A Weighted k-Nearest-Neighbors-Based Spatial Framework of Flood Inundation Risk for Coastal Tourism—A Case Study in Zhejiang, China

Round 1

Reviewer 1 Report

Dear Authors, 

it was a pleasure to read your paper also for the complexity of the research.

Just a question: what about the airports in the area of interest? 

Regards 

Author Response

Response to Reviewer #1

Comment: what about the airports in the area of interest?

Response:

Appreciate your extremely useful comments. We have added the following context to discuss the airports’ situation of FIR in the study area and this has been added to the end of the fifth paragraph of Section 4.4 on page 14.

Figure 5. Tourist facilities located in FIR areas, (a) hotels, (b) medical treatment institutions, (c) parks, (d) parking places, (e) Restaurants, and (f) national and provincial roads and airports.

Airports play an important role in modern tourism, such as providing an easier way for tourists to travel and increasing tourist arrivals. In the study area, although airports are impacted by flood inundation, most of them are located in the low or very low FIR (Figure 5f). There are only four airports covered by Medium FIR including Zhoushan Putuoshan Airport (Number 3), Ningbo LiShe International Airport (Number 4), Yiwu Airport (Number 5), and Lishui Airport (Number 8) since these airports are located in medium-high riks in precipitation, elevation and SWR. Two airports, Jiaxing Nanhu Airport (Number 1) and Quzhou Airport (Number 6), are located in Low FIR. The rest 3 airports, Hangzhou Xiaoshan International Airport (Number 2), Taizhou Luqiao Airport (Number 7), and Wenzhou Longwan International Airport (Number 9), are situated in the very-low risk. These airports in low or very low areas are of great importance in evacuating passengers in case of flood inundation disasters under extremes.

Author Response File: Author Response.pdf

Reviewer 2 Report

No further corrections are required.

Author Response

Response to Reviewer #2

Response:

Thank you very much for your hard work and contribution to the paper.

Reviewer 3 Report

Followings are my comments and suggestions.

Abstract:

The abstract provides a concise summary of the paper, highlighting the main objectives, methodology, and findings. It effectively conveys the purpose of the study and the significance of the proposed spatial framework. However, it would be helpful to include specific results or conclusions in the abstract to provide a clearer understanding of the paper's contribution.

Introduction:

The introduction provides a good background on the importance of flood risk assessment for coastal tourism and the challenges associated with it. The research gap is clearly identified, and the objectives of the study are well-defined. However, it would be beneficial to provide more context on the specific location (Zhejiang, China) and its relevance to the broader field of flood risk assessment.

Framework Development:

The section provides a clear explanation of the basic principle of kNN and its application in classifying examined objects. The two steps involved in classifying the categories of examined objects are well-described. However, it would be helpful to provide more details on the weighted kNN (WkNN) approach and how it improves upon the traditional kNN method. Additionally, referencing the source (Liu, Liu, and Tan [31]) for further information on the characteristics of examined objects would enhance the credibility of the framework.

Framework Conceptualization:

The section effectively presents the conceptual framework of the WkNN-based spatial framework for flood inundation risk (FIR) assessment. The three parts of the framework (input collection, model construction, and output classification) are clearly explained. The inclusion of Figure 1 enhances the understanding of the framework. However, it would be beneficial to provide more details on the specific flood-induced factors and flood hazard data used in the framework. Additionally, explaining the rationale behind the division of the standardized datasets into training and testing datasets would provide more clarity.

Case Study:

The section provides a comprehensive description of the study area (Zhejiang province) and its characteristics. The inclusion of Figure 2 and Table 1 enhances the understanding of the collected data and soil types. The section effectively highlights the flood risk faced by the study area due to sea levels, typhoons, and tropical cyclones. However, it would be helpful to provide more information on the historical records of flood events and their impact on tourism in the area. Additionally, including specific examples or statistics on the tourism facilities exposed to flood inundation would strengthen the case study.

Results and Discussion:

The section presents the results of the WkNN model and their implications for flood inundation risk in coastal tourism. The inclusion of Figure 5 provides visual representation of the distribution of tourism facilities in different risk areas. The discussion of the results in relation to the role of engineering measures and the significance of medical treatment institutions is insightful. However, it would be beneficial to provide more detailed analysis and interpretation of the results, including the sensitivity analysis conducted to explore the relationship between inputs and outputs of the model.

Conclusions:

The conclusions effectively summarize the main findings of the study and their implications for flood inundation risk assessment in coastal tourism. The mention of the innovative spatial framework and its integration of WkNN, GIS, and flood-relative indices is noteworthy. However, it would be helpful to provide more specific conclusions based on the results presented in the paper. Additionally, highlighting the limitations of the study and suggesting avenues for future research would enhance the completeness of the conclusions.

The English language used in the document is generally clear and understandable. The sentences are well-structured, and the vocabulary is appropriate for the subject matter. However, there are a few instances of grammatical errors and punctuation inconsistencies throughout the document. Proofreading for these errors would enhance the overall quality of the writing.

Author Response

Response to Reviewer #3

Followings are my comments and suggestions.

Comment 1.

Abstract:

The abstract provides a concise summary of the paper, highlighting the main objectives, methodology, and findings. It effectively conveys the purpose of the study and the significance of the proposed spatial framework. However, it would be helpful to include specific results or conclusions in the abstract to provide a clearer understanding of the paper's contribution.

Response:

Thanks for your extremely useful comments.

Among the matched areas, 80.14%, 90.13%, 65.50%, and 84.14% of the predicted categories using WkNN perfectly coincide with the MIE in high, medium, low, and very-low risk, respectively. For the whole study area, approximately 2.85%, 64.83%, 10.8%, and 21.51% are covered by high risk, medium risk, low risk, and very low risk of flood inundation. Precipitation and elevation make a negative contribution to high-medium risk, and drainage systems positively alleviate the regional stress of FIR. The evaluation results illustrate that in most inland areas, some tourism facilities are located in high-medium risk areas of flood inundation. However, most tourism facilities in coastal cities are at low or very low risk, especially from Hangzhou-centered northern coastal areas to southern Wenzhou areas.

The above specific results or conclusions have been added to the end of the abstract.

Comment 2.

Introduction:

The introduction provides a good background on the importance of flood risk assessment for coastal tourism and the challenges associated with it. The research gap is clearly identified, and the objectives of the study are well-defined. However, it would be beneficial to provide more context on the specific location (Zhejiang, China) and its relevance to the broader field of flood risk assessment.

Response:

Do appreciate your precious comments.

In Zhejiang, about 74.63% of the area is occupied by mountains and hills in which relatively steep terrain and extreme precipitation will easily cause flood inundation with the limitation of river flows and fast water accumulation [1]. Moreover, the whole area is deeply affected by a subtropical monsoon climate which brings heavy rainfall between June to October. Especially, Its eastern area is frequently impacted by typhoons which regularly originate between June and October[2]. This period happens to be the best tourist season in Zhejiang and China, which is heavily and widely influenced by the wet season [3-5].

The above context with references to the specific location of Zhejiang and its relevance to flood risk assessment has been added to the beginning of section 3.1.

Comment 3.

Framework Development:

The section provides a clear explanation of the basic principle of kNN and its application in classifying examined objects. The two steps involved in classifying the categories of examined objects are well-described. However, it would be helpful to provide more details on the weighted kNN (WkNN) approach and how it improves upon the traditional kNN method. Additionally, referencing the source (Liu, Liu, and Tan [31]) for further information on the characteristics of examined objects would enhance the credibility of the framework.

Response:

Thanks for your thorough review.

We have added the following content to the beginning of section 2.1.

The core of kNN is based on the similarity or distance between two points which means the properties or classification of query points are more affected by the closest points than those farther away.

We have added the following explanation to the end of section 2.2 to clarify how the weighted kNN (WkNN) approach can improve upon the traditional kNN method.

Equation (3) shows Inverse Distance Weighting, the weight of a neighbor is inversely proportional to its distance from the query point. Equation (4) shows that WkNN introduces the concept of assigning weights to the neighboring data points based on their proximity to the query point. These weights are used to influence the final classification or prediction. As a result, closer neighbors have a greater influence on the prediction, while farther neighbors have a reduced impact.

Comment 4.

Framework Conceptualization:

The section effectively presents the conceptual framework of the WkNN-based spatial framework for flood inundation risk (FIR) assessment. The three parts of the framework (input collection, model construction, and output classification) are clearly explained. The inclusion of Figure 1 enhances the understanding of the framework. However, it would be beneficial to provide more details on the specific flood-induced factors and flood hazard data used in the framework. Additionally, explaining the rationale behind the division of the standardized datasets into training and testing datasets would provide more clarity.

Response: Thanks for your extremely useful comments.

We did two main works based on your comments. First, we have provided more details on every flood-induced factor and flood hazard data in section 3.2. Please refer to the added contents in red in section 3.2. Second, we further add an explanation of why the study chose 70% and 30% as the rationale behind the division of the standardized datasets into training and testing datasets.

The standardized datasets within the extent of MIE were divided into two parts: 70% training dataset and 30% testing dataset [6], which is not an inflexible rule. It can vary depending on the size of your dataset and problem. Usually, The larger portion of the data is allocated to training because the model needs to learn from a significant amount of information. A larger training set can help the model capture the underlying patterns and relationships in the data.

The above explanation has been added to the third paragraph of section 2.3.

Comment 5.

Case Study:

The section provides a comprehensive description of the study area (Zhejiang province) and its characteristics. The inclusion of Figure 2 and Table 1 enhances the understanding of the collected data and soil types. The section effectively highlights the flood risk faced by the study area due to sea levels, typhoons, and tropical cyclones. However, it would be helpful to provide more information on the historical records of flood events and their impact on tourism in the area. Additionally, including specific examples or statistics on the tourism facilities exposed to flood inundation would strengthen the case study.

Response:

These are very useful assessments.

We have added the following context to the third part of section 3.1.

Newspapers are extremely important data sources for data collection and verification. It can be employed as truly historical data records since its reliability and timeliness are relatively high. Therefore, historical flood disaster data for the study were collected and analyzed between 1950 and 2022 from Zhejiang Daily (https://zjrb.zjol.com.cn/). The statistical results show flood disaster events have negative impacts on Zhejiang province and its tourism facilities. For example, more than 250 and 219 kilometers of roads were washed away caused by continuous rainfall in Jiaxing city in July 1983. In June 1989, the transmission line in Jingning County was interrupted for 25 hours, and two hydropower stations were shut down by rainstorms, causing a temporary shutdown of industries in the county. The Yunhe section of the highway between Jingning to Yunhe collapsed seriously, and houses, warehouses, and shops were flooded by continuous rainfall. In July 1999, more than 900 enterprises ceased production and semi-suspended, of which 94.6 percent of those with sales revenues of more than 5 million yuan were affected by the flood inundation disaster in Changxing county.

Comment 6.

Results and Discussion:

The section presents the results of the WkNN model and their implications for flood inundation risk in coastal tourism. The inclusion of Figure 5 provides visual representation of the distribution of tourism facilities in different risk areas. The discussion of the results to the role of engineering measures and the significance of medical treatment institutions is insightful. However, it would be beneficial to provide more detailed analysis and interpretation of the results, including the sensitivity analysis conducted to explore the relationship between inputs and outputs of the model.

Response: Thanks for your tremendously valuable comments.

We have added the following context to section 4.2.

The study explored the relationship between sampling times and the tendency of EA under  values. Overall Accuracy (OA) was chosen to evaluate the performance of kNN and WNN against MIE data. OA denotes the proportion of correct predictions made by models or systems over the total number of predictions [7,8], which can directly reflect EA and is easy to understand and use. Figure 4A shows the larger the k value, the higher the OA accuracy since the EA curve distribution ( = 5, blue points) is significantly lower than the EA curve distribution ( = 95, green points).

We have added the following context to the end of the first paragraph in section 4.4.

Elevation can contribute to flash flooding in mountainous regions. Heavy rainfall or rapid rainstorms at higher elevations can lead to the sudden release of large volumes of water downstream, causing flash floods in lower-lying areas. Meanwhile, heavy rainfall over a short period can overwhelm the capacity of rivers and stormwater systems to handle the water, leading to flash floods.

We have added the following context to the beginning of the second paragraph in section 4.4.

In areas with a steep slope, such as mountainous or hilly regions in the study area, water flows downhill more rapidly. When heavy rainfall occurs in these areas, the water can quickly run off the slopes and accumulate in lower-lying regions, potentially causing flash floods. The steepness of the terrain can lead to a high runoff speed and increased water volume downstream. Soils with high water retention, such as clay soils in the study area, may have slower infiltration rates. This can lead to increased surface run-off during heavy rainfall events, which may contribute to flash flooding if the rainfall rate exceeds the soil's infiltration capacity.

We have added the following context to the middle of the fourth paragraph in section 4.4.

That is because high-risk areas are often scenic and attractive due to their proximity to water bodies, such as rivers, lakes, or the ocean. Some flood-prone areas may have historical or cultural significance, such as old towns or heritage sites. All these make the number of medical treatment institutions in high-risk areas more than other facilities for responding to disaster relief and casualties.

Comment 7.

Conclusions:

The conclusions effectively summarize the main findings of the study and their implications for flood inundation risk assessment in coastal tourism. The mention of the innovative spatial framework and its integration of WkNN, GIS, and flood-relative indices is noteworthy. However, it would be helpful to provide more specific conclusions based on the results presented in the paper. Additionally, highlighting the limitations of the study and suggesting avenues for future research would enhance the completeness of the conclusions.

Appreciate your enormously beneficial comments. We have added the following context to the first part of section 5.

It was illustrated using in Chinese case study in Zhejiang province where flood inundation risk is highly related to environmental variabilities and extreme weather events such as typhoons which bring about long-term or intensive rainfall. All these environmental criteria have diverse and complicated contributions to flood hazards. In the study, the weights, derived from the distance between query points and their k nearby neighbors, were selected as the fit-for-purpose evaluation criteria in the WkNN-based FIR assessment. The evaluation results show that precipitation and elevation make a huge contribution to high-medium risk, and drainage systems positively alleviate the regional stress of FIR.

Comment 8.

Comments on the Quality of English Language

The English language used in the document is generally clear and understandable. The sentences are well-structured, and the vocabulary is appropriate for the subject matter. However, there are a few instances of grammatical errors and punctuation inconsistencies throughout the document. Proofreading for these errors would enhance the overall quality of the writing.

Response: Thanks for your valuable comments.

We will grammatical errors and punctuation inconsistencies throughout the document in the process of proofreading.

1. Li, L.; Li, Z.; He, Z.; Yu, Z.; Ren, Y. Investigation of storm tides induced by super typhoon in Macro-Tidal Hangzhou Bay. Frontiers in Marine Science 2022, 9, doi:10.3389/fmars.2022.890285.
2. Cui, Y.-l.; Hu, J.-h.; Xu, C.; Zheng, J.; Wei, J.-b. A catastrophic natural disaster chain of typhoon-rainstorm-landslide-barrier lake-flooding in Zhejiang Province, China. Journal of Mountain Science 2021, 18, 2108-2119.
3. Zhang, W.; Wang, W.; Lin, J.; Zhang, Y.; Shang, X.; Wang, X.; Huang, M.; Liu, S.; Ma, W. Perception, knowledge and behaviors related to typhoon: A cross sectional study among rural residents in Zhejiang, China. International Journal of Environmental Research and Public Health 2017, 14, 492.
4. Liao, X.; Xu, W.; Zhang, J.; Qiao, Y.; Meng, C. Analysis of affected population vulnerability to rainstorms and its induced floods at county level: A case study of Zhejiang Province, China. International Journal of Disaster Risk Reduction 2022, 75, 102976.
5. Cao, F.; Xu, X.; Zhang, C.; Kong, W. Evaluation of urban flood resilience and its space-time evolution: A case study of Zhejiang Province, China. Ecological Indicators 2023, 154, 110643.
6. Batchuluun, G.; Nam, S.H.; Park, K.R. Deep learning-based plant-image classification using a small training dataset. Mathematics 2022, 10, doi:10.3390/math10173091.
7. Foody, G.M. Status of land cover classification accuracy assessment. Remote Sensing of Environment 2002, 80, 185-201.
8. Liu, C.; Frazier, P.; Kumar, L. Comparative assessment of the measures of thematic classification accuracy. Remote Sensing of Environment 2007, 107, 606-616.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

All of my comments have been addressed.

Quality of english language is fine