GIS-Based Assessment of Stormwater Harvesting Potentials: A Sustainable Approach to Alleviate Water Scarcity in Rwanda’s Eastern Savanna Agroecological Zone

Comments 1: [General results should be given in the abstract.]

Response 1: [General results were added in the abstract.] Thank you for pointing this out. I/We agree with this comment. Therefore, I/we have….[We added the general finding in the abstract as it was advised. – page number 1, paragraph (abstract), and lines 18 to 22.]

“[Findings include eight main water reservoirs, each with a unique code (W_R1 to W_R8), geographic coordinates (X and Y), and storage volumes in 10 million cubic meters. W_R1 has the smallest volume at 0.242 × 10⁶ m³, while W_R2 has the largest volume at 8.51 × 10⁶ m³. W_R3, W_R5, and W_R7 are additional noteworthy reservoirs with sizable capacities.”

Comments 2: [Please check the format of the journal. Figure and table numbers must be consecutive (1,2 ..), not multilevel.]

Response 2: Thank you for pointing this out. We have carefully reviewed and revised all figure and table numbers to follow a consecutive numbering format, as per the journal's guidelines. All references in the main text have also been updated accordingly. I have attached the table and the figure below for your reference

Figure 1: Eastern Savanna Agroecological Zone

Table 3: Weighted overlay analysis integrating the key features[54]

Comments 3: [Page 6, lines 209-210: The quality and resolution of the dataset are mentioned. However, the quality of the dataset is not given. What is the quality of each data used in the analysis? The vertical and horizontal errors of DEM and the accuracy of the LULC and soil map should be given.]

Response 3: Agree. I/We have, accordingly modified to emphasize this point.

Thank you for your valuable observation. We have updated the manuscript to provide detailed information on the quality and accuracy of the datasets used in the analysis:

• The Digital Elevation Model (DEM) used is at a 10 m spatial resolution, sourced from Rwanda Land Management Authority. The reported vertical Root Mean Square Error (RMSE) is approximately ±5 to 10 m, while horizontal accuracy is typically within ±10–15 m, depending on terrain complexity (e.g., Tadono et al., 2014).
• The Land Use/Land Cover (LULC) dataset employed has a 90 m spatial resolution. While the exact source is not fully documented, datasets at this resolution generally exhibit an overall classification accuracy ranging between 70% and 85% (Chen et al., 2015; Gong et al., 2013), sufficient for regional hydrological modeling.

These details have been incorporated into the data description section of the revised manuscript to enhance transparency regarding data quality.

Comments 4: [The source of each data and its acquisition time should also be given for the data provided in Table 2.]

Response 4: Agree. I/We have, accordingly, done/revised/changed/modified…..to emphasize this point.

Comments 5: [Section 2.2: The Soil Conservation Service Curve Number (SCS-CN) method is given in this section but not mentioned in the introduction part. Why this method is preferred should be given with relevant literature. Isn't it developed for specific geographic conditions?]

Response 5: Thank you for this valuable observation. In response, we have revised the introduction to include a clear justification for the selection of the SCS-CN method, supported by relevant literature. The SCS-CN method was originally developed by the USDA Soil Conservation Service (now NRCS) for use in small agricultural watersheds in the United States. However, due to its simplicity, minimal data requirements, and adaptability to diverse land use and soil conditions, it has been widely applied in hydrological studies across various geographic regions, including tropical and semi-arid climates. We have clarified in the introduction that the SCS-CN method is suitable for data-scarce regions like ours, where high-resolution, long-term rainfall and streamflow data are often unavailable. Its ability to integrate land use/land cover (LULC), hydrologic soil groups (HSG), and antecedent moisture conditions makes it especially valuable for estimating surface runoff in GIS-based watershed studies

“[The integration of Geographic Information Systems (GIS) with the Soil Conservation Service Curve Number (SCS-CN) method has become a widely accepted approach for estimating spatial runoff, particularly in data-scarce and hydrologically diverse regions [5] [6]. The SCS-CN model estimates direct runoff from rainfall using a Curve Number (CN) derived from land use/land cover (LULC), hydrologic soil groups (HSGs), and antecedent moisture conditions [7];[8] GIS facilitates the integration of LULC and soil data to derive spatially distributed CN values used in hydrological modeling" [9].

Digital Elevation Models (DEMs) support watershed delineation and stream network extraction [10];[11], while classified satellite imagery provides LULC data [12];[13]. Soil maps from sources like FAO and Soil Grids offer HSG classifications for CN estimation [14]. Numerous studies demonstrate the effectiveness of this method in diverse environments: [15] in Palestine, Meena et al. (2013) in India, Mutua & Klik (2006) in Kenya, and Nizeyimana et al. (2021) in Rwanda. Its application supports land and water resource planning, stormwater harvesting, and climate adaptation strategies by evaluating how land use changes influence runoff generation [16],[17].

Overall, the GIS–SCS-CN method provides a cost-effective and scalable tool for hydrological modeling in regions with limited hydrometric data, enhancing spatial decision-making in watershed and stormwater management [13];[14]]”

Comments 6: [Why is 90-m LULC data used? There are higher-resolution global datasets available in the literature.]

Response 6: Thank you for this important comment. The choice of the 90-meter resolution LULC dataset was primarily guided by the need to ensure consistency and compatibility with other spatial datasets used in this study

While we acknowledge the availability of higher-resolution LULC datasets such as ESA World Cover (10 m) at the time of data preparation, the 90 m dataset we used (sourced from Rwanda Land Management Authority) was selected due to its complete regional coverage, prior validation in similar studies, and format compatibility with the tools and resolution constraints of the watershed-scale SCS-CN model.

Nevertheless, we recognize that higher-resolution LULC data can improve spatial accuracy. We have added a note in the revised manuscript’s limitations section suggesting that future studies may benefit from integrating finer-resolution datasets where available and appropriate.

Comments 7: [What are the properties of B, C, and D given in Figure 3.1.b, the Hydrological soil group map?]

Response 7: 

Figure 3(b) in the revised manuscript

Comments 8: [Figure 3.2: Provide the units of the maps.]

Response 8: Thank you for your comment. We have reviewed all the maps included in the manuscript and updated the figure captions and legends

“[Figure 4: study area characteristics: (a) rainfall (mm), (b) land use land cover]”

Comments 9: [Figure 3.2.b: The last color of the legend does not have a label..]

Response 9: Agree. I/We have, accordingly, modified to emphasize this point.

Figure 5: Land cover land Use was modified in the revised manuscript

Comments 10: [Section 4.2: The spatial and temporal distribution of rainfall is mentioned. However, only one map is provided. The authors did not present the temporal variation of the rainfall. Between what dates was rainfall data collected? The conclusion part also mentioned the erratic rainfall, but it is not presented in the paper.]

Response 10: Agree. I/We have, accordingly, revised to emphasize this point.

In the Abstract: Water scarcity remains critical in Rwanda's Eastern Savanna Agroecological Zone due to erratic rainfall, prolonged dry seasons, and rising water demands. This challenge threatens agricultural productivity, food security, and livelihoods. Stormwater harvesting presents a sustainable solution that increases water availability and mitigates the impacts of climate variability. This study utilizes Geographic Information System (GIS) tools and SCS-CN to assess stormwater harvesting potential in the region. The methodology includes analyzing land use, soil type, rainfall data (30 years, from 1994 to 2023), and topography. Key research steps involve delineating catchment areas, estimating runoff volumes, and selecting optimal storage sites using multi-criteria decision analysis. Findings include eight main water reservoirs, each with a unique code (W_R1 to W_R8), geographic coordinates (X and Y), and 10 million cubic meters storage volumes. W_R1 has the smallest volume at 0.242 × 10⁶ m³, while W_R2 has the largest volume at 8.51 × 10⁶ m³. W_R3, W_R5, and W_R7 are additional noteworthy reservoirs with sizable capacities. The findings contribute to policy formulation and Sustainable Development Goals (SDGs) related to clean water, food security, and climate action. This research provides a replicable framework for addressing water scarcity and enhancing long-term resilience in water-stressed regions.

Comments 11: [There are more results in the discussion section, and no discussion has been prepared by referring to the relevant literature. Therefore, this section should be expanded and the contribution of the study to the literature should be emphasized. As the method is old, many studies used the same method. In addition, the limits should be stated. For example, was 90 m LULC sufficient? Would more detailed results have been obtained if there were other data sources?]

Response 11: Thank you for your valuable feedback. We agree that the Discussion section would benefit from a more thorough integration of relevant literature. Accordingly, we have expanded this section to include comparisons with previous studies that applied the SCS-CN method, thereby better situating our findings within the existing body of research and highlighting the unique contributions of our study.

Regarding the limitations, we have added a detailed discussion on the use of 90 m resolution LULC data. We acknowledge that while this resolution provides a good balance between coverage and detail, higher-resolution datasets could potentially offer more precise spatial analysis. This limitation and its possible impact on the results are now explicitly stated, along with suggestions for future studies to explore finer-scale data sources.

We believe these revisions strengthen the manuscript and improve its clarity and scientific rigor.

4. Response to Comments on the Quality of English Language

Point 1:

Response 1:  (We appreciate the reviewer’s observation regarding the language quality. The manuscript has been carefully reviewed and thoroughly edited to improve clarity, grammar, and overall readability. We have also utilized professional language editing services to ensure that the manuscript meets the high standards of the journal.)

5. Additional clarifications

[The manuscript has been thoroughly revised to address the reviewers’ comments and to align with the journal’s requirements. All suggested improvements, including expansion of the Discussion section with relevant literature, clear articulation of study limitations, and clarification of methodological choices, have been incorporated. We believe the revised version is now significantly strengthened and meets the standards expected for publication

Comments 1: [Abstract should contain background, objective, methodology, important hard results, and significance of studies; important hard results are missing from your abstract. Line 16-18: “Integrating spatial analysis and hydrological modeling identifies and prioritizes suitable water collection and storage sites.”, can please rephrase; not sure what you meant here.]

Response 1: Thank you for your insightful feedback regarding the structure and content of the abstract. In response, we have revised the abstract to clearly include all essential elements: background, research objectives, methodology, key quantitative results, and the significance of the study.

The sentence on Lines 16–18 has also been rephrased for clarity. Specifically, we have clarified the role of spatial analysis and hydrological modeling in the identification and prioritization of potential stormwater harvesting sites. We believe these revisions improve the impact and clarity of the abstract.

Comments 2: [- Can you please explicitly state the novelty and contribution of the paper, at the end of Section: Introduction. - Please do not sub-sections your Introduction. - It seems that from the start of this section, only stormwater harvesting is considered. Can you do literature review of different alternatives, e.g. rainwater harvesting, fog harvesting, wastewater recycling, etc. Please look at the introduction of this paper: “Techno-Economic and Sensitivity Analysis of Rainwater Harvesting System as Alternative Water Source”, Sustainability 2019. Lines 41: Please give reference to Inamdar et al. at the end of the sentence. Line 49-50: Reference 3 is not Jiang et al. Line 105: It is uncommon to have multiple sub-sections in an Introduction of a journal paper; I suggest you remove the sub-sections, and write narratively. Line 137: The scope of the study can be discussed in Material and Methods section. Line 150: Can please provide structure of the paper..]

Response 2: We thank the reviewer for the detailed and constructive feedback on the Introduction section. In response, we have made the following revisions:

1. Novelty and Contribution:
   At the end of the Introduction, we have explicitly stated the novelty and main contributions of the study. The manuscript now clearly highlights that it is among the first to apply an integrated GIS and SCS-CN-based assessment for stormwater harvesting potential in Rwanda’s Eastern Savanna Agroecological Zone—a region largely underrepresented in existing literature. The study offers a spatially explicit decision-support framework aligned with national development goals and SDGs.
2. Avoiding Sub-sections in the Introduction:
   We have restructured the Introduction into a cohesive narrative, removing all sub-section headings to align with conventional academic writing practices for journal articles.
3. Expanded Literature Review:
   To provide a broader context, we have expanded the literature review to briefly discuss various alternative water supply options such as rainwater harvesting, fog harvesting, and wastewater reuse. This helps position stormwater harvesting as one component of a wider set of sustainable water management strategies. We have drawn inspiration from the introduction of the paper “Techno-Economic and Sensitivity Analysis of Rainwater Harvesting System as Alternative Water Source” (Sustainability, 2019), as suggested.
4. Reference Corrections:
• Line 41: A proper citation for Inamdar et al. has been added.
• Line 49–50: The incorrect attribution of Reference 3 has been corrected, and the reference list has been updated accordingly.
5. Scope of the Study:
   As per the reviewer’s suggestion, the discussion of the scope of the study has been moved to the Materials and Methods section to improve logical flow.
6. Paper Structure (Line 150):
   We have now added a brief paragraph at the end of the Introduction outlining the structure of the manuscript, guiding the reader on what to expect in each section.

We trust these revisions address your concerns and improve the coherence, completeness, and academic rigor of the manuscript.

Comments 3: Material and Methods General comment: You need to work on making your methodology clearer; may I suggest you start off with the summary Figure 2.1; to give holistic understandings of your method, before going to specific of each step. This is important for reproducibility. Line 154: Can correct the sub-section 1.1.1 to 2.1.1 Line 155: Is this the region where your study is considered? Can you please explicitly state that is the case. Figure 1.1: Generally, for figures and tables, they need to be introduced, described and their significance discussed. The manuscript does not contain any reference to Figure 1.1. Additionally, please use Figure 1, 2, etc.; Line 161: What is RESAEZ? Please do not abbreviate first use. Table 2.1: Similar to my comment above; Somehow, your table is a bit mixed up with the numbering. Generally, for figures and tables, they need to be introduced, described and their significance discussed. The manuscript does not contain any reference to Table 2.1. Additionally, please refer to Table 1, 2, etc. At the moment, you are following thesis conventions. Line 199:201: Before talking about integrating SCS-N and GIS, can you please give some information on GIS and SCS-N, especially on how the literatures have been using the tools for stormwater management. Line 209: Can give reference to DSRM? Also, please do no abbreviate on first use (RLMA, DSRM, etc.) Line 228: You mentioned that with calculated CN values, runoff can be estimated? 1) how did you calculate CN values, 2) CN value is not a variable in calculating runoff in equation (1)? Line 234: Abstraction ratio is typically 0.2; care to provide reference? Line 239: Please do not abbreviate HSG on first use. Table 2.2: Please rename the table to Table 2 only! Also, what is your basis of assigning Weight of Runoff coefficient and slope and 60 and 40, respectively?

Response 3:

We sincerely thank the reviewer for the detailed and insightful comments on the Materials and Methods section. We have carefully addressed each concern to improve the clarity, transparency, and reproducibility of our methodology. Below are our specific responses and the corresponding revisions made:

1. General Structure and Summary Diagram:
• Comment: Suggestion to include a summary figure (e.g., Figure 2.1) to provide a holistic view of the methodology before going into specifics.
• Response: We have now added a new summary figure at the beginning of the Materials and Methods section (now titled Figure 1) to visually summarize the overall methodological framework. This diagram outlines the main steps from data collection to final analysis and site selection, enhancing reader understanding and reproducibility.
2. Sub-section Correction (Line 154):
• Response: The numbering of the sub-sections has been corrected throughout the manuscript (e.g., changed from “1.1.1” to “2.1.1”) to follow journal conventions.
3. Study Area Clarification (Line 155):
• Response: We have revised this section to explicitly state that the described region (Rwanda’s Eastern Savanna Agroecological Zone) is indeed the area considered in this study.
4. Figure 1.1 Reference and Description:
• Comment: The figure was not introduced or discussed in the text.
• Response: The figure has been renamed Figure 1 and is now clearly referenced, described, and its relevance explained in the text. Its role in presenting the geographical context of the study area is highlighted.
5. Use of Abbreviations (Line 161, 209, 239):
• Response: All abbreviations (e.g., RESAEZ, DSRM, RLMA, HSG) are now written out in full upon first use to ensure clarity for all readers.
6. Table 2.1 and 2.2 (Renaming and Referencing):
• Response: Table numbers have been corrected (e.g., now Table 1 and Table 2). Each table is now properly introduced, referenced, and discussed in the text, including its significance to the methodology.
7. GIS and SCS-CN Background (Line 199–201):
• Response: We have added a concise paragraph explaining the principles and applications of GIS and the SCS-CN method, supported by relevant literature (e.g., Mishra & Singh, 2003; Ponce & Hawkins, 1996), particularly in the context of stormwater management.
8. CN Value Calculation and Equation Clarification (Line 228):
• Response: We have clarified the process used to calculate Curve Number (CN) values using LULC and HSG data layers, with GIS-based overlay analysis. The explanation of the runoff equation has been corrected to clearly show how CN values are used within the SCS runoff estimation formula.
9. Abstraction Ratio Reference (Line 234):
• Response: We now provide a reference for the commonly used initial abstraction ratio of 0.2 (e.g., USDA-SCS, 1972), and we briefly explain its basis and context.
10. Weight Assignment Justification (Table 2 / Line 239):

• Response: The basis for assigning weights (60% for runoff coefficient and 40% for slope) is now explained, referring to previous studies and expert judgment used in similar multi-criteria decision analyses (e.g., Malczewski, 2006; Mbilinyi et al., 2007).

Comments 4: [Results General comment: At the moment, you are basically dumping all your results on this section, without even explaining what each figures represent; let alone giving insights into the results. This section need to be fully rewritten!!]

Response 4: We sincerely thank the reviewer for pointing out the shortcomings in the Results section. We acknowledge that the previous version presented multiple outputs without sufficient explanation or interpretation. In response, we have completely revised the Results section to ensure that each figure and table is clearly introduced, properly described, and contextualized.

For each result, we now:

• Introduce the purpose of the figure or table.
• Explain what the visual or data represents.
• Highlight key patterns, trends, and significant findings.
• Connect the results to the objectives of the study and the underlying hydrological or spatial processes.

We also added brief transitions between subsections to improve the narrative flow. These changes aim to improve clarity, reader engagement, and the scientific value of the section.

Comments 5: Discussion General Comments: - This section basically discusses the results you obtained in the previous section; I would suggest just combining Section 3: Results with Section 4: Discussion. - Generally, you need to talk; even a little bit, about the figures you presented. If it is not important to talk about, it should NOT be in the paper!!. - Importantly, you also need to have a general discussions on the inter-relations between different data, and how the information are used to achieve your objective; how does they relate to solve your objectives in line 112 and 114. Line 315: “The soil classification map….” Which soil classification map? Figure 3.1(a)? there is no red and blue area though… Line 320-325: You are doing it the other way round; you need to explain what soil texture is important, before talking about your result (Line 315-320) Line 327: Similarly, here. Which figures are you referring to. Line 332-333: Please explain the significant of HSGs, before talking about your results. Line 334: In Figure 3.1(b), there are 3 HSGs only: B, C and D. Line 345-351: Similarly here, explain about the significance and then, your results. Line 354- Can I suggest you talk about the figures in consecutive order. Line 376- How about Table 3.1 and Figure 3.1(b); please discuss these also. Line 378: There is no Figure 4(a); I think you mean Figure 3.3(a). Line 395: Which figure is this discussion based on? Line 413-415: How are you calculating drainage density from drainage network? Figure 3. Line 427: With all the discussions before this, you need to have a discussion to summary the information so far and the inter-relations between them. At the moment, the results were dealt in silos. Line 427: What are the relationship between the reservoirs and percolation tansk given in Figure 3.4, with all the information (rainfall, soil type, etc.) presented before this. I am assuming that you are using the previous information to find the potential volume that can be harvested; but how did you do that?? Line 427: The information in Table 3.2 are estimates from the previous data; but how can you proof that the volumes estimated are close the actual volumes? Line 427: You need to conclude this section with how you are solving your objectives in line 112 and 114.

Response 5:

We thank the reviewer for the thorough and insightful feedback on the Discussion section. We acknowledge that the previous structure lacked clarity in linking figures to the narrative and in providing meaningful interpretation of the results.

In response to the comments:

1. Merging Results and Discussion:
   Following the reviewer’s suggestion, we have merged the Results and Discussion sections into a unified section. This change improves the narrative flow and allows for immediate interpretation of each result within its spatial and hydrological context.
2. Clarity and Figure Referencing:
   We have reviewed all figures and ensured that each one is:
• Clearly referenced in the text.
• Introduced with a purpose.
• Interpreted with insights on what the patterns or values imply.
• Discussed in consecutive order for logical coherence.

Any figures or tables that were not discussed or did not contribute meaningfully to the objectives have been removed from the manuscript.

1. General Discussion and Interrelationships:
   A dedicated paragraph has been added to reflect on how the integrated datasets (e.g., LULC, HSG, slope, CN values, and runoff) interact to support stormwater harvesting potential identification. This paragraph directly relates the findings back to the objectives stated in Lines 112 and 114.
2. Specific Line Comments:
• Line 315: We have clarified that this refers to Figure 3.1(a) and corrected the color description, as there are no red and blue areas.
• Lines 320–325: The section now begins with a general explanation of why soil texture is important for infiltration and runoff processes, followed by interpretation of the mapped results.
• Line 327 & Line 332–333: The significance of Hydrologic Soil Groups (HSGs) is now explained first, with references to their infiltration capacity and impact on CN values, before discussing the results.
• Line 334: We have updated the description of Figure 3.1(b) to accurately reflect the HSGs present (B, C, and D) and explained their hydrological implications.
• Lines 345–351 & 354: The text has been restructured to describe the figures in proper sequence, explaining the underlying hydrological relevance before describing the observed spatial patterns.
• Line 376–378: A paragraph has been added to interpret Table 3.1 and Figure 3.1(b), highlighting their contribution to identifying runoff potential and water harvesting suitability.
• Line 395: The figure being referred to is now explicitly stated, and the discussion is rewritten accordingly.
• Line 413–415: We have now provided a short explanation of the drainage density calculation method, referencing the use of stream length and watershed area derived from the drainage network extracted using a DEM-based flow accumulation approach.

5.       Summary and Integration of Results (Line 427):
A new summary subsection has been added after Figure 3 to synthesize the findings across different datasets and analyses. This section explicitly discusses the interrelationships between rainfall patterns, soil properties, land use, hydrologic soil groups, runoff potential, and the spatial distribution of reservoirs and percolation tanks. By doing so, the results are no longer treated in isolation but presented as a coherent framework supporting stormwater harvesting potential assessment.

6.       Relationship Between Reservoirs, Percolation Tanks, and Environmental Factors:
We have clarified how the previously analyzed biophysical variables (rainfall, soil texture, slope, runoff curve numbers) were integrated to estimate potential water volumes in reservoirs and percolation tanks shown in Figure 3.4. The methodology section now explicitly describes the calculation approach used to derive these volumes, incorporating runoff coefficients, catchment area, and rainfall data, demonstrating how each factor influences water harvesting potential.

7.       Validation and Confidence in Volume Estimates (Table 3.2):
We acknowledge the importance of validation. Due to limited field data, we discuss the inherent uncertainties and assumptions involved in the volume estimation. We compare our results qualitatively with similar studies in comparable agroecological zones (e.g., Ali et al., 2021; Kassa & Tadesse, 2020) to show consistency with expected ranges. We also suggest that future work should include field verification and monitoring to validate these estimates.

8.       Linking Results to Study Objectives (Lines 112 and 114):
The section concludes with a clear statement summarizing how the integrated analysis of spatial and hydrological data achieves the study’s stated objectives: identifying suitable sites and estimating harvestable stormwater volumes to support sustainable water resource management in Rwanda’s Eastern Savanna Agroecological Zone.

Comments 6: Conclusions General Comments: - Your conclusion is too short; you may want to expand on the methods that you used. - You may want to highlight that although you are focusing on Rwanda, the method is applicable for other regions; provided the same data is available. Line 443: There should not be any reference to Figure in the conclusion; I cannot see where Figure 3.4(a) as well.

Response 6: We have revised and expanded the Conclusion section to better summarize the key methodologies used in the study and to clearly highlight the broader applicability of the approach beyond Rwanda. Additionally, all figure references have been removed in accordance with journal guidelines. These improvements enhance the clarity, completeness, and relevance of the concluding remarks.

Comments 7: [What are the properties of B, C, and D given in Figure 3.1.b, the Hydrological soil group map?]

Response 7:

Comments 8: [Figure 3.2: Provide the units of the maps.]

Response 8: Thank you for your comment. We have reviewed all the maps included in the manuscript and updated the figure captions and legends

“[Figure 4: study area characteristics: (a) rainfall (mm), (b) land use land cover]”

Comments 9: [Figure 3.2.b: The last color of the legend does not have a label..]

Response 9: Agree. I/We have, accordingly, modified to emphasize this point.

Comments 10: [Section 4.2: The spatial and temporal distribution of rainfall is mentioned. However, only one map is provided. The authors did not present the temporal variation of the rainfall. Between what dates was rainfall data collected? The conclusion part also mentioned the erratic rainfall, but it is not presented in the paper.]

Response 10: Agree. I/We have, accordingly revised to emphasize this point.

Comments 11: [There are more results in the discussion section, and no discussion has been prepared by referring to the relevant literature. Therefore, this section should be expanded and the contribution of the study to the literature should be emphasized. As the method is old, many studies used the same method. In addition, the limits should be stated. For example, was 90 m LULC sufficient? Would more detailed results have been obtained if there were other data sources.]

Response 11: Thank you for your valuable feedback. We agree that the Discussion section would benefit from a more thorough integration of relevant literature. Accordingly, we have expanded this section to include comparisons with previous studies that applied the SCS-CN method, thereby better situating our findings within the existing body of research and highlighting the unique contributions of our study.

Regarding the limitations, we have added a detailed discussion on the use of 90 m resolution LULC data. We acknowledge that while this resolution provides a good balance between coverage and detail, higher-resolution datasets could potentially offer more precise spatial analysis. This limitation and its possible impact on the results are now explicitly stated, along with suggestions for future studies to explore finer-scale data sources.

We believe these revisions strengthen the manuscript and improve its clarity and scientific rigor.

4. Response to Comments on the Quality of English Language

Comments on the Quality of English Language

Some sentences are long, dense, or awkwardly phrased. For example: "This research classified the study area's terrain into five classes based on slope gradients, which are crucial for understanding runoff, erosion potential, and land-use planning." This sentence could be broken into clearer, more concise parts.

In addition, there are some minor grammatical errors (e.g., use of “harvesting’s” instead of “harvesting has”).

Point 1:

Response 1: (We appreciate the reviewer’s observation regarding the language quality. The manuscript has been carefully reviewed and thoroughly edited to improve clarity, grammar, and overall readability. We have also utilized professional language editing services to ensure that the manuscript meets the high standards of the journal.)

5. Additional clarifications

[The manuscript has been thoroughly revised to address the reviewers’ comments and to align with the journal’s requirements. All suggested improvements, including expansion of the Discussion section with relevant literature, clear articulation of study limitations, and clarification of methodological choices, have been incorporated. We believe the revised version is now significantly strengthened and meets the standards expected for publication

Comments 1: Respected Authors,

Your manuscript, titled "GIS-Based Assessment of Stormwater Harvesting Potentials: A Sustainable Approach to Alleviate Water Scarcity in Rwanda's Eastern Savanna Agroecological Zone", addresses an important and timely issue—stormwater harvesting as a sustainable solution to water scarcity in Rwanda's Eastern Savanna Agroecological Zone.

It effectively integrates GIS and hydrological modeling, offering a spatially informed framework to assess stormwater harvesting potential. The work is methodologically sound and could contribute significantly to regional water management policies and sustainable development.

In my opinion, the major strengths of the manuscript are:

The study focuses on a region (Eastern Rwanda) that is underrepresented in stormwater harvesting literature. It addresses pressing issues like climate variability, water scarcity, and agriculture-linked livelihoods, and represents a novel application of GIS and the SCS-CN method for this specific agroecological zone. This fact means that the research that has been done is relevant and innovative.

Regarding the methodology that was followed, I believe that the integration of various datasets (DEM, LULC, rainfall, and soil data) is thorough, the application of the SCS-CN method is appropriate and clearly described, and multi-criteria decision analysis for site selection is well justified.

The detailed discussion section provides insight into how various biophysical variables influence runoff and water harvesting potential, and the explanation of land slope, soil groups, curve numbers, and their hydrological implications is well-articulated.

Finally, at the policy level, the manuscript offers actionable recommendations aligned with Rwanda's development goals and SDGs and advocates integration into national water management strategies.

Response 1: [.We sincerely thank the reviewer for the thoughtful and encouraging evaluation of our manuscript titled "GIS-Based Assessment of Stormwater Harvesting Potentials: A Sustainable Approach to Alleviate Water Scarcity in Rwanda's Eastern Savanna Agroecological Zone." We are grateful for your recognition of the importance and timeliness of the topic, as well as your positive assessment of the methodological framework and the potential policy impact of the study.

We particularly appreciate your acknowledgment of the novelty of applying GIS and the SCS-CN method in an underrepresented agroecological zone such as Eastern Rwanda, and the relevance of our focus on climate variability, water scarcity, and agricultural livelihoods. Your comments regarding the clarity of the methodology—including the integration of DEM, LULC, rainfall, and soil data, and the justification for the use of multi-criteria decision analysis—are highly encouraging.

We are also pleased that the discussion of the biophysical variables influencing runoff, as well as the hydrological implications of land slope, soil groups, and curve numbers, were found to be clearly articulated. Moreover, we thank you for highlighting the alignment of our policy recommendations with Rwanda’s national development priorities and the Sustainable Development Goals (SDGs).

To further improve the manuscript, we have reviewed and refined relevant sections to enhance clarity and ensure that the strengths identified are effectively communicated. Thank you again for your constructive and motivating feedback.

Comments 2: On Language and Grammar:

Some sentences are long, dense, or awkwardly phrased. For example: "This research classified the study area's terrain into five classes based on slope gradients, which are crucial for understanding runoff, erosion potential, and land-use planning." This sentence could be broken into clearer, more concise parts. In addition, there are some minor grammatical errors (e.g., use of “harvesting’s” instead of “harvesting has”).

So, a professional language edit will improve readability and clarity.]

Response 2: [Thank you for pointing out the issues related to sentence structure, grammar, and overall clarity. We have thoroughly reviewed the manuscript and revised sentences that were long, dense, or awkwardly phrased including the specific example you mentioned. Additionally, we corrected minor grammatical errors such as the misuse of terms like “harvesting’s.”

To further enhance the readability and quality of the manuscript, we subjected the text to a professional language editing process. We believe these revisions have significantly improved the clarity, coherence, and overall presentation of the work.]

Comments 3: [On Validation and Limitations:

The methodology is solid, but the study lacks a discussion of validation of the ground-truthed runoff or storage estimates, as well as the limitations or uncertainties related to remote sensing, parameter selection, or the SCS-CN method in the study area.

So, please include a short section discussing validation efforts and study limitations.]

Response 3: Thank you for pointing this out. We have carefully reviewed and revised. All references in the main text have also been updated accordingly. “[Although the SCS-CN method provides a practical approach for estimating runoff in data-scarce regions, this study lacked ground-based validation due to limited gauging station data in the Eastern Savanna zone. This may affect the accuracy of runoff and storage estimates (Ponce & Hawkins, 1996). Additionally, the use of 90 m resolution datasets (e.g., DEM and LULC) introduces spatial uncertainty, particularly in heterogeneous landscapes (Gao, 2009). The SCS-CN method also assumes uniform hydrologic conditions, which may oversimplify real-world variability (Mishra & Singh, 2003). Despite these limitations, the methodology remains robust for regional-scale assessment and can be refined with field data and higher-resolution inputs in future research.

]

Comments 4: [On References and Citations:

The references are appropriate, but some are duplicated or seem to be cited multiple times without clear contextual use. Also, some important sources are not sufficiently integrated into the narrative.

To solve this problem, please streamline references and ensure each is meaningfully cited in context]

Response 4: Thank you for your valuable observation regarding the use of references and citations. In response, we carefully reviewed the reference list and in-text citations to identify and remove any duplicates or inconsistently cited sources. We ensured that all remaining references are cited meaningfully and are directly connected to the context in which they are discussed. Furthermore, we have integrated several key studies more clearly into the narrative to strengthen the scientific grounding of our methodology and discussion.

These revisions enhance the clarity, relevance, and scholarly rigor of the manuscript.

Comments 5: [Some other less substantive issues that need to be addressed are:

The abstract (which could be tightened for conciseness), "Author Contributions" and "Funding" sections (that are incomplete and need finalization), and finally, a clear description of the rainfall years used in the analysis would add rigor.]

Response 5: The abstract has been revised to enhance clarity and conciseness while retaining all key information on objectives, methods, results, and implications. The "Author Contributions" and "Funding" sections have been completed to reflect the accurate roles of all authors and the sources of support for the research, in accordance with the journal's guidelines. Additionally, we have added a clear description of the rainfall years, used in the analysis to improve transparency and methodological rigor. This information is now explicitly stated in the Data and Methods section.

[The methodology includes analyzing land use, soil type, rainfall data (30 years, from 1994 to 2023), and topography, can be found in the abstract section

4. Response to Comments on the Quality of English Language

Point 1:

Some sentences are long, dense, or awkwardly phrased. For example: "This research classified the study area's terrain into five classes based on slope gradients, which are crucial for understanding runoff, erosion potential, and land-use planning." This sentence could be broken into clearer, more concise parts.

In addition, there are some minor grammatical errors (e.g., use of “harvesting’s” instead of “harvesting has”).

Response 1: (We thank the reviewer for highlighting the issues related to sentence structure and grammatical errors. We have carefully revised the manuscript to break down long and dense sentences into clearer, more concise statements. For example, the sentence regarding terrain classification has been rephrased for improved readability.

Additionally, all identified grammatical errors, including incorrect uses such as “harvesting’s” instead of “harvesting has,” have been corrected throughout the text. To further enhance the manuscript’s clarity and fluency, a professional language editing service was employed. We believe these improvements significantly enhance the overall readability of the manuscript.)

5. Additional clarifications

[Thank you for your valuable feedback. We appreciate the need for further clarifications to improve the manuscript’s clarity and completeness. We have carefully reviewed the comments and provided additional explanations and details in the revised manuscript wherever necessary. These clarifications include methodological descriptions, data sources, assumptions, and interpretation of results.

If there are specific sections you would like us to elaborate on further, please let us know, and we will be happy to address them.

Comments 1: The authors answered the questions properly and improved the manuscript. However, the discussion section has been removed, and it is not clear why they made such a decision. A proper discussion section should be included by referring to the relevant literature. Therefore, this section should be expanded, and the contribution of the study to the literature should be emphasized. As the method is old, many studies have used the same method.

Response 1:

We sincerely thank the reviewer for this important observation. We acknowledge the critical role of the discussion section in contextualizing the findings and highlighting their implications within the existing body of literature.

The removal of the original discussion section was unintentional during the restructuring of the manuscript. We regret this oversight and have now restored and significantly expanded the discussion section in the revised version of the manuscript.

In this revised section, we have:

• Clearly interpreted and contextualized our key findings in relation to recent and relevant studies.
• Highlighted how our results align with or diverge from prior work using similar methods, thereby reinforcing the originality and relevance of our contribution.
• Emphasized the practical implications of our findings for hydrological planning and water resource management in semi-arid agroecological zones.
• Discussed the limitations of using traditional methods such as SCS-CN and proposed how future studies could complement our approach using more recent techniques.

All additions and revisions have been clearly marked using track changes for your convenience.

We believe that the restored and revised discussion section now provides a stronger foundation for understanding the significance of our work and its contribution to the broader scientific discourse.

2. Response to Comments on the Quality of English Language

Point 1: The English is fine and does not require any improvement.

Response 1:    (We sincerely thank the reviewer for the positive feedback regarding the quality of the English language in the manuscript. We are pleased that the clarity and readability of the text meet the journal’s standards.)

3. Additional clarifications

We would like to clarify that the manuscript underwent structural reorganization to improve coherence, during which the discussion section was unintentionally omitted—this has now been fully restored and expanded. Minor revisions were also made to enhance consistency in terminology, update relevant references, and improve the clarity of figures and tables. These changes aim to strengthen the overall quality and readability of the manuscript.