As climate change continues to worsen, coastal areas are increasingly vulnerable to more frequent and severe storm surges. This poses a significant risk to economic entities, particularly in areas that have undergone rapid development. However, quantitative assessment of economic losses from storm surge disasters in China has been challenging due to limited exposure and vulnerability data. This study proposes a framework for comprehensive economic losses assessment of storm surge disasters using open data, focusing on Zhoushan City as an example. The study quantifies economic loss ratios caused by storm surges by identifying essential urban land use/cover (EULUC) and considering the water depth of different EULUC types for quantitative vulnerability assessment. The study then calculates direct economic losses using the loss ratio maps and gridded gross domestic product data and quantifies indirect economic losses (IEL) using an input–output model to account for inter-industry correlation. Results show that under the scenario of a super typhoon intensity (915 hPa), the total economic loss can reach 131 million CNY, with IEL accounting for 60% of the total. The construction and industrial sectors experience higher IEL due to excessive dependence on upstream and downstream industries, with IEL accounting for approximately 70%.

  • A comprehensive economic losses assessment framework for storm surge flooding is proposed.

  • A quantitative vulnerability assessment method using open data is proposed.

  • The direct economic loss is calculated based on the refined sub-sectors GDP gridded data.

  • The calculation of indirect economic losses is critical to the quantitative risk assessment of storm surge flooding.

Storm surge disasters frequently cause significant economic losses and casualties in coastal areas of China (Fang et al. 2014, 2017; Hamilton et al. 2019). From 1998 to 2019, a total of 382 storm surges occurred along the Chinese coast, resulting in 181 disasters and 962 fatalities. The direct economic losses (DEL) incurred by these events amounted to 223.814 billion Chinese Yuan (CNY), making up 91.7% of all DEL caused by marine disasters (Ministry of Natural Resources of China 2021). Furthermore, in the context of global climate change, it is expected that the frequency and intensity of storm surge events in coastal areas will further increase in the future (Muis et al. 2018; Wu et al. 2022). While advances in early-warning and forecasting technology of storm surges have helped to mitigate casualties, the increasing number of vulnerable economic entities in coastal regions poses a challenge to sustainable development (Chan et al. 2013; Pastor-Paz et al. 2020; Hauer et al. 2021). Thus, there is an urgent need for accurate and scientific assessments of economic losses resulting from storm surge disasters. These assessments can be divided into two categories: direct economic loss and indirect economic loss (IEL).

DEL caused by storm surges refers to the physical damages and reduction in economic value caused directly by storm surge disasters (Wang et al. 2021a). These losses can be determined by considering the interaction between the geographical coverage of storm surge hazards, the elements exposed to these hazards, and the vulnerability of the exposed elements, based on the three fundamental elements of the risk assessment (Roy et al. 2021; Islam et al. 2023). Over the past few decades, reliable ocean models have become a popular tool in assessing the hazards of storm surges due to their accuracy and ability to simulate real-world processes (Kirkpatrick & Olbert 2020; Shi et al. 2020a; Yin et al. 2021). Researchers have also developed extreme typhoon scenarios to gauge the severity of storm surge disasters, with the Chinese ocean disaster prevention and reduction researchers using past events to create a standard calculation method to predict storm surge flooding depth and extent. This technical directive has been widely implemented in China's coastal areas, leading to a better understanding of storm surge hazards (Shi et al. 2020b; Wang et al. 2021b; Chen et al. 2023). However, compared to the abundant methods and data available for hazard assessment, obtaining detailed spatial distribution data for the components of exposure and vulnerability has proven to be challenging (Romah & Bloetscher 2015; Wang et al. 2021a).

Recently, various refined estimation methods have been employed by researchers to quantify exposure elements using open social perception data. These data sources include points of interest (POI), nighttime light (NTL), and open street map (OSM) data, which are utilized to disaggregate statistical gross domestic product (GDP) data into spatial grid cells for refined economic exposure quantification. For instance, Chen et al. (2021) utilized the random forest (RF) method and POI data to create 1-km gridded maps of three sub-sectors of China's GDP in 2010. Additionally, gridded GDP data have been utilized to assess flood losses. Tang et al. (2021) estimated the economic losses caused by a one-in-a-thousand-year storm flooding in Shanghai using publicly available 1-km gridded GDP data from 2015. However, the currently available gridded GDP data have certain limitations, such as infrequent updates, lower resolutions, and limited spatial coverage, making it challenging to directly apply them for the fine-scale assessment of DEL from storm surges (Li et al. 2023). Therefore, there is a need to update gridded maps of GDP in both temporal and spatial scales based on open data and utilize exposure estimation methods to meet the requirements of DEL assessment.

Additionally, quantitative assessment of vulnerability for exposure elements requires the use of vulnerability curves and essential urban land use and cover (EULUC) data (Yin et al. 2012; Wang et al. 2021a). However, compared to the rich availability of vulnerability curves provided by scholars (de Moel et al. 2012), publicly available EULUC data in China are relatively limited. While remote sensing imagery can effectively classify land cover using spectral and texture information (Van de Voorde et al. 2011; Wellmann et al. 2020), it is challenging to differentiate categories closely related to human societal activities (Gong et al. 2020; Wang et al. 2023), such as commercial, residential, and industrial land, hindering accurate quantitative assessment of vulnerability for exposure elements. In recent years, research on identifying EULUC using open data such as land cover, POI, OSM, and area of interest (AOI) data has gained increasing attention and has been widely applied in various urban studies (Xu et al. 2021; Izzo et al. 2022). However, there is limited research considering EULUC mapping in the quantitative assessment of vulnerability for storm surge disasters, and few studies have combined high-resolution GDP gridded data and EULUC to assess DEL from storm surge disasters.

The economic losses caused by storm surge disasters encompass not only the DEL incurred through the suppression of economic activities such as commodity production but also the IEL within the economic system resulting from supply-demand imbalances after the disaster (Hallegatte et al. 2011; Parent et al. 2023). Although the methods and principles for assessing DEL and IEL differ, the assessment processes are complementary (Zhuang et al. 2024). The evaluation of IEL from disasters typically relies on sector-based DEL and utilizes modeling approaches related to IEL to estimate the indirect impacts of disasters on the economy (MacKenzie et al. 2012). Initially, statistical techniques like case study analysis, scaling factors, and multiple regression analysis were used for IEL assessment (Haas et al. 1977). Subsequently, economic theories were introduced, and the evaluation gradually evolved from the initial econometric model to the input–output (IO) model and the computable general equilibrium (CGE) model (Oosterhaven 1988; Tirasirichai & Enke 2007). To further quantify IEL, the key lies in disaggregating DEL data by industry and integrating relevant IEL assessment approaches. However, due to data availability and methodological limitations, the quantitative assessment of economic losses caused by storm surge disasters in China rarely incorporates the assessment of IEL.

Building upon the concerns above, this study aims to conduct a comprehensive assessment of economic losses caused by storm surge disasters in the urban core area of Zhoushan City, a renowned archipelagic city in China, utilizing multiple open datasets, vulnerability curves, and the IO model. This study builds upon previous research, specifically a published paper (Chen et al. 2023), which defined five different intensity scenarios for typhoon events in the study area and employed the Jelesnianski wind model and finite volume community ocean model (FVCOM) to simulate storm surge flooding for hazard assessment. Subsequently, the RF method was employed to generate a 30-m sub-sector gridded GDP dataset for exposure assessment. This paper aims to further conduct quantitative vulnerability assessment, DEL assessment, and IEL assessment based on the previous research. Firstly, various open datasets (i.e., POI, OSM, land cover, AOI) were utilized to identify and map different types of EULUC, and quantitative vulnerability assessment was conducted by combining storm surge flooding maps with vulnerability curves. Subsequently, the economic loss ratio maps were combined with the gridded GDP dataset to quantitatively assess DEL. Finally, an IEL assessment was carried out based on sector-based DEL and the IO model.

Study area

Zhoushan City, with latitudes ranging from 29°32′N to 31°04′N and longitudes ranging from 121°30′E to 123°25′E, is located in the eastern Zhejiang Province of China (Figure 1(a) and 1(b)). It is one of the first prefecture-level cities in China built in the form of archipelagos, with a total of 2,085 islands, accounting for 47.93% of the total number of islands in Zhejiang Province. Among them, Zhoushan Island stands out as the largest island in the city, covering an area of 503 km2, and ranking as the fourth largest island in China. The densely populated and economically prosperous areas on Zhoushan Island are predominantly situated in low-lying coastal districts with elevations below ten meters. By 2020, the residential population of Zhoushan had surpassed 1,157,817 persons, with the total GDP reaching approximately 170,362 million CNY.
Figure 1

Study area: (a) map of China; (b) map of Zhoushan City; (c) the urban core area of Zhoushan City.

Figure 1

Study area: (a) map of China; (b) map of Zhoushan City; (c) the urban core area of Zhoushan City.

Close modal

Due to its geographical location, Zhoushan City is frequently subjected to the periodic impact of storm surges caused by typhoons. Historical data from 1949 to 2020 indicate that an average of approximately three typhoon events per year occur (Figure S1). The coastal areas of Zhoushan are characterized by numerous tourism zones, residential buildings, large ports, and crude oil transportation bases, making them highly vulnerable to storm surge disasters. This study focuses on the urban core area of Zhoushan City, specifically Lincheng, Qiandao, Donggang, and Shenjiamen Streets in the southeastern part of Zhoushan Island (Figure 1(c)). Lincheng and Qiandao Streets are the most high-intensity construction areas, while Donggang and Shenjiamen Streets are popular tourist destinations and home to the largest marine fishing port and seafood market.

Data sources

This study utilized various datasets, including storm surge flooding scenarios dataset, open access geospatial data (such as POI, AOI, land cover, OSM, and administrative boundaries), and open access socio-economic data (such as gridded GDP data, statistical yearbook, and IO table). The description, features, and sources of the datasets are as follows:

The dataset for storm surge flooding scenarios: This dataset was obtained from a published paper (Chen et al. 2023). The published paper defined typhoon events in the study area based on different scenarios of typical typhoon intensity, represented by minimum central pressure classes of 915, 925, 935, 945, and 955 hPa. The Jelesnianski typhoon model and FVCOM were used to simulate wind, water level, and current. In the numerical model, the coastal defense measures were also considered in detail through the dike–groyne module. With the numerical modeling system, the storm surge flooding in the five scenarios was simulated, including inundation parameters such as extent and depth, and the corresponding flood maps were created.

POI and AOI: The POI and AOI were acquired from the Baidu Map Open Platform (https://lbsyun.baidu.com/) (accessed on 3 January 2022), which is a prominent provider of digital map content, navigation, and location services solutions in China. These datasets proved to be invaluable in assigning relevant information to each plot in the study area, facilitating the identification and classification of different types of EULUC.

Land cover: The land cover data were sourced from the European Space Agency (ESA) World Cover dataset (https://esa-worldcover.org/en), which offers cutting-edge global land cover products at a 10-meter resolution for the year 2020 (Zanaga et al. 2022). These data were used to provide natural categories information of the EULUC.

OSM: The OSM dataset was sourced from the esteemed OSM platform (https://www.openstreetmap.org/). It provided detailed information about the road infrastructure, including road types, connectivity, and transportation networks.

Administrative boundaries: The administrative boundary data at the township/street level and village/community level in the study area were obtained from the Zhoushan Natural Resources and Planning Bureau. The study area comprised 4 street levels and 55 villages/communities.

Gridded GDP data: The 30-m sub-sector gridded GDP data for the year 2020 was obtained from a published paper (Chen et al. 2023). The generation process employed a combination of the RF model and the dasymetric mapping method, ensuring a robust and reliable estimation of GDP at a highly detailed level.

Statistical yearbook: The statistical yearbook was sourced from the People's Government of Zhoushan Municipality (https://www.zhoushan.gov.cn/). The yearbook was used to determine the proportion of different industries.

IO table: The IO table was sourced from the Zhejiang Provincial Bureau of Statistics (https://tjj.zj.gov.cn/). This dataset provided a comprehensive overview of the interdependencies and linkages between various sectors within the economy. It helped in analyzing the economic ripple effects triggered by disruptions caused by storm surge disasters.

The study presents a multidisciplinary framework for conducting a comprehensive assessment of economic losses caused by storm surge disasters using open data. The framework comprises three main components: quantitative assessment of vulnerability, DEL assessment, and IEL assessment. The process is shown in Figure 2.
Figure 2

Flow chart of comprehensive quantitative assessment for economic losses of storm surge disasters.

Figure 2

Flow chart of comprehensive quantitative assessment for economic losses of storm surge disasters.

Close modal

Quantitative vulnerability assessment

The quantification of storm surge economic loss vulnerability primarily relies on the storm surge flood scenario simulation, the EULUC classification method, and vulnerability curve analysis. The result of quantitative vulnerability assessment is expressed in the form of loss ratio (Yin et al. 2012; Tang et al. 2021; Shi et al. 2024). Storm surge flood scenario simulations provide information on the extent of inundation and flood depths during storm surge events in the study area. The EULUC classification was based on existing literature and previous research (Gong et al. 2020; Wang et al. 2023), which utilized POI density analysis, OSM-based space units, and land cover data to categorize the EULUC of study area. Furthermore, to improve the spatial accuracy of existing OSM and POI-based EULUC sources, a new open dataset (i.e., AOI) was introduced. The process involves several steps. Firstly, the POI and AOI were reclassified into five types: residential (RE), administrative and public facilities (AP), commercial and business facilities (CB), industrial (IN), streets, and transportation (ST). Then, space units were delineated using OSM, AOI, and administrative boundaries. Subsequently, the POI density of each space unit was calculated, and the space units were classified based on the concentration and distribution patterns of relevant POI categories. Finally, the EULUC was generated by combining the space units, OSM, land cover, and AOI. More details about the methodology can be found in the supporting information.

In the quantitative assessment of storm surge vulnerability, vulnerability curves play a fundamental role as data that represent the correlation between storm surge intensity and the extent of damage (Yin et al. 2012; Shi et al. 2024). The intensity in vulnerability curves can be expressed as inundation depth, water flow velocity, flow direction, or duration. In most cases, inundation depth is considered the most common intensity indicator for calculating the direct tangible losses of a set of affected elements. In this study, vulnerability curves (depth-loss ratio curves) for different EULUC types (Table S2) were sourced from the report titled ‘Evaluation of Flood Control and Disaster Reduction Capacity of Water Conservancy Projects in Zhejiang Province’ published by the Zhejiang Institute of Hydraulics and Estuary (2007). These curves depicted the loss ratios at different inundation depths and were calculated for agricultural land, residential land, industrial land, transportation land, public land, and other land categories. In addition, due to the lack of vulnerability curve data for mixed-use areas, this study employed the average vulnerability values of the two EULUC types as the vulnerability values for mixed-use areas.

DEL assessment

The DEL assessment method enables the comprehensive consideration of the storm surge hazards and their adverse consequences, quantifying the losses in monetary terms (Yin et al. 2011; Tang et al. 2021). These consequences arise from the interaction between the geographic extent of the hazards, the degree of element exposure to the hazards, and the vulnerability of the exposed elements. In this study, high-resolution economic distribution data in the form of a 30 m resolution sub-sectors gridded GDP dataset were utilized. The gridded GDP data represent the economic value of each grid, allowing for a more detailed and spatially explicit analysis of storm surge-related economic losses (Chen et al. 2023).

Regarding different industries, this study established distinct statistical rules to determine the impact of storm surge flooding on GDP. For the primary sector GDP (GDP1), it was assumed that a single storm surge flood event would have a sustained effect on the annual GDP. This is because significant flood events can lead to irreversible losses such as farmland inundation, reduced crop yields, forest destruction, and depletion of fish populations. These losses are not easily repaired or recovered in subsequent stages. On the other hand, for the secondary sector GDP (GDP2) and tertiary sector GDP (GDP3), flood disasters would not have a sustained impact on the annual GDP. Although work and production may experience temporary interruptions, GDP2 and GDP3 exhibit a certain degree of resilience. Therefore, calculating the DEL for GDP2 and GDP3 using annual GDP would result in overestimated losses, which do not align with the actual situation. To address this issue, this study employed the daily interruption method to estimate the DEL for GDP2 and GDP3. This approach considers the temporary disruptions in work and production caused by storm surge flooding, allowing for a more realistic assessment. The calculation formulas are shown in the following equations.
(1)
(2)
(3)
(4)
where is the DEL of the primary industry; is the DEL of the secondary industry; is the DEL of the tertiary industry; is the total DEL; is the value of the primary industry in the i-th grid; is the value of the secondary industry in the i-th grid; is the value of the tertiary industry in the i-th grid; R(i,h) is the vulnerability value (loss ratio) of the i-th grid at the height of h water depth, %; Di is the number of inundation days of the i-th grid affected by the storm surge disaster.

Subsequently, the DEL was further divided into nine industries using the GDP classification method from the ‘Zhoushan Statistical Yearbook’ (Su et al. 2022). These industries include agriculture, forestry, animal husbandry, and fishery (INAG), industry (ININ), construction (INCO), wholesale and retail trade (INWR), transportation, warehousing, and postal services (INTR), accommodation and catering (INAC), finance (INFI), real estate (INRE), and other services (INOS). According to this classification, the DEL for the agricultural sector (INAG) corresponds to the DEL of GDP1, while the DEL for ININ, and INCO corresponds to the DEL of GDP2. The DEL for INWR, INTR, INAC, INFI, INRE, and INOS corresponds to the DEL of GDP3. This allocation method enables a relatively accurate breakdown of the DEL into the nine industries.

IEL assessment

The IEL assessment utilized an IO model. The IO model is an economic mathematical model that comprehensively analyzes the interdependence between input and output quantities in economic activities (Oosterhaven 1988; Parent et al. 2023). It consists of two components: the IO table and a system of mathematical equations established based on the balanced relationship derived from the IO table. The calculation process involves several steps. Firstly, the IO table's 42 industries were consolidated into 9 relevant industries based on the industry classification provided by the ‘Zhoushan Statistical Yearbook’. Then, the direct consumption coefficient matrix (DCCM) was computed, representing the input proportions consumed by each sector to produce one output unit. Subsequently, using the DCCM, the complete consumption coefficient matrix (CCCM) was derived, capturing the direct and indirect consumption relationships between sectors and their chain reactions. Next, the IEL coefficients were calculated based on the CCCM, considering both demand-side equilibrium and supply-side equilibrium. These coefficients represent the inter-industry linkages that result in IEL for an industry due to disruptions in another industry. The DEL was then multiplied by the corresponding IEL coefficients to quantify the IEL for each industry. Finally, the DEL and the corresponding IEL were summed to obtain the total economic loss (TEL), providing an overall estimation of the economic impact. The calculation formula is shown in Equation (5). More details about the methodology can be found in the supporting information.
(5)
where TEL is the total economic loss; is the IEL on the demand side; is the IEL on the supply side.

Quantitative vulnerability assessment

Figure 3 displays the EULUC map, which was compiled from various open data sources. The study area has been categorized into nine functional types, including five single-functional types and four mixed-functional types. Furthermore, there are seven land types that correspond to natural features within the area.
Figure 3

Essential urban land use/cover categories division map in the urban core area of Zhoushan City (residential land (RE); administration and public services (AP); commercial and business facilities (CB); industrial land (IN); street and transportation (ST)).

Figure 3

Essential urban land use/cover categories division map in the urban core area of Zhoushan City (residential land (RE); administration and public services (AP); commercial and business facilities (CB); industrial land (IN); street and transportation (ST)).

Close modal

According to Figure 3, the EULUC in coastal low-lying areas consists mainly of residential, commercial, public service, and industrial areas. Lincheng Street and Qiandao Street have a higher concentration of EULUC compared to Donggang Street and Shenjiamen Street. Residential lands are concentrated in the middle section of Qiandao Street, the eastern section of Lincheng Street, Donggang Street, and the area behind the embankment in Shenjiamen Street. Commercial and business facilities are integrated with residential and administrative/public areas. The southwestern parts of Lincheng Street, the southwestern parts of Donggang Street, and the western parts of Shenjiamen Street are characterized by integrated industrial parks with active port transportation, oil and gas industries, and other industrial activities due to their advantageous geographical locations. Administrative and public services intertwine with residential and commercial areas. Mixed-functional areas primarily combine commercial and residential functions in terms of area and percentage and are mainly located in the city center. Mixed-functional areas with public service functions are typically found near schools and government buildings. Industrial and residential mixed-use areas are located near low-cost industrial zones, providing convenience for commuting workers.

The study area encompasses not only land types associated with human activities, but also seven land types that are linked to natural features, namely forests, grasslands, bare land, water bodies, farmland, shrubs, and wetlands. Forests and grasslands are primarily found in the highland areas that are located far from the coastline and urban regions. Bare land represents recently reclaimed areas and is predominantly situated on the right side of the embankment behind Donggang Street. Farmland is mainly concentrated in the higher elevation areas on the western side of Donggang Street. Water bodies consist of reservoirs situated in high-altitude regions, as well as small rivers, canals, and lakes in urban zones. The study area has relatively limited shrublands and wetlands.

By utilizing storm surge flooding maps for varying typhoon intensities, in conjunction with the spatial distribution of EULUC types, the loss ratio maps were computed through vulnerability curves, as depicted in Figure 4. The split of loss ratios in the legend of Figure 4 was based on previous research (Shi et al. 2024). Generally speaking, the loss ratios remain modest at lower typhoon intensities. However, as the typhoon intensity escalates, the scope of flooding expands, leading to a wider distribution range and increased values of the loss ratios.
Figure 4

Loss ratio distribution maps for the designed typhoon scenarios: (a) 915 hPa, (b) 925 hPa, (c) 935 hPa, (d) 945 hPa, and (e) 955 hPa (unit: %).

Figure 4

Loss ratio distribution maps for the designed typhoon scenarios: (a) 915 hPa, (b) 925 hPa, (c) 935 hPa, (d) 945 hPa, and (e) 955 hPa (unit: %).

Close modal

Based on Figure 4, it can be observed that at a moderate typhoon intensity of 955 hPa, the loss ratios in the study area are concentrated near the coastline and in areas with low flooding in Donggang Street where flood protection embankments are lacking. The distribution of high-value loss ratios is below 10%. However, at a strong typhoon intensity of 945 hPa, the high-value loss ratios of approximately 45% are mainly concentrated on the left and central areas behind the embankment in Donggang Street. These areas have lower levels of embankment construction and low-lying terrain, leading to increased flood depths due to storm surges. At a strong typhoon intensity of 935 hPa, the range of loss ratios expands to the entire study area, with high-value areas of approximately 55% at the intersection of Lincheng Street and Qiandao Street, as well as the left and central areas behind the embankment in Donggang Street. High-loss ratios are also observed on the left side of the intersection between Donggang Street and Shenjiamen Street and behind the embankment in Shenjiamen Street, reaching around 45%.

At a super typhoon intensity of 925 hPa, the range of loss ratios and high-value areas expands correspondingly, spreading to the farmland on the left side of Donggang Street and the industrial park on the left side of Lincheng Street. Notably, the highest-loss ratio of 100% is observed behind the embankment in the left and central areas of Donggang Street. At a super typhoon intensity of 915 hPa, the range of high-value loss ratio areas further expands. High-loss ratio areas are mainly concentrated in low-lying areas of Qiandao Street, with loss ratios reaching approximately 64% in the central business center area between Lincheng Street and Qiandao Street. A large area behind the embankment in Qiandao Street also experiences a loss ratio of 100%.

DEL assessment

Using the economic loss ratios resulting from various levels of typhoon intensity and the spatial distribution of GDP in the study area, distribution maps of DEL were generated for different typhoon scenarios, as depicted in Figure 5. In general, the DEL is relatively modest at lower typhoon intensities. However, as the intensity of the typhoon increases, the extent of flooding expands, resulting in a wider distribution range and higher values of DEL.
Figure 5

Direct economic loss distribution maps for the designed typhoon scenarios: (a) 915 hPa, (b) 925 hPa, (c) 935 hPa, (d) 945 hPa, and (e) 955 hPa (unit: thousand CNY/900 m2).

Figure 5

Direct economic loss distribution maps for the designed typhoon scenarios: (a) 915 hPa, (b) 925 hPa, (c) 935 hPa, (d) 945 hPa, and (e) 955 hPa (unit: thousand CNY/900 m2).

Close modal

These maps offer detailed and informative insights into the distribution patterns of DEL under different scenarios. The split of DEL in the legend of Figure 5 was based on the natural breakpoint classification method from Geographic Information System (GIS) (Zhang et al. 2021; Celedón et al. 2023). According to Figure 5, it can be observed that at a moderate typhoon intensity of 955 hPa, there are only a few DEL concentrated near the coastline or low-hazard areas of Donggang Street. However, at a strong typhoon intensity of 945 hPa, the DEL primarily concentrates in the left and central areas behind the Donggang Street embankment. This includes high-value DEL of approximately 2.11 thousand CNY/900 m². Due to the lower level of embankment construction and low-lying terrain, storm surges result in flooding, which overlaps with the commercial and residential centers of Donggang Street, leading to greater economic losses. At a strong typhoon intensity of 935 hPa, the distribution range of DEL expands to the entire study area. The high-value areas include the aforementioned areas, the commercial center area at the intersection of Lincheng Street and Qiandao Street, and the left side of Shenjiamen Street. The highest losses in the left and central areas behind the embankment of Donggang Street can exceed 4.51 thousand CNY/900 m². At a super typhoon intensity of 925 hPa, the range of DEL widens, particularly toward the agricultural fields on the left side of Donggang Street and the industrial park area on the left side of Lincheng Street. Although the scope of DEL expands, the range of high DEL does not increase. At a super typhoon intensity of 915 hPa, the range of high-value DEL further expands. The new high-value areas mainly concentrate in the low-lying areas of Qiandao Street and the commercial center area between Lincheng Street and Qiandao Street. Additionally, there is an extensive high DEL in the left and central areas behind the Donggang Street embankment.

Figure 6 presents the statistical analysis of DEL across various typhoon scenarios. At a moderate typhoon intensity of 955 hPa, the DEL amounts to approximately 0.61 million CNY. However, during a super typhoon intensity of 915 hPa, the DEL significantly increases to around 49 million CNY, which is approximately 80 times higher than the moderate typhoon intensity. The DEL in the nine industries was obtained and classified based on GDP data from the statistical yearbook (Table S5). Among these categories, INAG stands out with the highest DEL, averaging 5.703 million CNY across five typhoon scenarios. This disparity can be attributed to variations in statistical standards. For instance, DEL for the primary industry is calculated based on annual GDP, while DEL for the secondary and tertiary industries is calculated based on daily average GDP. Within the secondary and tertiary industries, ININ exhibits the highest DEL, with an average of 4.878 million CNY across five typhoon scenarios, while INAC has the lowest average DEL of 0.296 million CNY.
Figure 6

The histogram statistic presents the direct economic losses for the designed storm surge-related flood scenarios.

Figure 6

The histogram statistic presents the direct economic losses for the designed storm surge-related flood scenarios.

Close modal

IEL assessment

To calculate the IEL caused by inter-industry linkages in different sectors, the relevant sectors in the IO table were merged based on the GDP classification in the statistical yearbook of Zhoushan City. The merged IO table was used to compute the DCCM and the CCCM (Figure S6). The IEL coefficients between different industries were then calculated separately from the demand side and the supply side using the CCCM (Figure S7). By utilizing the IEL coefficients and the DEL data from five typhoon scenarios, the IEL values for different scenarios were obtained (Table S7). According to the statistical results presented in Table S9, the industries with the highest average TEL are ININ (16.390 million CNY), INAG (13.403 million CNY), and INOS (13.273 million CNY), accounting for 26, 21, and 21% of the total IEL, respectively. Figure 7 presents a statistical analysis of both DEL and IEL. At a moderate typhoon intensity of 955 hPa, the TEL amounted to approximately 1.6 million CNY. However, at a super typhoon intensity of 915 hPa, the TEL significantly increased to around 131 million CNY, which is 82 times higher than the moderate typhoon intensity. Additionally, the proportion of DEL, IELDemand, and IELSupply is approximately 4:3:3, with IEL accounting for about 60%. These findings highlight the importance of calculating IEL in assessing economic losses.
Figure 7

The histogram statistic presents the total economic losses for the designed typhoon scenarios.

Figure 7

The histogram statistic presents the total economic losses for the designed typhoon scenarios.

Close modal

According to the histogram statistics presented in Figure S8, the proportions of DEL and IEL in TEL were determined across various industries. The results show that ININ has the smallest proportion of IELDemand, accounting for only 5.7%, while INCO has the highest proportion of IELDemand at 68%. The remaining industries have IELDemand proportions ranging from 29 to 55%. Additionally, ININ boasts the highest proportion of IELSupply at 65%, while ININ has the smallest proportion of IELSupply at 2.1%. The proportions of IELSupply in other industries range from 7 to 29%. The cumulative IEL across all industries surpasses DEL, especially in ININ and INCO sectors, which rely heavily on upstream and downstream industries, accounting for up to 70%. In other industries, IEL ranges from 55 to 64%.

By analyzing the CCCM of nine industries through a heat map (Figure S6) and considering the IEL coefficients (Figure S7), effective disaster risk reduction measures can be developed to mitigate the damages caused by storm surge hazards. Among the industries, INCO has the highest IELDemand coefficient, which has a significantly greater impact on other industries than its own consumption. Therefore, INCO should prioritize implementing measures to prevent losses from spreading to other sectors. This may include measures to enhance supply chain resilience, promote business continuity planning, and facilitate coordination and communication among industries to minimize chain reactions. On the other hand, ININ has the highest IELSupply coefficient and multiple industries consume ININ's products in their production processes, resulting in the largest proportion of IELSupply. As a result, ININ should primarily focus on reducing losses within its own industry. This could involve implementing measures such as improving flood control infrastructure, and early-warning systems, and developing emergency response plans specifically tailored to ININ.

Storm surge risk zoning map

The storm surge risk zoning map is a valuable tool for decision-makers to determine which areas are most vulnerable to severe storm surge impacts (Wang et al. 2021a). This allows them to create effective evacuation plans and reduce economic losses. By calculating the average land-based TEL for each village/community at varying typhoon intensities, the risks were categorized into five levels: high, relatively high, moderate, relatively low, and low. To ensure the accuracy of the results, the natural breakpoints classification method was utilized to avoid any extreme values affecting the classification process (Zhang et al. 2021; Celedón et al. 2023). More details about the breakpoint ranges can be found in the supporting information.

A visual overview of the risk levels across the entire study area is presented in Figure 8. The chart identifies the villages that require immediate attention and improvements in flood control measures. The zoning results for economic loss levels in the study area encompassing 55 villages/communities are as follows: At a moderate typhoon intensity of 945 hPa, except Pingyang and Siwan Villages, all other coastal villages are classified as low loss levels. At a strong typhoon intensity of 945 hPa, Yongxing and Chenjiahou Villages, located behind the Donggang embankment, are initially classified as relatively low-loss villages. Luxiasha Village is categorized as moderate loss level. At a strong typhoon intensity of 935 hPa, Waigoushan Village is identified as the highest-loss level village. Xitang, Wansan, Pudong, Luxi, and Zhongnong Villages are classified as high-loss level villages. At a super typhoon intensity of 925 hPa, Shengshan, Hengtang, Shuangyang, and Lusha Villages are classified as relatively high-loss level villages. At a super typhoon intensity of 915 hPa, Xitang, Shengshan, Wansan, and Luxiasha Villages have higher loss levels. The findings of the study provide valuable insights to enhance flood control measures in the region.
Figure 8

Economic loss risk level zoning maps at the village/community scale for the designed typhoon scenarios: (a) 915 hPa, (b) 925 hPa, (c) 935 hPa, (d) 945 hPa, and (e) 955 hPa.

Figure 8

Economic loss risk level zoning maps at the village/community scale for the designed typhoon scenarios: (a) 915 hPa, (b) 925 hPa, (c) 935 hPa, (d) 945 hPa, and (e) 955 hPa.

Close modal

Given the findings, officials should give precedence to enforcing high embankment standards in Luxiasha, Yongxing, and Chenjiahou Villages situated behind the Donggang Street embankment. Moreover, closely monitoring high-loss communities such as Waigoushan, Xitang, Shengshan, Wansan, and Lusha is encouraged. In addition, it is wise to avoid placing valuable assets in regions susceptible to flooding caused by storm surges during future urban development initiatives.

Research implications

This study proposes a framework for a comprehensive economic losses assessment of storm surge disasters using open data. The focus is on quantifying the economic loss ratios, DEL, and IEL caused by storm surge disasters. This framework can be widely applied to other coastal areas.

Previous storm surge disaster vulnerability studies in China's disaster risk reduction field have recommended the risk matrix method for qualitative risk assessment of storm surges, due to limited exposure and vulnerability data. This method has been adopted by many coastal cities in China (Shi et al. 2020a; Wang et al. 2021b, 2021c), but it has drawbacks, such as subjective vulnerability definitions and the inability to quantify damages and risks in monetary terms (Wang et al. 2021a). Additionally, previous studies also relied on land cover data alone, making it difficult to assess specific vulnerabilities of different impermeable surfaces in urban areas (Su et al. 2022). This study uses multiple open data sources, including land cover, POI, AOI, and OSM, along with the EULUC classification and vulnerability curve methods, to create a data foundation for quantitative vulnerability assessment of storm surge disasters.

Previous studies on quantitative economic loss assessment have encountered challenges due to the lack of detailed socio-economic distribution data. This has resulted in the use of statistical data on GDP or publicly available gridded GDP data for DEL assessment (Tang et al. 2021). However, such data sources have limitations, such as lacking detailed spatial distribution or infrequent updates, low accuracy, and limited spatial coverage (Li et al. 2023). Furthermore, there has been limited research on incorporating IEL assessment in quantitative economic loss assessment of storm surge disasters (Wang et al. 2021a), and few studies have provided recommendations on reducing such losses from an industry perspective. In the related research of this study (Chen et al. 2023), a combination of multiple open data sources, including POI, OSM, and NTL data, was employed. Utilizing the RF model, three sub-sectors of gridded GDP data with a spatial resolution of 30-m were derived. These data, along with vulnerability assessment results and IO models, facilitated the quantitative evaluation of both direct and IEL stemming from storm surge disasters. Furthermore, industry perspectives and loss zoning were taken into account to provide recommendations for reducing storm surge disaster losses. Notably, the use of open data in this study offers advantages such as free accessibility, transparency, frequent updates, and low data collection costs (Gouett-Hanna et al. 2022). This approach enables rapid acquisition of the necessary data for vulnerability and quantitative loss assessment. Additionally, it allows for timely updates of vulnerability and quantitative loss maps, which help identify areas with high vulnerability and potentially significant losses within urban settings.

To better compare the three components in quantitative DEL assessment, the relationship between them is analyzed using an example of a 915 hPa typhoon storm surge scenario, as shown in Figure 9. A high level of storm surge hazard is observed in inundated areas like Zone A and Zone B, but their economic exposure and loss ratios are relatively low, resulting in lower DEL. This indicates that high-hazard areas may not necessarily correspond to regions with high levels of loss. On the other hand, some inundated areas like Zone C have moderate or lower hazard levels, resulting in lower loss ratios, but their economic exposure is relatively high, leading to higher DEL. In contrast, areas like Zone D are classified as high-hazard areas for storm surges, but their economic exposure is relatively low, resulting in moderate DEL despite having higher loss ratios. Therefore, reducing quantitative economic losses should not focus solely on a single component but should encompass a comprehensive consideration of hazard, exposure, and vulnerability to minimize losses from storm surge disasters. These findings highlight the importance of detailed hazard assessments, exposure evaluations, and quantitative vulnerability assessments for precise quantitative economic loss evaluations.
Figure 9

The assessment results for the super typhoon intensity of 915 hPa scenario, including (a) hazard assessment, (b) exposure assessment, (c) vulnerability assessment, and (d) direct economic loss assessment.

Figure 9

The assessment results for the super typhoon intensity of 915 hPa scenario, including (a) hazard assessment, (b) exposure assessment, (c) vulnerability assessment, and (d) direct economic loss assessment.

Close modal

Limitations and future works

There are several areas where this study could be improved. Firstly, the study only considered inundation depth as the primary factor for assessing loss ratios, even though flow velocity and direction are also important parameters in the assessment of storm surge hazards (de Moel et al. 2012; Wang & Marsooli 2021). Future research should take these flooding hazard factors into account. Secondly, while the use of land use data, OSM, POI, and AOI provides valuable information for distinguishing different types of EULUC, they are insufficient for differentiating the structure and condition of buildings within urban areas (Hadi et al. 2021; Shi et al. 2024). To achieve a more accurate and comprehensive quantitative assessment of vulnerability, it is necessary to integrate street view image data to identify and classify the structure and condition of buildings.

Moreover, vulnerability can be divided into two interconnected elements: physical vulnerability and social vulnerability, as noted by Nahar et al. (2023). However, this study solely focused on physical vulnerability. Social vulnerability is typically evaluated through socio-economic analyses of communities, taking into account factors such as age, gender, and education, which determine the community's susceptibility and ability to withstand destructive impacts, as mentioned by Nguyen et al. (2019). In the future, incorporating multi-criteria decision-making techniques and data sources such as statistical yearbooks and geospatial big data would enable the integration of social vulnerability into comprehensive storm surge risk assessments.

Several datasets used in this study can be upgraded in future research. In the sectoral classification utilized in this study, the categorization of DEL solely relied on the proportion of GDP derived from statistical yearbooks. A more detailed breakdown of these losses by industry would provide a more comprehensive understanding of trade relationships between sectors (Zhuang et al. 2024). As such, future research will incorporate an analysis of the relationship between EULUC and industries, utilizing this data to allocate DEL to more specific industries. This will enable a more nuanced assessment of IEL. Additionally, it is important to note that extreme disasters can have far-reaching impacts on the economy, particularly in densely populated cities with complex transportation and sectoral systems. With increasing interconnectedness between industries, the effects of losses can spread beyond local areas and impact related industries in other regions, resulting in regional spillover losses (Ding & Wu 2023). Moving forward, multi-regional IO tables will be utilized to calculate these losses more accurately.

With the ongoing global climate change, it is anticipated that coastal regions will experience a rise in the frequency and intensity of storm surge events in the future. However, as coastal areas continue to experience rapid economic and social development, the proportion of economic entities vulnerable to storm surge disasters is also growing. Despite the wealth of simulated methods and information available on storm surge hazards, obtaining detailed spatial distribution data for quantitative exposure and vulnerability remains a challenge. This results in a lack of comprehensive assessments of the economic losses caused by storm surge disasters. Furthermore, few studies have incorporated the evaluation of IEL into the quantitative assessment of storm surge economic losses. To address this gap, this study introduces a multidisciplinary framework that utilizes open data to comprehensively assess the economic losses caused by storm surge disasters. The focus is on quantifying the economic loss ratios of storm surge flooding, as well as the DEL and IEL.

Based on the spatially explicit analysis of the storm surge-related economic losses, it has been found that embankments demonstrate effectiveness in reducing the risk of economic losses during moderate typhoon intensity (955 hPa). However, they are less effective during strong and super typhoon intensities (915–945 hPa). The eastern part of Donggang Street has been identified as a high and extremely high occurrence area for 945 hPa typhoons. Additionally, scenarios involving strong or super typhoon intensities (915–935 hPa) result in significant economic losses throughout the coastal area. To address this issue, it is recommended that local governments prioritize strengthening the coastal flood control capabilities of Luxiasha, Yongxing, and Chenjiahou Villages. These areas are located behind the Donggang embankment and are therefore vulnerable to flooding. Furthermore, attention should be given to high-risk villages such as Waigoushan, Xitang, Shengshan, Wansan, and Luxiasha Villages, with a focus on enhancing their resilience against coastal flooding. The quantitative loss zoning map generated for the study area can assist decision-makers in identifying areas that are more likely to be severely affected by storm surges. This information enables the development of evacuation strategies aimed at minimizing economic losses.

According to the quantitative assessment results, the TEL is approximately 1.6 million CNY during moderate typhoon intensity (955 hPa). However, during super typhoon intensity (915 hPa), the TEL significantly increases to around 131 million CNY, which is 82 times higher than that of moderate typhoon intensity. Additionally, the proportion of IEL exceeds that of DEL in all industries, with ININ and INCO industries being particularly affected, as they rely heavily on upstream and downstream industries, accounting for up to 70% of the losses. Therefore, to reduce storm surge disaster losses, it is important to develop different disaster prevention and mitigation measures for different industries. For example, the industrial sector can focus on reducing losses within the sector itself, while the construction sector, which has a significant impact on other industries, can prevent losses from further expanding into other sectors. This study enriches the assessment methods for storm surge disaster losses and provides technical support for disaster prevention and mitigation decision-making, as well as post-disaster industrial structural adjustments.

We would like to thank the comments and suggestions by anonymous reviewers for improving the original manuscript.

This work was supported by the National Key Research and Development Program of China (Grant No. 2021YFE0206200), the Science Foundation of Donghai Laboratory (Grant No. DH-2022KF01012), and the HPC Center of Zhejiang University (Zhoushan Campus).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.

Celedón
V.
,
Del Río
L.
,
Ferreira
Ó.
,
Costas
S.
&
Plomaritis
T. A.
2023
Identification of risk hotspots to storm events in a coastal region with high morphodynamic alongshore variability
.
Natural Hazards
115
(
1
),
461
488
.
Chan
F. K. S.
,
Adekola
O. A.
,
Mitchell
G.
&
McDonald
A. T.
2013
Appraising sustainable flood risk management in the Pearl River Delta's coastal megacities: A case study of Hong Kong, China
.
Journal of Water and Climate Change
4
(
4
),
390
409
.
Chen
Q.
,
Ye
T.
,
Zhao
N.
,
Ding
M.
,
Ouyang
Z.
,
Jia
P.
,
Yue
W.
&
Yang
X.
2021
Mapping China's regional economic activity by integrating points-of-interest and remote sensing data with random forest
.
Environment and Planning B: Urban Analytics and City Science
48
(
7
),
1876
1894
.
Chen
B.
,
He
J.
,
He
Z.
,
Li
L.
,
Chen
Q.
,
Li
F.
,
Chu
D.
,
Cao
Z.
&
Yang
X.
2023
Potential impacts of storm surge-induced flooding based on refined exposure estimation: A case study in Zhoushan island, China
.
Geomatics, Natural Hazards and Risk
14
(
1
),
2232080
.
de Moel
H.
,
Asselman
N. E. M.
&
Aerts
J. C. J. H.
2012
Uncertainty and sensitivity analysis of coastal flood damage estimates in the west of The Netherlands
.
Natural Hazards and Earth System Sciences
12
(
4
),
1045
1058
.
Fang
J.
,
Sun
S.
,
Shi
P.
&
Wang
J. A.
2014
Assessment and mapping of potential storm surge impacts on global population and economy
.
International Journal of Disaster Risk Science
5
,
323
331
.
Fang
J.
,
Liu
W.
,
Yang
S.
,
Brown
S.
,
Nicholls
R. J.
,
Hinkel
J.
,
Shi
X.
&
Shi
P.
2017
Spatial-temporal changes of coastal and marine disasters risks and impacts in Mainland China
.
Ocean & Coastal Management
139
,
125
140
.
Gong
P.
,
Chen
B.
,
Li
X.
,
Liu
H.
,
Wang
J.
,
Bai
Y.
,
Chen
J.
,
Chen
X.
,
Fang
L.
&
Feng
S.
2020
Mapping essential urban land use categories in China (EULUC-China): Preliminary results for 2018
.
Science Bulletin
65
(
3
),
182
187
.
Gouett-Hanna
C.
,
Oulahen
G.
,
Henstra
D.
&
Thistlethwaite
J.
2022
Flood risk assessment data access and equity in Metro Vancouver
.
Canadian Water Resources Journal/Revue canadienne des ressources hydriques
47
(
4
),
202
215
.
Haas
J. E.
,
Kates
R. W.
&
Bowden
M. J.
1977
Reconstruction following disaster
.
Reconstruction Following disaster
331
331
.
Hadi
T.
,
Islam
M. S.
,
Richter
D.
&
Fakhruddin
B.
2021
Seeking Shelter: The factors that influence refuge since cyclone gorky in the coastal area of Bangladesh
.
Progress in Disaster Science
11
,
100179
.
Hallegatte
S.
,
Ranger
N.
,
Mestre
O.
,
Dumas
P.
,
Corfee-Morlot
J.
,
Herweijer
C.
&
Wood
R. M.
2011
Assessing climate change impacts, sea level rise and storm surge risk in port cities: A case study on Copenhagen
.
Climatic Change
104
(
1
),
113
137
.
Hamilton
D.
,
Zhang
H.
&
Eccles
R.
2019
A review of the effects of climate change on riverine flooding in subtropical and tropical regions
.
Journal of Water and Climate Change
10
(
4
),
687
707
.
Hauer
M. E.
,
Hardy
D.
,
Kulp
S. A.
,
Mueller
V.
,
Wrathall
D. J.
&
Clark
P. U.
2021
Assessing population exposure to coastal flooding due to sea level rise
.
Nature Communications
12
(
1
),
6900
.
Islam
M. S.
,
Islam
A. F.
,
Bahar
S.
,
Rahman
S. M. A.
,
Sharif
N. H.
,
Hadi
T.
,
2023
A risk based approach of assessing cyclone shelter locational needs in the coastal region of Bangladesh
. In:
Proceedings of the 2nd International Symposium on Disaster Resilience and Sustainable Development
(
Pal
I.
,
Shaw
R.
,
Ichinose
T.
,
Yonariza
T.
&
Oda
, eds).
Springer Nature Singapore
,
Singapore
, pp.
105
131
.
Izzo
S.
,
Prezioso
E.
,
Giampaolo
F.
,
Mele
V.
,
Di Somma
V.
&
Mei
G.
2022
Classification of urban functional zones through deep learning
.
Neural Computing and Applications
34
(
9
),
6973
6990
.
Kirkpatrick
J. I. M.
&
Olbert
A. I.
2020
Modelling the effects of climate change on urban coastal-fluvial flooding
.
Journal of Water and Climate Change
11
(
S1
),
270
288
.
MacKenzie
C. A.
,
Santos
J. R.
&
Barker
K.
2012
Measuring changes in international production from a disruption: Case study of the Japanese earthquake and tsunami
.
International Journal of Production Economics
138
(
2
),
293
302
.
Ministry of Natural Resources of China
2021
Chinese Marine Disaster Bulletin 2020
.
Beijing
.
Muis
S.
,
Haigh
I. D.
,
Guimarães Nobre
G.
,
Aerts
J. C. J. H.
&
Ward
P. J.
2018
Influence of El Niño-Southern Oscillation on global coastal flooding
.
Earth's Future
6
(
9
),
1311
1322
.
Nahar
N.
,
Islam
S.
,
Sarwar
G. M.
,
Perveen
F.
,
Bhuiyan
S. H.
,
2023
Development of a disaster and climate risk atlas in Bangladesh: Methodology for quantification of risk. Chapter 3
. In:
Multi-hazard Vulnerability and Resilience Building: Cross Cutting Issues
(
Pal
I.
&
Shaw
R.
, eds).
Elsevier
, pp.
27
44
.
Oosterhaven
J.
1988
On the plausibility of the supply-driven input-output model
.
Journal of Regional Science
28
(
2
),
203
217
.
Parent
O.
,
Hofe
R. v.
,
Oh
S.
,
Ervin
P.
,
Khalid
M. A.
&
Yeghiazarian
L.
2023
Estimating the socioeconomic impacts of flooding on regional economies with a computable general equilibrium model
.
Water Resources Research
59
(
9
),
e2022WR033474
.
Pastor-Paz
J.
,
Noy
I.
,
Sin
I.
,
Sood
A.
,
Fleming-Munoz
D.
&
Owen
S.
2020
Projecting the effect of climate change on residential property damages caused by extreme weather events
.
Journal of Environmental Management
276
,
111012
.
Romah
T.
&
Bloetscher
F.
2015
Tools for assessing sea level rise vulnerability
.
Journal of Water and Climate Change
6
(
2
),
181
190
.
Roy
B.
,
Khan
M. S. M.
,
Saiful Islam
A. K. M.
,
Khan
M. J. U.
&
Mohammed
K.
2021
Integrated flood risk assessment of the Arial Khan River under changing climate using IPCC AR5 risk framework
.
Journal of Water and Climate Change
12
(
7
),
3421
3447
.
Shi
X.
,
Jufei
Q.
,
Bingrui
C.
,
Xiaojie
Z.
,
Haoshuang
G.
,
Jun
W.
&
Zhuyuan
B.
2020a
Storm surge risk assessment method for a coastal county in China: Case study of Jinshan District, Shanghai
.
Stochastic Environmental Research and Risk Assessment
34
(
5
),
627
640
.
Shi
X.
,
Yu
P.
,
Guo
Z.
,
Sun
Z.
,
Chen
F.
,
Wu
X.
,
Cheng
W.
&
Zeng
J.
2020b
Simulation of storm surge inundation under different typhoon intensity scenarios: Case study of Pingyang County, China
.
Natural Hazards and Earth System Sciences
20
(
10
),
2777
2790
.
Shi
X.
,
Lv
Y.
,
Dong
D.
,
Jia
N.
,
Ge
J.
&
Yin
J.
2024
Quantitative assessment of building risks and loss ratios caused by storm surge disasters: A case study of Xiamen, China
.
Applied Ocean Research
145
,
103934
.
Su
X.
,
Shao
W.
,
Liu
J.
,
Jiang
Y.
,
Shao
R.
&
Wang
K.
2022
Dynamic assessment of economic losses from flood disasters based on scenario simulations
.
Journal of Tsinghua University: Natural Science Edition
62
(
10
),
1606
1617
.
Tang
J.
,
Li
W.
,
Fang
J.
,
Zhang
Z.
,
Du
S.
,
Wu
Y.
&
Wen
J.
2021
Scenario-based economic and societal risk assessment of storm flooding in Shanghai
.
International Journal of Climate Change Strategies and Management
13
(
4/5
),
529
546
.
Van de Voorde
T.
,
Jacquet
W.
&
Canters
F.
2011
Mapping form and function in urban areas: An approach based on urban metrics and continuous impervious surface data
.
Landscape and Urban Planning
102
(
3
),
143
155
.
Wang
S.
,
Mu
L.
,
Qi
M.
,
Yu
Z.
,
Yao
Z.
&
Zhao
E.
2021a
Quantitative risk assessment of storm surge using GIS techniques and open data: A case study of Daya Bay Zone, China
.
Journal of Environmental Management
289
,
112514
.
Wang
S.
,
Mu
L.
,
Yao
Z.
,
Gao
J.
,
Zhao
E.
&
Wang
L.
2021b
Assessing and zoning of typhoon storm surge risk with a geographic information system (GIS) technique: A case study of the coastal area of Huizhou
.
Natural Hazards and Earth System Sciences
21
(
1
),
439
462
.
Wang
Y.
,
Guo
Z.
,
Zheng
S.
,
Zhang
M.
,
Shu
X.
,
Luo
J.
,
Qiu
L.
&
Gao
T.
2021c
Risk assessment for typhoon-induced storm surges in Wenchang, Hainan Island of China
.
Geomatics, Natural Hazards and Risk
12
(
1
),
880
899
.
Wellmann
T.
,
Lausch
A.
,
Andersson
E.
,
Knapp
S.
,
Cortinovis
C.
,
Jache
J.
,
Scheuer
S.
,
Kremer
P.
,
Mascarenhas
A.
&
Kraemer
R.
2020
Remote sensing in urban planning: Contributions towards ecologically sound policies?
Landscape and Urban Planning
204
,
103921
.
Wu
K.
,
Wang
C.
,
Wu
L.
,
Zhao
H.
&
Cao
J.
2022
Slowdown in landfalling tropical cyclone motion in South China
.
Geophysical Research Letters
49
(
21
),
e2022GL100428
.
Yin
Z.
,
Yin
J.
,
Xu
S.
&
Wen
J.
2011
Community-based scenario modelling and disaster risk assessment of urban rainstorm waterlogging
.
Journal of Geographical Sciences
21
(
2
),
274
284
.
Yin
J.
,
Yin
Z.
,
Yu
D.
&
Xu
S.
2012
Vulnerability analysis for storm induced flood: A case study of Huangpu River basin
.
Scientia Geographica Sinica
32
(
9
),
1155
1160
.
Yin
J.
,
Lin
N.
,
Yang
Y.
,
Pringle
W. J.
,
Tan
J.
,
Westerink
J. J.
&
Yu
D.
2021
Hazard assessment for typhoon-induced coastal flooding and inundation in Shanghai, China
.
Journal of Geophysical Research: Oceans
126
(
7
),
e2021JC017319
.
Zanaga
D.
,
Van De Kerchove
R.
,
Daems
D.
,
De Keersmaecker
W.
,
Brockmann
C.
,
Kirches
G.
,
Wevers
J.
,
Cartus
O.
,
Santoro
M.
&
Fritz
S.
2022
ESA WorldCover 10m 2021 v200
.
Zhejiang Institute of Hydraulics and Estuary
2007
Evaluation of Flood Control and Disaster Reduction Capacity of Water Conservancy Projects in Zhejiang Province, Hangzhou
.
Zhuang
L.
,
Zhang
Z.
,
Wang
M.
,
Liu
K.
,
Liu
C.
&
Wu
W.
2024
Investigating the role of refining urban flood direct loss in enhancing the accuracy of indirect economic loss assessment
.
International Journal of Disaster Risk Reduction
102
,
104285
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).

Supplementary data