21: Done as suggested. Thank you for the reviewer's comments.

Comments 22: In section ‘3.5. Annual Integrated Drought Index Spatial Distribution’ I guess that drought – LU relationship must be in connection with spatial LU, of different regions of the basin (presented in lines 521-536 and in Figure 17). I mean the drought pattern could be mainly explained by regions more than LU. 
Response 22: Done as suggested. Thank you for the reviewer's comments. This is something I didn't consider. We analyzed it after listening to your suggestions. See line 730–743 in the discussion section of 4.3 for details.

Comments 23: Subsections are not needed.
Response 23: Thank you for your valuable feedback. I believe that dividing the discussion into subsections helps make different viewpoints and content clearer, making it easier for readers to understand.

Comments 24: More explanations are needed to discuss winter drought: the study area showed an increase in temperature (Figure 1, line 156) which should impact vegetation growth. However, the authors explain winter VCI drought by low temperature (lines 600-608). This must be reviewed. 
Response 24: Done as suggested. Thank you for your valuable feedback. The study area is located in a mid-to-high latitude region, where low temperatures during winter cause vegetation to enter dormancy and wither. According to the VCI index, winter severely limits vegetation growth, resulting in low VCI values. In remote sensing observations, this dormancy often appears as red areas, reflecting deteriorating vegetation conditions. While this may resemble drought conditions, it is primarily due to low temperatures and seasonal vegetation dormancy, and does not necessarily indicate severe meteorological drought. Please refer to line 692–702 of the manuscript for further details.

Comments 25: The field validation is a good idea but, all used VCI values are high (Figure 20): the shown VCI values are higher than 0.76 (except the first one which is not readable for me). This can be an interesting validation for high VCI and not for the low one.
Response 25: Done as suggested. Thank you for your valuable feedback. See Figure 19 of the manuscript for details.

Comments 26: Are land management practices (afforestation, soil conservation practices, and conservation tillage) applied in the study area? What is the extension of irrigated land? These aspects are mentioned in the discussion (lines 614-648) but what about the regions – locations where they are implemented? What is their contribution in drought explanation? 
Response 26: Done as suggested. Thank you for your valuable comments. In response to your suggestion, I have clarified the implementation of land management practices in the study area, particularly in the Heilongjiang and URB regions. Measures such as afforestation, soil conservation practices, and conservation tillage are indeed applied in these regions to improve soil moisture retention, reduce soil erosion, and enhance drought resilience. These practices play a significant role in mitigating drought impacts, especially in forested and agricultural areas that are vulnerable to water scarcity. Please refer to line 721–729 of the manuscript for further details.

Comments 27: Line 62: check the reference 
Response 27: Done as suggested. Thank you for your valuable feedback.

Comments 28: Line 77: add the reference number
Response 28: Done as suggested. Thank you for your valuable feedback.

Comments 29: Line 87: add the reference number
Response 29: Thank you for the reviewer's comments. The 45th reference is.

Comments 30: Line 118: add the reference number 
Response 30: Thank you for your feedback. We have added references, please refer to the 53th reference for details.

Comments 31: Line 204: check the reference
Response 31: Done as suggested. Thank you for your valuable feedback.

Comments 32: Line 233: check the reference
Response 32: Done as suggested. Thank you for your valuable feedback.

Comments 33: Line 240: check the reference
Response 33: Done as suggested. Thank you for your valuable feedback.

Comments 34: Line 272: check the reference
Response 34: Done as suggested. Thank you for your valuable feedback.

Comments 35: Line 579: check the reference
Response 35: Done as suggested. Thank you for your valuable feedback.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report (New Reviewer)

Comments and Suggestions for Authors

I have reviewed the revised version of the manuscript titled "Analysis of Spatio-temporal Characteristics of Drought in Transboundary Watersheds of Northeast Asia Based on com-prehensive Indices" and carefully assessed the authors' responses to the reviewers' comments. The authors have adequately addressed all the concerns and suggestions raised during the review process, and the revisions have significantly improved the quality of the manuscript.

In my opinion, the manuscript is now suitable for publication without further revisions.

Author Response

Thank you for your positive feedback and for taking the time to review the revised manuscript. We appreciate your thoughtful evaluation of the changes made, and we're glad to hear that the revisions have addressed your concerns. We are pleased that the manuscript is now suitable for publication.

Reviewer 2 Report (New Reviewer)

Comments and Suggestions for Authors

The authors improved deeply the manuscript and consider the main comments. They provide additional information to better understand the findings and their explanations. I think the revised paper is more suitable. However, the authors omitted some comments or were not convincing in others. Following some remaining concerns:

-The first comment (in the introduction section: The authors present an interesting state-of-the-art considering drought indices and techniques, with a focus on recent papers. However, they don’t bring the novelty of their study. The authors should state the scientific question of the paper) was not considered in the responses: ‘The authors present an interesting state-of-the-art considering drought indices and techniques, with a focus on recent papers. However, they don’t bring the novelty of their study. The authors should state the scientific question of the paper.’

- The response to the first comment related to section Materials and Methods (The authors don’t indicate if the precipitation data distribution was checked or not? This should be well stated). Thank you for the additional information provided by the authors. Indeed, they might indicate if the used gauges meet the gamma distribution, before applying SPI and SPEI.

- For the last comment of the same section (The authors should indicate the threshold for sporadic missing values), the authors provide additional information, to better understand how the missing values had been filled. However, the meaning of this comment is: For this kind of analysis, a threshold must be adopted to drop gauges with high missing values (more than 10 or 15%) otherwise the gauge must be dropped.’ In this version, the author mentioned, if I understand well, that missing data were interpolated but this is not possible for long missing values. Also, the missing values should be not zero but there is recorded data.

- For figures 6 and 7: Just remove the non available classes in your study area from the legend, so easy!

Author Response

Comments 1: The first comment (in the introduction section: The authors present an interesting state-of-the-art considering drought indices and techniques, with a focus on recent papers. However, they don’t bring the novelty of their study. The authors should state the scientific question of the paper) was not considered in the responses: ‘The authors present an interesting state-of-the-art considering drought indices and techniques, with a focus on recent papers. However, they don’t bring the novelty of their study. The authors should state the scientific question of the paper.’

Response 1: Done as suggested. Thank you for the reviewer's comments. Based on their suggestions, we have made revisions. Please refer to lines 135–141 of the manuscript for details.

Comments 2: The response to the first comment related to section Materials and Methods (The authors don’t indicate if the precipitation data distribution was checked or not? This should be well stated). Thank you for the additional information provided by the authors. Indeed, they might indicate if the used gauges meet the gamma distribution, before applying SPI and SPEI.

Response 2: Done as suggested. Thank you for the constructive feedback from the reviewer. We have made the necessary revisions, please refer to lines 193–196 of the manuscript for details.

Comments 3: For the last comment of the same section (The authors should indicate the threshold for sporadic missing values), the authors provide additional information, to better understand how the missing values had been filled. However, the meaning of this comment is: For this kind of analysis, a threshold must be adopted to drop gauges with high missing values (more than 10 or 15%) otherwise the gauge must be dropped.’ In this version, the author mentioned, if I understand well, that missing data were interpolated but this is not possible for long missing values. Also, the missing values should be not zero but there is recorded data.

Response 3: Done as suggested. Thank you for the reviewer's comments. For further details, please refer to lines 183–187 of the manuscript.

Comments 4: For figures 6 and 7: Just remove the non available classes in your study area from the legend, so easy!

Response 4: Done as suggested. Thank you for the reviewer's comments. Please refer to Figure 7 in the manuscript for details.

Author Response File: Author Response.pdf

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

Dear Authors,

The article is like a case study without any novelty or new contribution to the academic field. The authors employed SPEI and VCI with classical trend analysis methods, which are well-documented and can be found in several articles worldwide. Given the lack of originality and significant contribution, the article cannot be accepted in high-quality journals like Water, which holds a Q1 ranking. I believe that the article must be rejected in its current form.

Major comments:

·       The percent match according to the iThenticate report is about 24%, which exceeds the acceptable limit; please revise and modify the article accordingly.

·       The methodology section lacks the appropriate references, with several equations and statements requiring proper citation. To ensure academic evaluation, the authors must include references for all equations and statements.

·       In the title, the authors mentioned “composite indices”, what did you mean by composite? Actually, the authors conducted SPEI and VCI.

·       The abstract should include quantified results as part of the main takeaways rather than just discussing the conclusions in qualitative terms.

·       The introduction is insufficient and fails to adequately establish a comprehensive background on the subject. It lacks a clear presentation of the significance and relevance of the topic, which is crucial for engaging the readers about the importance of the research. Also, its structure and flow are poor.

·       The quality of all figures is poor, and one can not be read (For example, Fig. 1). In Figure 1, the meteorological station symbol is very small, and the precipitation figures are of low quality. All figures must be revised and improved (Figures 1, 2, 3, 8, 13, 14, 15, and 17).

·       How did you check the quality of historical observation data? This point must be expanded.

·       Each of the used data has a different spatial resolution; how did you overcome and address this issue?

·       Table 1: The reference is wrong; the authors must use the original drought classification reference.

·       Simplify the Run theory in the methodology section. It only depends on different thresholds, which is preferably 0.

·       A flowchart must be added to the methodology section.

·       Table 2: the reference must be added to this table.

·       For the Sen slope and Mann-Kendall methods, the appropriate references must be added for each statement and equation.

·        Figures 3 – 5: All of these Figures are unclear and must be improved in terms of quality, results, and caption.

·       Lines 374 – 388: This paragraph must be revised.

·       In Figure 15, nothing can be observed.

• The discussion section must be revised and expanded to cover more previous studies, limitations, and future implications.

·       The conclusion section should be more concise, providing a brief summary of the article followed by the key findings listed point by point. Kindly revise the conclusion accordingly to streamline the presentation and enhance the clarity of the main outcomes.

Sincerely,

Comments on the Quality of English Language

Moderate editing of English language required.

Reviewer 2 Report

Comments and Suggestions for Authors

Q1. In the abstract, could the authors numerically label the findings, e.g., (1), (2), etc.?

Q2. L285-95: Several points within these paragraphs need improvement, as follows: The current findings are very brief and unclear. Could the authors revisit and expand on them (see L20-23)? Additionally, please provide 1-2 sentences explaining how these findings contribute to the current literature or support stakeholders/authorities in disaster management and prevention.

Q3. In several sections, the authors demonstrate redundancy in writing. For example, the phrase “extreme climate events are becoming increasingly frequent” and the similar phrase “extreme weather events such as droughts and floods have become increasingly frequent” repeat the same idea unnecessarily.

Q4. L31-33: The sentence “its formation is influenced by multiple factors, including exposure, vulnerability, and differences in drought resistance capabilities” is vague. Specifically, the phrase “differences in drought resistance” is unclear and needs further explanation.

Q5. L52-54: The sentence "Recognizing limitations in single meteorological drought indices like SPI and comprehensive indices like PDSI, which led to the development of the SPEI index by integrating the strengths of both in 2010" is grammatically incomplete and needs restructuring for clarity.

Q6. L76-77: The objective to “...gain a more comprehensive and in-depth understanding of the impacts of drought on ecosystems and water resources” is too general. It lacks specificity about what “more comprehensive” entails. Please revisit this statement.

Q7. In this work, the integration of SPEI and VCI indices is presented as the main method, but the authors do not clearly indicate why this combination is novel or what gap it fills in the current literature. Additionally, the statement “provide more scientific and effective support for drought monitoring, early warning, and management” is vague without clearly stating what kind of support or advancements are expected. Please improve by adding more specific details.

Q8. Some statements lack references. For example: “…with drought being a particularly severe natural disaster globally (suggest 10.1016/j.scitotenv.2024.174289).” “These characteristics make drought a challenging natural disaster to predict, mitigate, and manage effectively (suggest 10.3850/978-90-833476-1-5_iahr40wc-p1339-cd).” “The occurrence of drought can be categorized into meteorological (suggest 10.1016/j.jenvman.2024.121375), agricultural, hydrological (10.1007/s00382-024-07319-7), and socioeconomic.”

Q9. More effective methods have been introduced in recent years. For example, why was the original Mann-Kendall method used instead of the modified version? Please provide more justification with evidence from previous works, such as 10.3390/w14213367.

Q10. The results are based on data from 58 stations, which may not sufficiently cover the spatial variability of drought across such a large and diverse region. The authors used Inverse Distance Weighting (IDW) interpolation but did not explain why this method was chosen over alternatives. Please clarify.

Q11. While the Vegetation Condition Index (VCI) is used, there is no comprehensive validation against ground-based observations. This may affect the reliability of the spatial drought estimates. Please revisit and address this point.

Q12. The authors mention that seasonal drought levels are primarily associated with seasonal vegetation wilting, but there is no detailed analysis of how different vegetation types respond to drought during different seasons. This weakens the explanation of VCI-based seasonal drought variation. Please revisit and address this point.

Q13. There is no clear discussion of the uncertainties associated with using the SPEI and VCI indices. A more rigorous assessment of the confidence in the drought metrics is suggested, which would strengthen the scientific validity of the findings.

Q14. Please include a section discussing the study’s limitations and suggestions for future work.

Q15. Please separate the findings into (1), (2), etc., for clarity.