Integrating Hydrological Models for Improved Flash Flood Risk Assessment and Mitigation Strategies in Northeastern Thailand
, Kittiwet Kuntiyawichai 3, Jirawat Supakosol 4, Krit Sriworamas 5, Ounla Sivanpheng 6 and Anongrit Kangrang 1,*Round 1
Reviewer 1 Report (New Reviewer)
Comments and Suggestions for AuthorsThe paper refers to a critical issue related to flash flood risks in Northeastern Thailand by using HEC-RAS. While the paper addresses a critical topic, the novelty of the research is insufficient to be considered groundbreaking. The integration of existing models and methodologies, though well-applied, does not introduce substantial new developments or techniques. Moreover, to be consider for publication a lot of details regarding the simulations are missing.
Historical simulations and scenario-based projections are made but the authors should include a schematic picture showing what will be presented and what are the most important information for each case.
The study focuses on Thailand so authors should explain how the findings regarding the impacts of land use changes, calibration of hydrological models, and flood mitigation strategies can be extrapolated to similar regions globally.
The HEC-RAS model’s potential inaccuracies in complex terrains and discrepancies in satellite imagery data could be better discussed.
Uncertainty analysis, such as Monte Carlo simulations or sensitivity analysis of key parameters, is missing. For instance, explain how the spatial resolution limitations in satellite data might lead to errors in flood extent estimation.
As so many literature references were included in the introduction, the study could benefit from comparative analysis with other regions or models in the discussion section. With this author can highlight specific innovations or new contributions in methodology, data integration, or application to enhance the originality.
The paper occasionally relies on simple or obvious statements that do not contribute significantly to the discussion. For example: "Flash floods pose a significant threat to human life and infrastructure in flash flood-prone areas."
The paper lacks critical details about the model configurations, such as (for instance):
- Boundary conditions.
- Simulation type (steady/unsteady).
- Model dimensionality (1D/2D).
- Cross-section spacing (if 1D).
- Mesh configuration (if 2D).
Until line 94 author cite 45 references. This is excessive and reduces clarity.
In Figure 2, the meaning of the colors is unclear.
The screenshot in Figure 4 of Hec-Ras Mapper is not professional or informative. I suggest to include the script text as a code snippet or describe it in the text.
The meteorological stations used for the SWAT model are not detailed. Additionally, consider using ERA5-land data for improved precision.
Daily resolution may not adequately capture flash flood dynamics, which often occur over shorter time scales. Please explain.
The statement: "Flood hazard maps from historical storm events were used to further validate the model" lacks explanation.
Figures 5 and 6 require improvements in clarity and visual appeal.
Figure 7 appears to show the flooded area output from HEC-RAS rather than calibration data.
The hydrograph from the SWAT model, which serves as the input to HEC-RAS, is crucial for understanding the impacts of land cover changes.
In line 608, the term "annual runoff" needs clarification. If it represents an average value, it may not accurately reflect flood conditions.
The use of a 1D model to capture 2D phenomena in large floodplains is questionable.
The paper contains inconsistent font sizes, such as in lines 157 and 159.
Author Response
see the attached file
Author Response File: 
Author Response.pdf
Reviewer 2 Report (New Reviewer)
Comments and Suggestions for Authors
Comments and Suggestions for Authors
The manuscript entitled: “Integrating hydrological models for improved flash flood risk assessment and mitigation strategies in Northeastern Thailand” is interesting. The topic of the paper is very important. The aim of the study was to assess flash flood risks using the HEC-RAS model in northeastern Thailand. The research specifically focused on the Lam Saphung, Phrom, and Chern River Basins, which are prone to flash floods. Authors have emphasized the significance of proactive risk management and targeted flood mitigation measures to enhance the resilience of community. Authors have highlighted that good planning in the management of water resources, e.g. efficient monitoring and drainage systems can help reduce the risk of flooding in the long run. According to Authors, accurately estimating flash flood risks using the HEC-RAS model can provide essential insights for stakeholders, e.g. urban planners, local agencies, as well as emergency management officials. The results of the study have confirmed the importance of understanding the geological and hydrological, characteristics of the region, as well as the river flow and annual (or monthly) rainfall patterns, which are crucial for assessment of the flood risk. The findings emphasize the need for customized flood risk mitigation strategies which address the specific characteristics of each drainage basin.
Below are presented specific comments to the Manuscript: water-3417673.
· Lines 184, 190, and 199: In these lines there are placed the units “MCM”. Please, explain (or change) the unit MCM. Is it million cubic metres (106 m3)?
· Lines 285 and 292: Please, check carefully the equations 1 and 2. There is probably an error in equation 2 because it is identical to the equation 1.
· Lines 364 – 365: Please, correct the sentence: “The precise locations of these stations are indicated in Figure [insert figure 9]”.
· Lines 325, 384, and 391: The text in Figures 4, 5, and 6 is poorly readable. This negatively affects the understanding of the information contained in these Figures. Would it possible to improve the quality of the text in these drawings?
· Lines 328 – 344: In the Chapter 3: Results and Discussion, Authors write that the results of the study provide a useful database for developing flood management. Reading this chapter, it seems that this conclusion is repeated several times. Please, analyse the text carefully and consider whether certain fragments can be shortened by removing some repetitive information.
· Lines 488 – 492, 540 – 544, 572 – 576, and 655 – 659: Please, check Figures 10 and 12, 14, and 16. Especially check the description of X-axes (Year). What do the abbreviations: Q/7, Q/5, Q/3, Q/., Q.6, and Q.4 mean? Check also the names of Y-axes. These names (especially the units) are not clear.
· The references should be edited in accordance with the indications for Authors.
Comments for author File: Comments.pdf
Author Response
see the attached file
Author Response File: Author Response.pdf
Reviewer 3 Report (New Reviewer)
Comments and Suggestions for AuthorsThe manuscript addresses a critical issue in flood risk management, particularly in a flood-prone region of Thailand. By combining HEC-RAS and SWAT models, the study offers valuable insights into the role of land-use changes and historical flooding events. However, significant revisions are required to enhance clarity, methodological rigor, and practical relevance.
Specific Comments:
1. The introduction outlines the importance of flood risk management but fails to clearly state the specific research objectives or hypotheses. These should be explicitly defined to give readers a focused understanding of the study's aims and contributions (Section 1).
2. The manuscript mentions the Lam Saphung, Phrom, and Chern River Basins but does not explain why these specific basins were selected. Provide a detailed justification, emphasizing their significance in terms of unique hydrological, topographical, or socio-economic characteristics (Section 2.1).
3. While data sources such as SWAT and HEC-RAS are mentioned, details about data quality, resolution, and pre-processing steps are insufficient. Elaborate on the validation methods for input datasets (e.g., land use, rainfall) to improve transparency and reliability (Sections 2.2–2.3).
4. The manuscript briefly mentions model calibration but lacks detail on calibration metrics, parameter adjustments, and performance outcomes. Include specifics about the statistical methods used (e.g., NSE, R²) and their thresholds for determining model adequacy (Section 2.3.1).
5. The two-case simulation strategy (land-use changes and historical flood events) is an interesting approach but needs further explanation. For example, clarify how the land-use scenarios were chosen and whether they reflect realistic future conditions. This would help strengthen the applicability of the findings (Section 3).
6. The study does not address potential biases, such as limitations in satellite imagery resolution or challenges in applying the HEC-RAS model in complex terrains. Discuss these constraints and their implications for model accuracy and results interpretation (Section 4.3).
7. The results section is overly descriptive and lacks critical interpretation. For instance, explain why certain areas exhibited higher flood risks and how land-use changes contributed to these outcomes. Relate these findings to broader flood management practices (Sections 3.2–4.1).
8. While the study highlights mitigation strategies, these are not sufficiently detailed or actionable. Provide concrete recommendations for policymakers, urban planners, and local communities based on the findings, such as specific land-use policies or infrastructure improvements (Section 5).
Author Response
see the attached file
Author Response File: Author Response.pdf
Reviewer 4 Report (New Reviewer)
Comments and Suggestions for AuthorsDecision: Accept
Thank you for this review invitation.
This manuscript “Integrating hydrological models for improved flash flood risk assessment and mitigation strategies in Northeastern Thailand” addresses a scope of topic of flood risk assessment using the HEC-RAS model (scenario-based simulations and historical secnario simulations) in three river basins in northeastern Thailand.
The authors contributed to the sector of flood risk management by providing valuable insights into the impacts of land use changes on flood hazards. In my view, their methodology (HEC-RAS model via mapping and statistical analysis) was well-designed, utilizing two distinct approaches to evaluate flood conditions: simulating historical land use changes and modeling past occurred flood events. The use of the HEC-RAS model, coupled with SWAT-derived runoff data, provides a concrete analytical foundation.
The authors have meticulously calibrated and validated the model using appropriate statistical measures, including R², NSE, and P-BIAS, increasing the credibility and reliability of their results.
The main question in the study document came from the thought how to estimate and minimize flash flood hazards in Northeast Thailand, with an emphasis on the Lam Saphung, Phrom, and Chern River basins. Their work and results simulates flood flow behavior using the HEC-RAS model, which takes into account parameters such as river cross-section data, land use, and previous flood occurrences.
Their English writing looks fine with me, and the contents in results and discussion are well addressed based on the analytical design
I have no objections to this analytical logic. The paper is well-structured and clearly written, with a logical flow of ideas from introduction to conclusions.
The study addresses valuable insights that will be of interest to researchers, policymakers, and practitioners working in flood risk management and urban planning. I would suggest accepting this manuscript to our journal “Water.”
Author Response
see the attached file
Author Response File: Author Response.pdf
Reviewer 5 Report (New Reviewer)
Comments and Suggestions for Authors1: The abstract needs improvement by incorporating quantitative results to better support the manuscript's findings. For example, include specific data on flood depths, high-risk area percentages, or validation metrics (e.g., R², NSE) to demonstrate the model's performance and reinforce the study's conclusions.
2. The introduction needs to define the research question and scope more clearly. It is recommended to cite much more recent relevant publications to increase the readability and quality of the manuscript. The following latest and related work will further improve the study: “Error Correction Method based on Deep Learning for Improving the Accuracy of Conceptual Rainfall-runoff Model. Journal of Hydrology, 2024, 643:131992.”
3: Figure 2 should use a geographic coordinate system (e.g., latitude and longitude), and a compass (north arrow) should be added. The legend is incomplete and should explain the meaning of the colored areas, particularly the representation of blue zones (e.g., are they reservoirs or other water bodies?). Additionally, if human interventions such as hydraulic structures (e.g., dams, reservoirs) exist within the basin, they should be indicated on the map with a discussion of their impact on flood risk analysis.
4: The x-axis labels of Figures 10, 12, 14, and 16 are inaccurate and should be corrected. In addition, clear figure captions and explanations should be provided to ensure the context of the data is properly understood.
5: The manuscript does not include sufficient details about the SWAT calibration and validation. These results are critical for credibility and should be presented with statistical performance metrics. Additionally, since a daily model is used in SWAT, the smoothing effect on instantaneous flood peaks should be discussed. This could significantly influence the results and needs explanation, especially regarding how this affects the coupling with HEC-RAS.
6: The description of the HEC-RAS model setup is unclear and incomplete. Specifically:
1) Were 1D and 2D model coupling methods applied? If so, provide further details. If the topographic data for defining river channels were only extracted from the DEM, which often lacks sufficient accuracy, explain whether and how river cross-section data were refined or validated.
2) Explain how boundary conditions were set for HEC-RAS, particularly upstream and downstream boundaries, initial flood conditions.
3) The manuscript lacks details on the computational time step used for HEC-RAS simulations. Provide specific information on the time step settings (e.g., seconds, minutes, or hours) and explain how SWAT-generated daily runoff data were integrated into the HEC-RAS model for coupling. Since HEC-RAS simulations typically require shorter time steps to capture dynamic flood processes, there is concern about the appropriateness of directly coupling daily runoff data from SWAT. Discuss whether the daily runoff data were interpolated or processed to match the temporal resolution of HEC-RAS and justify the coupling approach used.
4) The calibration and validation of Manning’s roughness coefficient for HEC-RAS also needs to be clearly explained with validation results.
7: The study does not address whether human interventions (e.g., reservoir operation, levees) or water engineering structures were considered in the simulations. These factors could significantly impact flood risk and should be discussed, particularly in regions with active human management of hydrological systems.
Comments on the Quality of English LanguageThe quality of the English language in the provided document is generally high.
Author Response
see the attached file
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report (New Reviewer)
Comments and Suggestions for AuthorsFollowing the reviewers' comments, the paper has been revised and improved significantly. The new version is more clearer and can now be considered for publication. However, there are still a few aspects that require further attention:
1 - Please explicitly state the novelty of the paper in comparison to other similar studies. Highlight the unique contributions of your work to distinguish it clearly.
2 - As referred before, the authors should include an uncertainty analysis in the discussion to strengthen the reliability of the results and interpretations.
3 - As referred in round 1, the authors must provide more details about the model configuration and characteristics. This should include information on boundary conditions, flow regime, cross-section spacing, mesh configuration, and any other relevant parameters.
4 - The number of references, particularly in the introduction, should be reduced.
5 - The authors have not adequately explained how a simulation with daily discretization can effectively model flash floods. In response to my previous comment, the authors referred to lines 151–167, but this section does not address the issue. Please provide a clear explanation or make the necessary corrections.
Author Response
We really appreciate the reviewers' comments, which are very detailed and very helpful in improving our manuscript water-3417673 . We have made a minor revision to our manuscript for round 2. We have improved our manuscript following your comments. All of the changes have been modified in the revised manuscript. Please find the attached files of both revised version and response to reviewers.
Author Response File: Author Response.pdf
Reviewer 3 Report (New Reviewer)
Comments and Suggestions for AuthorsI have no significant concerns that would prevent its acceptance. I recommend it for publication in its current form.
Author Response
We really appreciate the reviewers' comments, which are very detailed and very helpful in improving our manuscript water-3417673 . We have made a minor revision to our manuscript for round 2. We have improved our manuscript following your comments. All of the changes have been modified in the revised manuscript. Please find the attached files of both revised version and response to reviewers.
Author Response File: Author Response.pdf
Reviewer 5 Report (New Reviewer)
Comments and Suggestions for AuthorsComments have been resolved.
Author Response
We really appreciate the reviewers' comments, which are very detailed and very helpful in improving our manuscript water-3417673 . We have made a minor revision to our manuscript for round 2. We have improved our manuscript following your comments. All of the changes have been modified in the revised manuscript. Please find the attached files of both revised version and response to reviewers.
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 AuthorsThis work entitled “Integrating HEC-RAS Modeling for Improved Flash Flood Risk Assessment” combines the SWAT and HEC-RAS models to assess the flash flood hazard in selected sub-basins of northeastern Thailand. The study evaluates flood hazard and vulnerability under different scenarios of land uses and past flood events, underscoring the significance of proactive risk management and mitigation measures. The topic of this research is impactful and the paper includes interesting ideas. However, some methodological aspects need to be further clarified, innovative aspects should be emphasized, and datasets used should be better described. Some major comments are given below:
1) Th abstract of the manuscript is quite confusing and distracts attention from the methodological framework and critical findings of this work. The abstract should be restructured to include the following information: i) scope and objectives of the study, ii) methodology of the work, iii) innovative aspects, iv) case studies selected, v) main extracted results, and vi) principal conclusions. The abovementioned sequence of presenting information should be followed in the abstract.
2) The manuscript does not clearly present its novelties. In the present form of this work, it is not clear what are its main innovations with respect to current literature, and its contribution in existing knowledge on flood hazard and risk assessment and management. Such innovative features could be made more distinct in the introductory section of the paper, distinguishing it from an advanced case study. A more detailed description of these features could significantly improve the quality of the manuscript.
3) In the paper, there is sometimes a misuse of the words “hazard” and “risk”. For a hazard to arise there must be a source or initiating event or a combination of events, and a pathway connecting the sources to the receptors of flood risks. Hazards are normally described as the probability of flood events with a certain magnitude and other features. The risk is a combination of the probability and the consequences of the hazard. More frequently, risk is described as a function of probability, exposure and vulnerability. For example, in Fig.1 the methodology as it is presented, results in assessing the flood hazard.
4) The authors should describe calibration and validation procedures of both the SWAT and HEC-RAS models in more details in section 2 of this work. How was calibration and validation of the models performed? What are the calibration parameters in both models? Which water level stations were used for this purpose? What is the time resolution of the data used for both procedures and what is the time extent of the datasets? Did the authors also use flood hazard maps corresponding to historic storm events (major flood events) for both the calibration and validation procedure? If so, which were these historic events for all sub-basins considered? What are the maximum allowable differences between observed and simulated data used for calibration purposes?
5) The rainfall datasets used in this work are poorly described. What is the time period of measurements? What is the time resolution of the observations and how were these datasets acquired (i.e., organizations that provided the data or sites to download them)?
6) Measures R^2, NSE, P-BIAS, RE and F-statistics should be included and shortly presented in the methodology section of this work.
7) In Section 3.3.1 it is mentioned that at Station E.85, where the simulated peak was higher than the observed, inaccuracies in the SWAT model's flow rate estimates might have led to an overestimation of the input data for the HEC-RAS simulation. Please explain what has possibly caused these inaccuracies observed in the SWAT model.
Some minor comments and corrections are given below:
1) Line 65: What do the authors mean by “such as rainfall land use”?
2) The word “data” is usually considered singular.
3) There is no need to mention the version of the HEC-RAS model used.
4) Lines 85-86 : Please rephrase “Design … flash floods.”
5) Lines 93-94: This sentence is a repetition of the previous one and should be erased.
6) Lines 96-99: Please rephrase “The study area … in the past.”
7) Include references for the SWAT model
8) The sub-basins E83, E93 and E85 mentioned section 2 (line 136) have not been previously defined.
9) Lines 161-162: How is it evident from Fig. 2 that the three sub-basins, despite their proximity, exhibit distinct physical characteristics?
10) Lines 168 & 180: Better replace “heightened” with “increased”
11) Line 206: “flash flood” is repeated twice
12) The site of line 207 should be put in parentheses
13) Write the resolution of the DEM in Table 1
14) What does the “Hydraulic Building” in Table 1 mean and what does “Shape file” include? Explain GISTDA.
15) Explain LDD DEM (line 215)
16) Line 230-232: Please rephrase “Detailed … study area.”
17) Flood curves presented in Fig. 3 are not used to create flood hazard maps. Flood hazard maps usually contain information on flood extent, flood depth and sometimes flood velocity. When information of Fig. 3 is included in the map, a flood vulnerability (or sometimes flood risk) map could be created.
18) Figs 5 and 6 should be made larger and their resolution should be improved.
19) Fig. 8 does not refer to September 27, 2021. Verification of Fig. 8 is performed for the data recorded on 20 October 2010.
20) Lines 346-347: Please rephrase “Results … conditions.”
21) Figs. 9 and 11 present flood hazard maps (with flood extent and flood depths). The Figure resolution should be increased.
22) Line 380: “… of 0.92%).”
23) I don’t understand the x-axis labels of Figs. 10, 12, 14 and 16.
24) Lines 450-452: Please rephrase “It depended … drought.”
25) Figs 13 and 15 do not represent just the extent of flooded areas, but also flood vulnerability in the affected areas.
26) Line 331: “3.3.2. The case …”
27) Lines 585-587: Please rephrase “Using … and 2006).”
28) Change the titles of Tables 6 and 7.
29) Line 684: “… and P-BIAS.”
30) Line 689: Higher resolution data refer to the DEM used in this work, or does it refer to the other datasets used?
Comments on the Quality of English LanguageComments and corrections on the quality of English used are given in the minor comments section of my review.
Author Response
Please find the attached file, we have responded the comments in the attached file.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe document is interesting but some concepts are not very clear. Please see the attached file.
Comments for author File: Comments.pdf
Author Response
Please find the attached file, we have responded the comments in the attached file.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe revised version of the manuscript is improved compared to its original version. The authors tried to adress the reviewers' comments in this new version. However, the authors should also consider the following comments before publication:
1) A more elaborate description of statistical measures and performance criteria should be included in the methodology section of this paper. The relevant formulas of these measures and range of possible values (where relevant) could be included.
2) The manuscript should be rechecked for unnecessary repetitions. These repetitions should be erased to improve the manuscript's clarity and consistency.
3) Do Figures 9 and 11 present only the flood extent or do they show the flood hazard? It is evident from both Figures that flood depth is also included in the flooded areas.
4) Why do Figures 13 and 15 have different captions in terms of which variables are presented therein? Please recheck the two captions for consistency.
5) It is better to use the term "flooded area" instead of "flood area"
6) Use sub-basin (recheck for the term subbasin) uniquely throughout the text.
7) The resolution of all Figures should be improved
8) Which Figure is mentioned in pg 10, line 330?
9) In pg 13, line 369: "Low RE and large F-statistic values demonstrate ..."
Comments on the Quality of English LanguageMinor editing is needed in the English syntax of the manuscript.
Author Response
Please see the attached file
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe last two references are incomplete. Please, check them
Author Response
Please see the attached file.
Author Response File: Author Response.pdf
