How Can Cities Respond to Flood Disaster Risks under Multi-Scenario Simulation? A Case Study of Xiamen, China
Round 1
Reviewer 1 Report
This is an interesting, timely, and useful paper.
I particularly like the construction of an index addressing risks, sensitivity, vulnerability, and capacity measures;
More explanation beyond the mathematical formulation of the entropy weight assignment is needed. It is not clear if the weighting has been validated or tested in terms of the underlying logic. More discussion of the data quality, accuracy, spatial resolution of the different data elements is needed.
I am concerned about the different scales and resolution of data on precipitation, flooding, and impacts given the diverse topography and land use and differences in terms of drainage systems and flood control and the effects in terms of absorption, runoff, detention, retention, etc.
Also, the description of the infrastructure system and investment is lacking. More discussion of the growth, development, and conversion of lands from natural to agricultural to urban uses would be helpful in better understanding the risks and natural and extent of flooding.
While it is not surprising to find a large coefficient of variation for the typhoon effects, it is not clear how best to manage these risks versus more routine flooding or problem associated with 3 year average precipitation levels.
Would it make sense to better formulate the risks to health and safety in terms of death/injury versus property losses versus disruption versus flooding of agricultural or less densely populated areas?
It's not clear also what the unit of analysis is - especially in terms of risk management, land use controls, drainage or flood control projects or more localized interventions.
The paper provides useful general background information, but lacks sufficient details on costs and benefits associated with more specific engineering, planning, or risk reduction projects. There are too many generalities with regard to inadequate or insufficient drainage/flood control projects. From a research perspective, the paper is useful, but from a more pragmatic and directed action-oriented perspective, it doesn't identify priorities, or types of projects nor investments which can be pursued, justified, and implemented. It seems that more environmental and public health dimensions could've been emphasized in developing actions or interventions to reduce risks from flooding. Also the tradeoffs between more frequent lower intensity events versus more rare larger impact events (typhoons) could be better clarified. While the paper makes some useful contributions especially in a conceptual sense, there is need for more testing, validation, and explanation of the relative values of the diverse indicators.
The literature review is adequate in terms of publications from China, but it could be expanded to consider research conducted in other countries and include more work on indices especially related to vulnerability as well as measurement of urban flood control capabilities.
While the English is acceptable, the manuscript needs editing to reduce redundancies. There are too many maps and figures which don't much to the paper.
Author Response
Comment 1:More explanation beyond the mathematical formulation of the entropy weight assignment is needed. It is not clear if the weighting has been validated or tested in terms of the underlying logic.
Response:Thanks for reviewers' useful suggestions.We have explained more about the entropy weight method. In order to reduce the subjective influence on the weight determination, we use the entropy weight method to determine the weights of the evaluation indices. The entropy weight method has the feature of strong objectivity. Compared with methods, the weights obtained by the entropy weight method have higher precision and more objectivity, and can better reflect the influence of the evaluation indicators on the results. The scientific nature of the model has been verified in other related studies. In this paper, from the calculation results of the weights for each indicator, it is also in line with the actual situation.
Comment 2:More discussion of the data quality, accuracy, spatial resolution of the different data elements is needed. I am concerned about the different scales and resolution of data on precipitation, flooding, and impacts given the diverse topography and land use and differences in terms of drainage systems and flood control and the effects in terms of absorption, runoff, detention, retention, etc.
Response:This suggestion is helpful. The various types of data used in this article are from different government departments of Xiamen, which can guarantee the quality and accuracy of the data. We have made notes on the data sources in the manuscript.In terms of the scale and resolution of the data, it is mainly based on the district-level scale in our manuscript. This is mainly limited by data statistics and collection. Simultaneously, in the following research, through more detailed data collection and acquisition, we will explore the risk assessment of flood disasters at smaller scales.
Comment 3:Also, the description of the infrastructure system and investment is lacking. More discussion of the growth, development, and conversion of lands from natural to agricultural to urban uses would be helpful in better understanding the risks and natural and extent of flooding.
Response:Thanks for reviewers' suggestions.We have added this content to the revised version. From the economic level and infrastructure status of Xiamen City, the Siming District and Huli District are the core urban areas of Xiamen. The economic level is the highest, and the economic aggregate accounts for more than 50% of Xiamen(from 2005 to 2015). Meanwhile, the infrastructure of these two regions is relatively complete, and the investment in infrastructure assets accounts for more than 30% of the city's total investment in assets (from 2005 to 2015). The economic aggregates of Jimei District, Haishu District, Xiang'an District and Tong'an District are relatively low, and the infrastructure level is relatively low compared with the first two districts.
In addition, in future research, we will incorporate the transformation and development of land form into the assessment of flood disaster risks based on the opinions of reviewers.
Comment 4:While it is not surprising to find a large coefficient of variation for the typhoon effects, it is not clear how best to manage these risks versus more routine flooding or problem associated with 3 year average precipitation levels.
Response:Thanks for reviewers' suggestions. As aforementioned, this paper constructs three flood risk assessment scenarios using the precipitation as a scenario parameter, and analyzes the flood risk levels for Xiamen under the three different precipitation scenarios(i.e., average annual precipitation, average precipitation during flood season and precipitation during typhoon). Based on different precipitation scenarios, the risk level of flood disasters under different scenarios can be calculated. Then, by superimposing the flood risk results of the three precipitation scenarios with the flood disaster data in history, we further simulate the flood prone locations in the urban built-up area of Xiamen under different precipitation scenarios. The FDRA facilitates the identification of the flooding risk environment in the city and the results of FDRA can also be used to guide safe urban development and construction.
Comment 5:Would it make sense to better formulate the risks to health and safety in terms of death/injury versus property losses versus disruption versus flooding of agricultural or less densely populated areas?
Response:Thanks for reviewers' suggestions.In this manuscript, we mainly consider the impacts of flood disasters on the social economy. In future research, we will incorporate risk to health and safety into the indicators, then we can more comprehensively assess the risk of flood disasters.
Comment 6:It's not clear also what the unit of analysis is - especially in terms of risk management, land use controls, drainage or flood control projects or more localized interventions.
Response:Thanks for reviewers' suggestions.It should be noted that the units and attributes of selected indicators are different, which need to be standardized for calculation.The data processed by the standardization has no unit and can be directly analysed and calculated.In the revised version we made a note on the units of each indicator (see Table 1).
Comment 7:The paper provides useful general background information, but lacks sufficient details on costs and benefits associated with more specific engineering, planning, or risk reduction projects. There are too many generalities with regard to inadequate or insufficient drainage/flood control projects. From a research perspective, the paper is useful, but from a more pragmatic and directed action-oriented perspective, it doesn't identify priorities, or types of projects nor investments which can be pursued, justified, and implemented. It seems that more environmental and public health dimensions could've been emphasized in developing actions or interventions to reduce risks from flooding. Also the tradeoffs between more frequent lower intensity events versus more rare larger impact events (typhoons) could be better clarified. While the paper makes some useful contributions especially in a conceptual sense, there is need for more testing, validation, and explanation of the relative values of the diverse indicators.
Response:Thanks for reviewers' suggestions.In this paper, based on the flood risk assessment, on the one hand, we further improve the safety standards of urban flood control and drainage systems to reduce the risk level of flood disaster. On the other hand, we propose more effective disaster preventions and emergency Response mechanisms to alleviate the loss of people's property caused by disasters and provide a strong shield for the safe development of the city. The countermeasures we proposed were for all similar cities, not for a specific city, so we did not provide details of specific engineering projects and cost benefits, which reviewers mentioned. In future research, we will further deepen our research to assess the risk of flood disasters from the perspective of economic losses caused by floods and the cost-effectiveness of post-disaster reconstruction.
Comment 8:The literature review is adequate in terms of publications from China, but it could be expanded to consider research conducted in other countries and include more work on indices especially related to vulnerability as well as measurement of urban flood control capabilities.
Response:Thanks for reviewers' suggestions. In the literature review section, we gave a detailed introduction to FDRA methods. The literature we quoted is the introduction and application of relevant methods, including both Chinese publications and other national publications.
Comment 9:While the English is acceptable, the manuscript needs editing to reduce redundancies. There are too many maps and figures which don't much to the paper.
Response:Thanks for reviewers' suggestions.During the major revision, we have invited a native speaker to edit the language of our manuscript.In addition, we have made some improvements to some of the figures and tables in the manuscript, including modifying the basic errors and adding risk grading standards to the figures.
Reviewer 2 Report
Manuscript ID: ijerph-423943
Type of manuscript: Article
Title: How can cities respond to flood disaster risks under Multi-Scenario Simulation? A Case Study of Xiamen, China
General comments:
This is an interesting study about flood risk assessment in the urban area. It uses precipitation factor as a scenario parameter. Flood risk management in an urban area is very complex, especially for developing countries. Risk assessment is a key step for flood risk management. Numerous tools have developed by the scholar to assess flood risk. Authors use index evaluation method to assess flood risk. However, I think some improvements are required and for this, I recommend major revisions.
1. The objective of the study is not clear throughout the study. The objective statement in the abstract, introduction section and so on is not consistent. This may mislead the readers to understand the rationale of the study.
2. The second paragraph of the ‘Introduction’ part (line 46-61) does not have a concise argument. Several concepts are misleading. For example, risk analysis and risk assessment/ hazards and disasters, they have different interpretations and they are different. In addition, authors need to be careful to adopt different terms related disaster (e.g. flood, hazard, risk, vulnerability, capacity and so on) throughout the manuscript. Authors may consult UNISDR Terminology to frame the concept map of their study.
3. The aim of the study mention in Lines 57-61, but it is an incomplete sentence. It remains unclear about the objective and key research questions of this manuscript. There should be a thin line between the conceptual framework and literature review. In fact, the literature review and conceptual framework is the blueprint of a study. However, the conceptual framework of the research is mixed up with method and result section which is not consistent with the standard practice. The theoretical framework of the study is unclear although authors mention one sentence in lines 206-207.
4. While authors try to assess flood disaster risk, they consider ‘risk’ as one of the aspects of risk assessment (see Line 207). This should be corrected. In addition, authors need to use similar phrases which they consider as risk contributors (e.g. hazard, sensitivity, vulnerability, and prevention and resilience) throughout the manuscript. Using different terms may mislead the non-expert readers.
5. The authors can use component and sub-component/ category and sub-category, and so on, instead of using ‘First-grade indicator’ and ‘Second-grade indicator’ in Table 1. How many indicators are used? We need to read an explanation column to understand indicators. This is not a usual practice. Authors can give unique identification of each indicator to overcome this misunderstanding. Moreover, each indicator should have clear definition and reasons for selecting them. A column (Table 1) could be added to show the data sources of corresponding indicators. Please also mention the scale of the indicators. I guess some indicators are qualitative in nature.
6. Methodology: A detailed description of the data normalization is required. How to aggregate the risk components to visualize them on a map? Was there any missing data in the precipitation data? In addition, there is no information on spatial analysis techniques (e.g. use of GIS) in the method section. Some text should be moved from the result section to the method section.
7. The limitations of the study (e.g. methodological, data availability etc) need to mention.
8. The discussion section is very elaborative and tired sum. Thus, the readability of the manuscript is not good enough.
Finally, be careful to use the technical terms of disaster studies. Overall quality of English is good and understandable, but in many cases, choice of words exaggerates the interpretation. (I am not an English native speaker myself).
Technical errors:
1. Line 15: Risk assessment factors are not similar (e.g. flood control and flood control) as mentioned in the main part of the manuscript.
2. In text citations are needed for the following sentences:
a. Lines 36-37: These disasters have caused a total of 22,989,400 ----- 290.28 billion USD.
b. Lines 39-41: For example, in 2012, ----- 122 houses were burned.
c. Lines 183-185: In 2017, the resident---- was 435.18 billion CNY.
3. Lines 34-36: Reference year (from 1990 to--)
4. Lines 43-44: It is necessary to mention the reference to support the argument.
5. Line 79: I am not sure why authors used the term ‘scope’.
6. Several abbreviations (e.g. GIS, RS, DTM, SWMM, etc.) are used in this manuscript which requires elaboration.
7. Line 179: twice use of ‘reference’ word.
8. Table 1: Vulnerability→ Economic vulnerability→ Per capita: Check the text
9. Line 280 and Line 285: Is the duration of precipitation data similar? They are different in these two lines.
10. Why is the reason to select a specific typhoon? Name of the typhoon is not similar throughout the manuscript (see Line 283/ 389/392)
11. In-text citations of the figures are required, for example, Figure 4 a, Figure 5 b, and so on. Also, do the same for equations.
12. Lines 338-342: Check the number of indicators. Rewrite the sentences.
13. Line 338: Authors claims that they consider four indicators but three indicators are found.
14. Line 361: Check the sentence ‘are is’.
15. Lines 408 – 410: rewrite the sentence in a meaningful way. Looks like ambiguous.
Author Response
Comment 1:The objective of the study is not clear throughout the study. The objective statement in the abstract, introduction section and so on is not consistent. This may mislead the readers to understand the rationale of the study.
Response:Thanks for reviewers' suggestions. We have perfected the purpose of the research objective statement in the abstract and introduction section.This paper aims to establish an applicable model to assess urban flood disaster risk for achieving an optimal outcome. A scientific understanding of urban flood disasters will allow us to plan in advance of the urban planning and construction, and minimize the damage caused by the disaster.
Comment 2:The second paragraph of the “Introduction” part (line 46-61) does not have a concise argument. Several concepts are misleading. For example, risk analysis and risk assessment/ hazards and disasters, they have different interpretations and they are different. In addition, authors need to be careful to adopt different terms related disaster (e.g. flood, hazard, risk, vulnerability, capacity and so on) throughout the manuscript. Authors may consult UNISDR Terminology to frame the concept map of their study.
Response:Thanks for reviewers' suggestions. We have adjusted and improved the second paragraph of the “Introduction” part.We have refined the interpretation of relevant concepts and unified the expression of related concepts. For specific modifications, see line 63-77 in the revised version.
Comment 3:The aim of the study mention in Lines 57-61, but it is an incomplete sentence. It remains unclear about the objective and key research questions of this manuscript. There should be a thin line between the conceptual framework and literature review. In fact, the literature review and conceptual framework is the blueprint of a study. However, the conceptual framework of the research is mixed up with method and result section which is not consistent with the standard practice. The theoretical framework of the study is unclear although authors mention one sentence in lines 206-207.
Response:Thanks for reviewers' suggestions. We have perfected the purpose of the research objective statement in the abstract and introduction section.This paper aims to establish an applicable model to assess urban flood disaster risk for achieving an optimal outcome. A scientific understanding of urban flood disasters will allow us to plan in advance of the urban planning and construction, and minimize the damage caused by the disaster. In ourpaper, based on the flood risk assessment, on the one hand, we further improve the safety standards of urban flood control and drainage systems to reduce the risk level of flood disaster. On the other hand, we propose more effective disaster preventions and emergency Response mechanisms to alleviate the loss of people's property caused by disasters and provide a strong shield for the safe development of the city. The research methods and countermeasures proposed in the manuscript are all around the aim of the study.
Comment 4:While authors try to assess flood disaster risk, they consider ‘risk’ as one of the aspects of risk assessment (see Line 207). This should be corrected. In addition, authors need to use similar phrases which they consider as risk contributors (e.g. hazard, sensitivity, vulnerability, and prevention and resilience) throughout the manuscript. Using different terms may mislead the non-expert readers.
Response:Thanks for reviewers' suggestions. We made adjustments to this part.When assessing the risk of flood disasters, we mainly consider from four aspects: hazard of disaster factors, sensitivity to environment, vulnerability of the disaster-bearing body and the city's disaster prevention capability. In addition, we have unified the terms in the entire manuscript.
Comment 5:The authors can use component and sub-component/ category and sub-category, and so on, instead of using ‘First-grade indicator’ and ‘Second-grade indicator’ in Table 1. How many indicators are used? We need to read an explanation column to understand indicators. This is not a usual practice. Authors can give unique identification of each indicator to overcome this misunderstanding. Moreover, each indicator should have clear definition and reasons for selecting them. A column (Table 1) could be added to show the data sources of corresponding indicators. Please also mention the scale of the indicators. I guess some indicators are qualitative in nature.
Response:Thanks for reviewers' suggestions. We have perfected the contents of Table 1. We use category and sub-category, and so on, instead of using ‘First-grade indicator’ and ‘Second-grade indicator’. We have increased the interpretation of each indicator (line746-833) and added units for each indicator(Table 1).
Comment 6:A detailed description of the data normalization is required. How to aggregate the risk components to visualize them on a map? Was there any missing data in the precipitation data? In addition, there is no information on spatial analysis techniques (e.g. use of GIS) in the method section. Some text should be moved from the result section to the method section.
Response:Thanks for reviewers' suggestions. We introduced the data standardization method (line846-900) in the manuscript. All the analysis and maps in the manuscript is based on GIS software (spatial analysis function). Since these operations are the basic operations in GIS, we have not specifically introduced the specific implementation operations. In addition, we have added all the standards of dividing the risk level of each indicator into the figures（figure2-7）.
Comment 7:The limitations of the study (e.g. methodological, data availability etc.) need to mention.
Response:Thanks for reviewers' suggestions. We have added a discussion of the limitations of the research in the conclusion section. Our studies expand the research on flood disaster risk assessment, and also provide reference for relevant cities to deal with flood disaster. Although the method proposed in this paper realizes the urban flood disaster risk assessment under different scenarios, however, it is still a static assessment, and is impossible to realize real-time monitoring and dynamic assessment of urban flood disasters. In the following research, we will further deepen the methods of simulating flood disaster scenarios and methods of flood disaster risk assessment, and continuously expand the research methods and scale on flood disaster risk assessment. Simultaneously, through more detailed data collection and acquisition, we will explore the risk assessment of flood disasters at smaller scales.
Comment 8:The discussion section is very elaborative and tired sum. Thus, the readability of the manuscript is not good enough.
Response:Thanks for reviewers' suggestions.We have modified and improved the language and content of the conclusion section.In addition, we have adjusted the content of the flood control zones to the results section.
Comment 9:Finally, be careful to use the technical terms of disaster studies. Overall quality of English is good and understandable, but in many cases, choice of words exaggerates the interpretation.
Response:Thanks for reviewers' suggestions. During the major revision, we have invited a native speaker to edit the language of our manuscript. Meanwhile, we have unified the academic terms in the manuscript.
Comment 10:Line 15: Risk assessment factors are not similar (e.g. flood control and flood control) as mentioned in the main part of the manuscript.
Response:Thanks for reviewers' suggestions. We have unified the risk assessment factors in the manuscript.We modified the "flood control and flood control" ,which mentioned by the reviewer, and revised it to "flood disaster prevention and resilience".
Comment 11:In text citations are needed for the following sentences: a. Lines 36-37: These disasters have caused a total of 22,989,400 ----- 290.28 billion USD. b. Lines 39-41: For example, in 2012, ----- 122 houses were burned. c. Lines 183-185: In 2017, the resident---- was 435.18 billion CNY.
Response:Thanks for reviewers' suggestions.We have revised and improved these three sentences. The specific modifications are as follows：a)the world has suffered 12,547 major natural disasters with a total of 22,989,400 deaths and economic losses of up to $290.28 billion. b)For example, in 2012, Hurricane Sandy caused widespread flooding of road networks along the coast of New York City, where firefighting vehicles could not reach fire sites and 122 houses were burned. c)n 2017, the resident population of Xiamen was 4.01 million, the urbanization rate reached 89.1% and its GDP was 435.18 billion CNY(approximately $64.95 billion ).
Comment 12:Lines 34-36: Reference year (from 1990 to--)
Response:Thanks for reviewers' suggestions. The reference year is from 1990 to 2017.
Comment 13:Lines 43-44: It is necessary to mention the reference to support the argument.
Response:Thanks for reviewers' suggestions. We have add the reference to the argument.
Comment 14:Line 79: I am not sure why authors used the term ‘scope’.
Response:Thanks for reviewers' suggestions. We didn't use the term ‘scope’, so we didn't make changes.
Comment 15:Several abbreviations (e.g. GIS, RS, DTM, SWMM, etc.) are used in this manuscript which requires elaboration.
Response:Thanks for reviewers' suggestions. We have added elaborations to these abbreviations.1) storm water management model (SWMM); 2)geographic information system (GIS); 3)digital elevation model (DEM); 4)remote sensing (RS)
Comment 16:Line 179: twice use of ‘reference’ word.
Response:Thanks for reviewers' suggestions. We have deleted the extra word "reference".
Comment 17:Table 1: Vulnerability→Economic vulnerability→Per capita: Check the text
Response:Thanks for reviewers' suggestions. We have checked and modified the related terms in Table 1.
Comment 18:Line 280 and Line 285: Is the duration of precipitation data similar? They are different in these two lines.
Response:Thanks for reviewers' suggestions.We modified the error on line 285 and changed the duration of precipitation data to “from 2005 to 2015”.
Comment 19:Why is the reason to select a specific typhoon? Name of the typhoon is not similar throughout the manuscript (see Line 283/ 389/392)
Response:Thanks for reviewers' suggestions.We have unified the name of the typhoon in the manuscript and changed it to "Meranti". According to the data in the Xiamen Statistical Meteorological Bulletin, during the typhoon Meranti in 2016, there was a strong precipitation in Xiamen. The cumulative precipitation in the whole process was 201.2 mm, which reached the magnitude of heavy storms. The typhoon caused serious casualties and direct economic losses. Typhoon Meranti is the most serious typhoon in Xiamen’s history, and it is used to simulate the flood disaster risk of Xiamen under extreme precipitation such as typhoon
Comment 20:In-text citations of the figures are required, for example, Figure 4 a, Figure 5 b, and so on. Also, do the same for equations.
Response:Thanks for reviewers' suggestions. The figures in the manuscript are original by the authors, so there is no citations.Meanwhile,we have added the citations to the formulas referenced in the methods section.
Comment 21:Lines 338-342: Check the number of indicators. Rewrite the sentences. Line 338: Authors claims that they consider four indicators but three indicators are found.
Response:Thanks for reviewers' suggestions. In this paper, four indicators are identified: population density, GDP per capita, building quality and building age. The population density is used for the analysis of economic vulnerability, the GDP per capita is used for the analysis of population vulnerability, and building quality and building age is used for the analysis of building vulnerability.
Comment 22:Line 361: Check the sentence ‘are is’.
Response:Thanks for reviewers' suggestions. We have perfected this sentence and changed it to “while the disaster prevention ability of Xiangan District and Tongan District in the urban fringe area is relatively weaker (see as Figure 6)”.
Comment 23:Lines 408 – 410: rewrite the sentence in a meaningful way. Looks like ambiguous.
Response:Thanks for reviewers' suggestions. We have perfected this sentence and changed it to "By superimposing the flood risk results of the three precipitation scenarios with the flood disaster data in history , we further simulate the flood prone locations in the urban built-up area of Xiamen under different precipitation scenarios. As shown in Figure 8, the flood prone locations in Xiamen are mainly distributed within the urban built-up area, and the flood prone locations distributed in the central urban area are more likely to occur than other areas. "
Reviewer 3 Report
This paper constructs indicator system and researches on flood disaster risk assessment, which is quite meaningful for urban flood management. The literature review in this paper is relatively detailed. And the index system selected from the perspectives of hazard factors, sensitivity to the environment, vulnerability, and disaster prevention and resilience is fairly comprehensive. However, I believe that the manuscript needs some improvements before being published. Major comments: 1. The index selection of the index system should be feasible, representative and independent. The indicators selected in this paper, such as population density, per capita GDP and urbanization level, have a certain correlation. River network distribution and river network density have partially overlapped. Thus the index system may seem to be not independent enough. It will be better to optimize or adjust the index system appropriately. 2. The method of dividing the risk level of each indicator should be further described. And the number of risk levels for the indicators is different. For example, the risk level of river factor is divided into four levels, while the population density level is divided into seven levels. It is necessary to clearly explain the specific method of dividing the risk level and the threshold of each index at different levels. 3. The difference in spatial distribution of risk level of land factor, building vulnerability and flood control capacity, is quite coarse. I don’t agree that the risk level of land factor is divided according to the administrative region. It should be more detailed. Please offer a more detailed spatial distribution of risk level of land factor, building vulnerability and flood control capacity. Otherwise, please give a reasonable explanation. 4. Flood disasters usually refer to natural disasters caused by short-duration heavy rainfall or long-duration rainfall events. The first and second rainfall scenarios selected in this paper are average rainfalls, which are not sufficiently correlated with flood disasters. Extreme rainfall events are better here, such as the maximum one day rainfall, the maximum three day rainfall, and so on. In addition, the paths of different typhoons vary greatly, which will make some differences in precipitations. What is the basis for this paper to select the precipitation during the Morante typhoon to represent the rainfall scenario under extreme disaster conditions? 5. The results of the risk assessment (Figure 7) indicate that the risk of flooding in the northern mountainous areas is higher, but historical flood vulnerable points and simulated flood vulnerable points (Figure 8) are concentrated in the southern built-up areas. Please explain this result in the discussion. In addition, the tools, methods, or principles used in the flood prone locations simulation in Section 4.5 should be supplemented. 6. From the current description, the discussion section seems not relevant to the results above, or it doesn’t use the results above. And some explanations should also be added to illustrate the division basis and variance of the flood control zone and the drainage zone. Please rewrite the discussion section with the current results. Minor comments: 1. Line 233 - There are some grammatical mistakes in the text, such as “flood control and flood control” (15 lines) and “per capita GDP and GDP per capita”. 2. The expression in Figure 9 is not clear enough and the corresponding legend should be added appropriately. 3. The typefaces in the figures should be with the journaAuthor Response
Comment 1:The index selection of the index system should be feasible, representative and independent. The indicators selected in this paper, such as population density, per capita GDP and urbanization level, have a certain correlation. River network distribution and river network density have partially overlapped. Thus the index system may seem to be not independent enough. It will be better to optimize or adjust the index system appropriately.
Response:These suggestions are helpful.We have refined the interpretation of the indicators and added the units for each indicator in Table 1.Although several indicators mentioned by reviewers have certain relevance, they are used for risk analysis in different aspects, and the analytical data used is different. In future research, we will improve the indicator system to increase the independence of each indicator.
Comment 2:The method of dividing the risk level of each indicator should be further described. And the number of risk levels for the indicators is different. For example, the risk level of river factor is divided into four levels, while the population density level is divided into seven levels. It is necessary to clearly explain the specific method of dividing the risk level and the threshold of each index at different levels.
Response:Thanks for reviewers' suggestions. In our article, we use the Natural Breaks method to classify disaster risk levels. The Natural Breaks method is based on the natural grouping inherent in the data. The classification interval is identified, the similarity values can be optimally grouped, and the differences between the classes can be maximized. We have added the standards for dividing the risk level of each indicator to the figures (Figure 2 to 7). In addition, We have made adjustments to the division of the population density level and divided it into five levels.
Comment 3:The difference in spatial distribution of risk level of land factor, building vulnerability and flood control capacity, is quite coarse. I don’t agree that the risk level of land factor is divided according to the administrative region. It should be more detailed. Please offer a more detailed spatial distribution of risk level of land factor, building vulnerability and flood control capacity. Otherwise, please give a reasonable explanation.
Response:Thanks for reviewers' suggestions. The data we use is collected on a district-level scale. Limited by the data, we analyzed the spatial distribution of land risk, building vulnerability and flood control capacity based on administrative units. This is also the limitation and deficiency of our manuscript. In future research, through more detailed data collection and acquisition, we will explore the risk assessment of flood disasters at smaller scales.
Comment 4:Flood disasters usually refer to natural disasters caused by short-duration heavy rainfall or long-duration rainfall events. The first and second rainfall scenarios selected in this paper are average rainfalls, which are not sufficiently correlated with flood disasters. Extreme rainfall events are better here, such as the maximum one day rainfall, the maximum three day rainfall, and so on. In addition, the paths of different typhoons vary greatly, which will make some differences in precipitations. What is the basis for this paper to select the precipitation during the Morante typhoon to represent the rainfall scenario under extreme disaster conditions?
Response:Thanks for reviewers' suggestions. In this paper, precipitation is used as a scenario factor to consider the level of urban flood risk under different precipitation conditions(annual average precipitation scenarios, multi-year flood season average precipitation scenarios and large typhoon precipitation scenarios). Among them, the annual average precipitation scenario represents the flood risk under general precipitation conditions, the multi-year flood season average precipitation scenario represents the flood risk under general disaster occurrence conditions, and the large typhoon precipitation scenario represents the flood risk under extreme disaster occurrence conditions. In the case of extreme disasters, we chose Typhoon Meranti as the basis for the simulation. According to the data in the Xiamen Statistical Meteorological Bulletin, during the Typhoon Meranti in 2016, there was a strong precipitation in Xiamen. The cumulative precipitation in the whole process was 201.2 mm, which reached the magnitude of heavy storms. The typhoon caused serious casualties and direct economic losses. Typhoon Meranti is the most serious typhoon in Xiamen's history, and it is used to simulate the flood disaster risk of Xiamen under extreme precipitation such as typhoon.
Comment 5:The results of the risk assessment (Figure 7) indicate that the risk of flooding in the northern mountainous areas is higher, but historical flood vulnerable points and simulated flood vulnerable points (Figure 8) are concentrated in the southern built-up areas. Please explain this result in the discussion. In addition, the tools, methods, or principles used in the flood prone locations simulation in Section 4.5 should be supplemented.
Response:Thanks for reviewers' suggestions.
Comment 6:From the current description, the discussion section seems not relevant to the results above, or it doesn’t use the results above. And some explanations should also be added to illustrate the division basis and variance of the flood control zone and the drainage zone. Please rewrite the discussion section with the current results.
Response:Thanks for reviewers' suggestions. Although the risk of flooding in the northern mountainous areas is higher, the existing disaster database records the floods in the urban built-up area of Xiamen City. Therefore, the recorded flood prone locations are also distributed in urban built-up areas. Meanwhile, in this paper, when simulating the flood prone locations in different scenarios, it is mainly targeted at urban built-up areas. According to the flood disaster risk assessment results obtained under different simulation scenarios and the flood disaster prone locations, flood control zones can be divided. When delineating the flood control zones, on the one hand, we take the urban construction land as the basis, and comprehensively consider the factors such as the layout of the water conservancy project, the flood prone locations and the administrative authority. On the other hand, we fully consider factors such as water system, terrain and drainage network. While maintaining the integrity of the water system, we rationally arrange the drainage methods in different terrain height areas to avoid dividing the high-lying and easily drained areas and low-lying areas into the same flood control zone.
Comment 7:Line 233 - There are some grammatical mistakes in the text, such as “flood control and flood control” (15 lines) and “per capita GDP and GDP per capita”.
Response:Thanks for reviewers' suggestions. We have revised these grammatical mistakes in the manuscript. a)“flood disaster prevention and resilience”（line 18）; b)“Per capita GDP(Unit: CNY)”(line 240).In addition, during the major revision, we have invited a native speaker to edit the language of our manuscript.
Comment 8: The expression in Figure 9 is not clear enough and the corresponding legend should be added appropriately.
Response:Thanks for reviewers' suggestions. We made adjustments to Figure 9 and added a legend to it.
Comment 9:The typefaces in the figures should be with the journal
Response:Thanks for reviewers' suggestions.According to the requirements of the journal, we have made adjustments to the fonts in all the figures.
Round 2
Reviewer 1 Report
While the paper is somewhat improved, I'm still concerned that sufficient details regarding the drainage and flood control system and the impacts of the built environment, especially roadways and other impervious surfaces should be better accounted for in the paper or at least acknowledged in terms of the limitations of the research. Because the paper focuses on a specific city, I think that there were missed opportunities to address more specific actions, costs and benefits of mitigation strategies. The authors could do more to make this research policy and practice relevant. While the English language is improved, there are still some syntax errors and excessive wordiness and redundancies that could be reduced.
Author Response
Comment 1:While the paper is somewhat improved, I'm still concerned that sufficient details regarding the drainage and flood control system and the impacts of the built environment, especially roadways and other impervious surfaces should be better accounted for in the paper or at least acknowledged in terms of the limitations of the research.
Response:Thanks for reviewers' useful suggestions.Different types of land use affect the natural water regulation capacity and the runoff of rainwater on the surface differently. The soil in the forest area has strong water seepage ability, and it is not easy to form stagnant water. The water seepage capacity of agricultural and forestry land is moderate. The ground surface of the urban built-up area is dominated by a hard surface and the water seepage capacity is the worst.However, on the city scale, it is relatively difficult to obtain the data on the impervious surface of the road. This paper uses the proportion of cultivated land is used to indicate the water seepage capacity of the land factor and obtain the evaluation results of the land factors.Although the method proposed in this paper realizes the urban flood disaster risk assessment under different scenarios, however, it is still a static assessment, and is impossible to realize real-time monitoring and dynamic assessment of urban flood disasters. In the following research, we will further deepen the methods of simulating flood disaster scenarios and methods of flood disaster risk assessment, and continuously expand the research methods and scale on flood disaster risk assessment. Simultaneously, through more detailed data(e.g. the drainage and flood control system and the impacts of the built environment)collection and acquisition, we will explore the risk assessment of flood disasters at smaller scales.
Comment 2:Because the paper focuses on a specific city, I think that there were missed opportunities to address more specific actions, costs and benefits of mitigation strategies. The authors could do more to make this research policy and practice relevant.
Response:This suggestion is helpful. In this paper, based on the flood risk assessment, on the one hand, we further improve the safety standards of urban flood control and drainage systems to reduce the risk level of flood disaster. On the other hand, we propose more effective disaster preventions and emergency Response mechanisms to alleviate the loss of people's property caused by disasters and provide a strong shield for the safe development of the city. The countermeasures we proposed were for all similar cities, not for a specific city, so we did not provide details of specific engineering projects and cost benefits, which reviewers mentioned. In future research, we will further deepen our research to assess the risk of flood disasters from the perspective of economic losses caused by floods and the cost-effectiveness of post-disaster reconstruction, and do more to make this research policy and practice relevant.
Comment 3:While the English language is improved, there are still some syntax errors and excessive wordiness and redundancies that could be reduced.
Response:Thanks for reviewers' suggestions. We have made several changes to the language of the manuscript to make it easier to understand.
Reviewer 2 Report
The author has improved the content and organization of the paper. However, there are a few issues that need to be dealt with.
1. Line 47: ‘resulting in 123 deaths in the state, including 14 from a local nursing home The hurricane has also’
A punctuation mark is required.
2. Line 79: ‘the quantitative analysis of the causes, probability, scope and intensity of flood disaster’
Check the references whether the word ‘scope’ is written.
Author Response
Comment 1:WThe author has improved the content and organization of the paper. However, there are a few issues that need to be dealt with.
1. Line 47: ‘resulting in 123 deaths in the state, including 14 from a local nursing home The hurricane has also’ -A punctuation mark is required.
2. Line 79: ‘the quantitative analysis of the causes, probability, scope and intensity of flood disaster’ -Check the references whether the word ‘scope’ is written.
Response:Thanks for reviewers' useful suggestions.We made changes to these two errors in the revised vision.
1) A punctuation mark isadded. The specific content is "In 2017, the hurricane Irma hit Florida, resulting in 123 deaths in the state, including 14 from a local nursing home. "
2) We deleted the word "scope" and changed the sentence to "The FDRA mainly focuses on the quantitative analysis of the causes, probability and intensity of flood disaster [8-11]."
Reviewer 3 Report
The author's opinions have been revised according to the evaluation opinions, and the requirements have been met.
Author Response
Comment 1:The author's opinions have been revised according to the evaluation opinions, and the requirements have been met.
Response:Thanks to reviewers for their contributions to the revision and improvement of the manuscript.
