Identification and Characterization of Highly Potential Post-Disaster Isolated Areas through Sorting Social Vulnerability

Identification and Characterization of Highly Potential Post-Disaster Isolated Areas through Sorting Social Vulnerability | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Identification and Characterization of Highly Potential Post-Disaster Isolated Areas through Sorting Social Vulnerability Jiuh-Biing Sheu, Yenming Chen, Kuo-Hao Chang, KUAN TING Li, Chih-Hao Liu, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3844488/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Identifying and characterizing post-disaster isolated areas are critical to the success of large-scale disaster management. A post-disaster isolated area (PDIA) refers to an area that can hardly be reached because of the destruction of traffic networks amid a disaster. Lacking relief and medical resources also inflicts psychological impacts on vulnerable dwellers in a PDIA. We believe humanitarian relief can be planned prior to disaster devastation. If a connected area has installed a relief facility, such as a hospital, the road damage may not severely affect the residents in PDIAs. This study enables the exploration of PDIAs characterized by the possibility of disaster occurrence and social vulnerability; and moreover, identifying the size of connected areas, also called responsible areas, to facilitate emergency relief supply and distribution in PDIAs amid a disaster. This research represents a joint venture with a national-level government agent, targeting a highly vulnerable territory that permits, efficiently and effectively, identifying and characterizing PDIAs from the perspective of social vulnerability. We adopt multi-data sources that incorporate socioeconomic, geographic, and disaster impact data gained and inputted from a national earthquake impact information platform. By conceptualizing and incorporating a syncretic disaster-risk index into the clustering metric, managerial endeavor becomes possible. We find that the chosen sizes of responsible areas of PDIAs are decisive, and by managing to maintain at least one relief facility in each PDIA, the impact on the dwellers can be mitigated. Earth and environmental sciences/Natural hazards Scientific community and society/Social sciences/Government Intelligent disaster management Social vulnerability Syncretic disaster risk index highly potential Post-disaster isolated area Responsible Area Data-driven spatial clustering National Science and Technology Center for Disaster Reduction Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction After a large-scale natural disaster, substantial influence exists between road network topologies and disaster risk (Zhou et al., 2023). Roads destroyed by earthquakes or landslides could isolate particular regions, preventing resource allocation evacuation of disaster victims. These cut-off areas, known as PDIAs, cannot be accessed from the outside, becoming more vulnerable than other regions due to insufficient disaster relief (Sugiura et al., 2023). In addition to road network vulnerability, social vulnerability has been essential in risk assessment for decades. Research on disaster mitigation has shifted from exploring the causes and factors of disasters to reducing vulnerability based on the total result of disaster mitigation (Briceño, 2015). In disaster management, vulnerability broadly refers to the possibility of suffering losses and injuries, including vulnerability based on social, economic, and political factors (Cutter, 2001). Children, pregnant women, the elderly, solitary seniors, and the disabled form vulnerable groups that increase the disaster risk through social vulnerability factors in the regions (Flanagan et al., 2011). The formation and characteristics of isolated areas on the vulnerability of road networks have not yet been discussed enough. The interactive influence of social vulnerability in the isolated area needs further investigation. Thus, a vital contribution of this study is the integration of the concepts of PDIA and social vulnerability to render a feasible dividing of manageable regions for administrative assignment. By providing a disaster risk assessment for each region, we offer an insight into the disaster mitigation efforts, PDIA, and social vulnerability. Results The investigation was performed on Tainan, Taiwan, a highly versatile metropolitan with highly asymmetric population density located at the joint of four high-risk faults. The city is a special municipality in southern Taiwan with 1.9 million registered residents living in an area of 2,191 km 2, and the elevation starts from 31 m to 1241 m. The complex living styles are governed by 37 administrative districts, each governed by its own district office. The districts vary in population, area, and density, reflecting Tainan's diverse history and culture. We acquired a list of demographic information from other data sources in 12,774 Basic Statistical Areas (BSA) and information about 35 hospitals, as shown in Supplementary Fig. 1. Tainan is known for its rich cultural heritage but also suffers from its unstable geological activities. The city is located in a tectonically active region where several faults have been identified. The Kousiaoli Fault, which last moved 12,670 years ago, is a 21 km long fault that runs through the city. The Chegualin Fault, which last moved 7,500 years ago, is a 25 km long fault that extends from Kaohsiung to Tainan. The Chusiang Fault, which last moved 13,500 years ago, is a 20 km long fault that crosses Nantou County. These faults are considered active because they have shown evidence of movement within the past 100,000 years and could move again. In practice, reliable records of road damage by earthquakes are difficult to collect. NCDR adopts a massive simulation platform to get the consequence of each event and frequently double-checks the simulation results with some limited reported damage cases. We collected earthquake data with magnitude five and above from 1993 to 2020, resulting in 2,800 events, and performed 100 simulations for each earthquake scenario to achieve significant statistical conclusions. A sample of the damage simulation before and after a historical earthquake event is compared in Fig. 1. The simulation illustrates the potential occurrence probabilities of PDIAs for each BSA in Fig. 2. PDIA with sporadic and higher PRI mostly overlapped. As shown in Fig. 3., regions with higher potential occurrence probability of PDIAs mostly have higher PRI values. We find that regions with high population density have higher PRI values. Therefore, the PRI captures the extreme conditions of exposure and vulnerability. It indicates that in addition to being located near rivers, reservoirs, and mountains, the lack of critical facilities can also cause the occurrence of PDIAs. Areas with fewer roads or areas near canals, such as Anping District, had a higher potential occurrence probability of PDIAs. We find that medical relief facilities are vital in rural areas. Due to a retraction from a hospital to a nursing home in Baihe District in 2022, a widespread and higher potential occurrence probability of PDIAs occurred in both Baihe and the nearby Houbi District. The PRI analysis agrees with the existing Social Vulnerability Index for Disasters of NCDR in the level of administrative demographics in Tainan, which reveals that our index is acceptable to administration infrastructure, as shown in Fig. 4. However, our indicator can effectively highlight potential PDIAs when accurate area clustering is properly applied. The government should keep at least a facility located in a responsible area. We conclude that regions with high social vulnerability generally have high PRI values. Conversely, regions with a high potential for PDIAs usually have high PRI values. If the closure of a responsible area is adequately chosen such that the above two clauses are commutable, the PRI value can effectively foresee the occurrence of vulnerable PDIA. To investigate the proper boundaries and size of responsible areas for allocating medical facilities, we have tried K-means, DBSCAN, and Hierarchical clustering methods with evaluation metrics the total within-cluster sum of square, average silhouette score, Davies-Bouldin index, and Calinski-Harabasz index. From the analysis, K-means generally exhibit good clustering metric values for most indexes. It is noticeable that the clustering corresponds to the administrative demographics and population concentration areas of Tainan City. DBSCAN provides intensive and robust clustering results, implying applicability for targeting high-risk areas. The clustering methods, therefore, have been used to find the region closure with a high potential of PDIA. To be accepted by top management, large clustering can provide acceptable allocation aligning with existing administration boundaries. Discussions Past research on the interactive effect of PDIA and social vulnerability has not yet been discussed sufficiently (Soden, Lallemant, and Kalirai, 2023). The primary contribution of this study lies in combining the two concepts, offering data-driven managerial support, and pioneering a new perspective in post-disaster management. Our findings indicate that PDIAs can be both highly and under-populated. Historical disaster events often expel residents from dangerous areas. However, superior economic factors often overwhelm risk factors such that positive population growth is often observed in some high-risk areas (Shu, Porter, Hauer, 2023). In contrast, inferior economic factors also hinder vulnerable dwellers from moving out of such areas (Rachunok and Nateghi, 2021 ; Do, McBrien, Flores, 2023). Proper clustering incorporating social vulnerability in this study shall facilitate accurate disaster reduction planning. At the planning level, it would be possible to understand which areas have lower transportation vulnerability and to strengthen road connectivity (León, Ordaz, and Haddad, 2022; Yarveysi, Alipour, and Moftakhari, 2023). By carefully choosing region boundaries, our PRI can also match the government's existing administrative structure, which the government can quickly adopt. In practice, through our suggestions, medical resources can be allocated to highly potential PDIAs by the responsive agents with highly vulnerable groups if the government needs to conduct an effective mitigation plan before the disaster (Newell, Rakow, and Yechiam, 2016). Methods A risk-based area clustering model is developed in this study. First, based on earthquake scenarios, we estimated the disaster relief population demand and potential occurrence probability of PDIAs for each area. By integrating socioeconomic data, we propose a potential risk index (PRI) and evaluate the values for each area. Finally, we determine proper clustering responsive areas for administration using these regional PRI values. The workflow of the risk-based area clustering model is shown in Fig. 6 . Isolated Area Simulation Model To define whether a road section fails after an earthquake, we adopt the definition of NCDR for the state of road damage (Liu, Wu, Li et al., 2014), as shown in Table 2 . Once a road section is determined to fail, it is considered inaccessible and becomes a potentially isolated area. Table 2 Road Damage State Damage State Definition Reduction efficacy Accessibility Failure Road No damage No deformation 0% Passable No Slightly damaged Slightly deformation (60cm) 90% Closed for several days or weeks Yes The Federal Emergency Management Agency (FEMA) fragility curve (1) can approximate the probability of road sections in damaged states. $${f}_{x}\left(x\right)=\frac{1}{\sqrt{2\pi }\beta x}exp\left[-\frac{1}{2}{\left(\frac{\text{ln}x-\text{ln}m}{\beta }\right)}^{2}\right], 0\le x\le \infty ,$$ 1 where \(x, m, \beta\) are permanent ground deformation, PGD, the damage state's median parameter, and the damage state, and the variance parameter of the damage state, respectively. The simulation took parameters from Table 3 , indicating the median and variance parameters corresponding to the damage state of different types of roads. Main roads are defined as those wider than 15 meters, and secondary roads are those less than 15 meters wide. We can obtain the probability of a road section being damaged by applying these parameters and the permanent ground deformation (PGD) to the fragility curve (1). Table 3 The median and variance parameters corresponding to the damage state of different types of roads Road Type Damage State Median m Variance \(\varvec{\beta }\) Main Road (Road width is more than 15m) Slightly damaged 12 0.7 Moderately damaged 24 0.7 Severely/Completely damaged 60 0.7 Secondary roads(Road width is less than 15m) Slightly damaged 6 0.7 Moderately damaged 12 0.7 Severely/Completely damaged 24 0.7 To obtain the PGD after an earthquake simulation, this study collaborated with the National Science and Technology Center for Disaster Reduction (NCDR). We adopt TERIA (Taiwan Earthquake Impact Research and Information Application Platform), developed by NCDR, to conduct earthquake simulations. We, therefore, acquired the PGD data of a region's road network by setting the earthquake scale, depth, and fault. Supplementary Fig. 5 illustrates the cumulative distribution function of the fragility curves for the main roads. The road damages were simulated based on the probability of the road being moderately damaged or above. For instance, if the probability of a road section being "severely/completely damaged" after the simulation is 30%, it implies a 30% chance that the road section will fail. The collection of road sections left after removing failed sections is the post-disaster road network. Once the post-disaster road network's damaged states were realized from the random simulation of the empirical distribution, we then determine whether an area becomes isolated. Because of aligning with the data collection for social vulnerability, this study uses the basic statistical area (BSA) defined by the Ministry of the Interior's National Land Information System for a basic unit of simulation. In the simulation, some road sections fail. If a network node cannot connect to key facilities (for example, hospitals), it is identified as an “isolated node.” If all network nodes in the BSA cannot reach key facilities, this BSA is deemed an isolated BSA. After applying the rules for determining isolated BSA, we then obtained a set of isolated areas. Taken from 28 significant earthquakes of magnitude five or above in Tainan from 1993 to 2020, we simulated the earthquake scenarios to the TERIA platform and accumulated 100 road damages from each scenario. The generated 2800 sets of post-disaster road network data were sufficient to calculate the probability of being PDIA. Potential Risk Index The Potential Risk Index, as an assessment of the disaster risk of a region, requires not only the potential occurrence probability of PDIAs (environmental vulnerability) but also needs to incorporate social vulnerability factors such as population od elderlies. Yang et al. ( 2014 ) suggest that the social vulnerability index is divided into two stages. The first stage is standardizing variables, such as z-score and relative comparison. The second stage calculates the weights of each variable. In contrast to the risk assessment framework of “Risk = Frequency x Population x Vulnerability” proposed by the United Nations Environment Program (UNEP), this study replaces frequency with the potential occurrence probability of PDIAs. It defines the population as the potential casualty population. The product of the two variables represents the expected value of the exposed population to casualties. However, considering the heterogeneity of population composition between regions, regions with higher proportions of vulnerable groups require more medical facilities than those with lower ones. Therefore, according to the government's announced “Basic Plan for Disaster Prevention and Rescue” compiled by the national agent of the Central Disaster Prevention and Rescue Committee, we define the potential risk index \({v}_{i}\) for area \(i\) : \({ v}_{i} ={p}_{i}\) * \({ x}_{i }\) * \((\) \({{w}_{o}O}_{i}+{{w}_{a}A}_{i}+{{w}_{c}C}_{i}+{w}_{d}{D}_{i}+{{w}_{p}P}_{i})\) , (2) where \({p}_{i}{, x}_{i}\) , \({O}_{i},{ A}_{i}, {C}_{i}, {D}_{i}, {P}_{i}\) are the estimated population number, probability of being isolated, population ratio over age 65, population ratio living alone, population ratio under 12, population ratio of disability, population ratio of pregnancy, in area \(i\) , respectively. The weighting \({w}_{*}\) of the corresponding indicator ranges from 0 to 1. Additionally, due to the varying degrees of casualties among different vulnerable groups, this study put entropy weights of the variables for the five vulnerable groups, allowing variables with higher data variability to have higher weights. The entropy and information for each variable are defined as \({e}_{j}=-\frac{1}{\text{ln}m}{\sum }_{i=1}^{m}\frac{{a}_{ij}}{{\sum }_{i=1}^{m}{a}_{ij}}\text{ln}\frac{{a}_{ij}}{{\sum }_{i=1}^{m}{a}_{ij}}\) , and \({d}_{j}=\) \({1-e}_{j}\) , respectively. A considerable number of BSAs are not manageable for administrative convenience. The management region or responsible area must be large enough to overlap with existing infrastructure and small enough to incorporate a relief facility. The PRI values in BSAs must be distributed as dots to implement the clustering algorithm, as shown in Fig. 8 . After the dots are distributed, they can be clustered through the selected clustering algorithm in this study (such as K-means, DBSCAN. Subsequently, to obtain stable clustering results, we performed clustering with 1000 realization and recorded cluster assignment with the lowest p-value as the final clustering result, as shown in Fig. 9 . Declarations Data Availability None of the data used in this study is publicly accessible. The data used in this study is provided by NCDR-S-111054. References Adger, W. N. (2006). Vulnerability. Global Environmental Change, 16, 268-281. Bohle, H. G., T. E. Downing & M. J. Watts (1994). Climate Change and Social Vulnerability. Toward a Sociology and Geography of Food Insecurity. Global Environmental Change, 4 (1), 37-48. Briceño, S. (2015). Looking back and beyond Sendai:25 Years of International Policy Experience on Disaster Risk Reduction. International Journal of Disaster Risk Science,6 (1) , 1-7. Burton, C., Rufat, S., & Tate, E. (2018). Social Vulnerability: Conceptual Foundations and Geospatial Modeling. In Vulnerability and Resilience to Natural Hazards . Cambridge University Press. Cutter, S. L., Ed. (2001). American hazardscapes: The Regionalization of Hazards and Disasters. Washington, D.C., Joseph Henry Press. Cutter, S. L., Mitchell, J. T., & Scott, M. S. 2000. Revealing the Vulnerability of People and Places: A Case Study of Georgetown County, South Carolina. Annals of the Association of American Geographers, 9 , 713-737. Cutter, S. L., B. J. Boruff & W. L. Shirley (2003). Social Vulnerability to Environmental Hazards. Social Science Quarterly, 84 (1), 242-261. Do, V., McBrien, H., Flores, N.M. et al. Spatiotemporal Distribution of Power Outages with Climate Events and Social Vulnerability in the USA. Nat Commun 14, 2470 (2023). https://doi.org/10.1038/s41467-023-38084-6 León, J.A., Ordaz, M., Haddad, E. et al. Risk Caused by the Propagation of Earthquake Losses through the Economy. Nat Commun 13, 2908 (2022). https://doi.org/10.1038/s41467-022-30504-3 Flanagan, B. E., Gregory, E. W., Hallisey, E. J., Heitgerd, J. L., & Lewis, B. (2011). A Social Vulnerability Index for Disaster Management. Journal of Homeland Security and Emergency Management, 8 (1), 4-6. Liu. Sheu-Yien, Wu. C. Carol, Li. Chin-Yen, Lee. et al. (2014). The Application of ArcGIS on Post-Earthquake Scenario Assessment for Road System, Power and Water Networks, and Emergency Relief and Response Systems. National Science and Technology Center for Disaster Reduction. Newell, B., Rakow, T., Yechiam, E. et al. Rare Disaster Information can Increase Risk-taking. Nature Clim Change 6, 158–161 (2016). https://doi.org/10.1038/nclimate2822 Rachunok, B., Nateghi, R. Overemphasis on Recovery Inhibits Community Transformation and Creates Resilience Traps. Nat Commun 12, 7331 (2021). https://doi.org/10.1038/s41467-021-27359-5 Shu, E.G., Porter, J.R., Hauer, M.E. et al. Integrating Climate Change Induced Flood Risk into Future Population Projections. Nat Commun 14, 7870 (2023). https://doi.org/10.1038/s41467-023-43493-8 Soden, R., Lallemant, D., Kalirai, M. et al. The Importance of Accounting for Equity in Disaster Risk Models. Commun Earth Environ 4, 386 (2023). https://doi.org/10.1038/s43247-023-01039-2 Sugiura, S., & Kurauchi, F. (2023). Isolation Vulnerability Analysis in Road Network: Edge Connectivity and Critical Link Sets. Transportation Research Part D: Transport and Environment , 119 , 103768. Wu, S.-Y., Yarnal, B., & Fisher, A. (2002). Vulnerability of Coastal Communities to Sea-level Rise: A Case Study of Cape May County, New Jersey, USA. Climate Research, 22, 255-270. Yang, H. H., Chen, Y. C. & Li, H. C. (2014). Establishing and Assessing Social Vulnerability Index (SVI) of Natural Disasters: Town Level Application. Journal of Disaster Management, 3 (2), 71-93. Yarveysi, F., Alipour, A., Moftakhari, H. et al. Block-level Vulnerability Assessment Reveals Disproportionate Impacts of Natural Hazards across the Conterminous United States. Nat Commun 14, 4222 (2023). https://doi.org/10.1038/s41467-023-39853-z Zhou, M., Yuan, M., Yang, G., & Mei, G. (2023). Risk Analysis of Road Networks under the Influence of Landslides by Considering Landslide Susceptibility and Road Vulnerability: A Case Study. Natural Hazards Research . Additional Declarations There is NO Competing Interest. Supplementary Files Supplementaryinformation.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3844488","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":267875334,"identity":"f0986fd8-3749-4ba0-86dc-16735a67a34c","order_by":0,"name":"Jiuh-Biing Sheu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvklEQVRIiWNgGAWjYBACxhkMbEDKBsLjIUFLGkQ1kJAgrEcCrOUwCVqYZ7c/e/Cx7XyevUQC44O3bQx1BgcIOWzOGXPDmW23i3kkEpgN57YxSBDWMiOHTZq37XZij0QCiMEgYUZYS/oz6b9t50Ba2H8TqSXBTJqx7QDYFmbitMw5YybZcy45sefMw2bJOeckJPcT0mIIDDGJH2V2ie3tyQc/vCmz4ZdsIKQFoYARxCQiJuUJKxkFo2AUjIIRDwC/6D0QzGleogAAAABJRU5ErkJggg==","orcid":"","institution":"National Taiwan University","correspondingAuthor":true,"prefix":"","firstName":"Jiuh-Biing","middleName":"","lastName":"Sheu","suffix":""},{"id":267875335,"identity":"e494b82a-3cc4-415a-948d-38c107451c9c","order_by":1,"name":"Yenming Chen","email":"","orcid":"","institution":"National Kaohsiung University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Yenming","middleName":"","lastName":"Chen","suffix":""},{"id":267875336,"identity":"993b216f-0c02-4479-aae6-b39d01397063","order_by":2,"name":"Kuo-Hao Chang","email":"","orcid":"","institution":"National Tsing Hua University","correspondingAuthor":false,"prefix":"","firstName":"Kuo-Hao","middleName":"","lastName":"Chang","suffix":""},{"id":267875337,"identity":"3286aacb-1ff1-4da6-b23b-97a14766e70f","order_by":3,"name":"KUAN TING Li","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"KUAN","middleName":"TING","lastName":"Li","suffix":""},{"id":267875338,"identity":"f4b9796e-9996-492e-8946-ba6d53058c58","order_by":4,"name":"Chih-Hao Liu","email":"","orcid":"","institution":"National Science and Technology Center for Disaster Reduction of Taiwan","correspondingAuthor":false,"prefix":"","firstName":"Chih-Hao","middleName":"","lastName":"Liu","suffix":""},{"id":267875339,"identity":"752e5d48-43ab-4712-a111-9b7bf8e20b13","order_by":5,"name":"Tzu-Yin Chang","email":"","orcid":"","institution":"National Science and Technology Center for Disaster Reduction of Taiwan","correspondingAuthor":false,"prefix":"","firstName":"Tzu-Yin","middleName":"","lastName":"Chang","suffix":""}],"badges":[],"createdAt":"2024-01-08 04:10:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3844488/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3844488/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":49858233,"identity":"24b4bb22-4f80-4fd2-8778-2dfa4277fe0e","added_by":"auto","created_at":"2024-01-19 08:16:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1519977,"visible":true,"origin":"","legend":"\u003cp\u003eThe sample image of pre-disaster and post-disaster road networks. Each area divided by black lines represents one BSA. The red lines represent the road network.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/aca4c07f42b90cbfe2286504.png"},{"id":49858365,"identity":"acbbd39d-f5a8-40b7-9ab1-306484e34187","added_by":"auto","created_at":"2024-01-19 08:24:58","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":452767,"visible":true,"origin":"","legend":"\u003cp\u003ePotential Occurrence Probabilities of PDIAs for each BSA. The blue dots represent hospitals. The saturation of the color represents the probability of being isolated from one BSA. The map implies three characteristics of an isolated area.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/eb7744af9afc71bf28136a33.png"},{"id":49858230,"identity":"a8f02d01-438f-46b7-a306-e2bbf424cca0","added_by":"auto","created_at":"2024-01-19 08:16:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":650540,"visible":true,"origin":"","legend":"\u003cp\u003ePRI values of the BSAs. The blue dots represent hospitals. The saturation of the color represents the value of PRIs. The map implies two characteristics of PRI.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/4fdc9ab04e594c4d643cda7d.png"},{"id":49857904,"identity":"bf774870-12f3-4af5-92bc-a2113575b8f2","added_by":"auto","created_at":"2024-01-19 08:08:58","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":715131,"visible":true,"origin":"","legend":"\u003cp\u003ePRI value in existing administration boundaries. The Exposure Index of SVID in NCDR is shown in the left map, revealing each area's exposurein Tainan City. The color represents the value of the Exposure Index. The Preparedness Index of SVID in NCDR is shown on the correct map, revealing each area's vulnerability in Tainan City. The color represents the value of the Preparedness Index.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/d3498caba12abe62b744e5a5.png"},{"id":49857896,"identity":"363137b6-467c-438d-9390-1dfd1ac20939","added_by":"auto","created_at":"2024-01-19 08:08:58","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":235384,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLine Chart of the Number of Clusters corresponding to the Index Value\u003c/strong\u003e. The four charts are the charts of each index compared to different numbers of clusters. The vertical axis in each chart is the value of indexes respectively. The horizontal axis in each chart is the number of clusters. Each line in each chart means one clustering algorithm.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/bbc89937cf56b4c23231fa58.png"},{"id":49857898,"identity":"b83dbb7f-a94f-4bc7-b038-195ac97ecdeb","added_by":"auto","created_at":"2024-01-19 08:08:58","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":236335,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow of Risk-based Area Clustering Model.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/bcedce6ff01e65dbbb76a344.png"},{"id":49858232,"identity":"ceb13cab-57f5-48b5-8e95-6cdd804db952","added_by":"auto","created_at":"2024-01-19 08:16:58","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":107917,"visible":true,"origin":"","legend":"\u003cp\u003eConnectivity losses of the road network before and after the disaster. Each dot represents an intersection. Each line represents a road.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/f7714c790985e2e96addc399.png"},{"id":49857900,"identity":"ffbc53b6-a326-44a5-a815-91f67532c284","added_by":"auto","created_at":"2024-01-19 08:08:58","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":84281,"visible":true,"origin":"","legend":"\u003cp\u003eConverting PRI values to dots for implementing clustering algorithms\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/9f3e0675787dccdad309e8d0.png"},{"id":49858364,"identity":"72c04b14-76b1-416f-ad55-c0a06d7860f0","added_by":"auto","created_at":"2024-01-19 08:24:58","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":89322,"visible":true,"origin":"","legend":"\u003cp\u003eGenerating Final Clustering Results. The method of allocating the cluster. In each area, the cluster allocation has the highest dot number.\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/b823ae7ae07d694bc2ce4ab9.png"},{"id":58607489,"identity":"8d6712c4-503b-421b-9ff6-077dad05b835","added_by":"auto","created_at":"2024-06-18 20:53:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4618779,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/d32b9975-c7fe-4398-a466-0c4a3058255d.pdf"},{"id":49857906,"identity":"440316f8-7c04-48af-a3bb-768ecb52bffa","added_by":"auto","created_at":"2024-01-19 08:08:59","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":13059054,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-3844488/v1/72ada07b00311dbc2c1fee74.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Identification and Characterization of Highly Potential Post-Disaster Isolated Areas through Sorting Social Vulnerability","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAfter a large-scale natural disaster, substantial influence exists between road network topologies and disaster risk\u0026nbsp;(Zhou\u0026nbsp;et al., 2023). Roads destroyed by earthquakes or landslides could isolate particular regions, preventing resource allocation evacuation of disaster victims. These cut-off areas, known as PDIAs, cannot be accessed from the outside, becoming more vulnerable than other regions due to insufficient disaster relief\u0026nbsp;(Sugiura\u0026nbsp;et al., 2023).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn addition to road network vulnerability, social vulnerability has been essential in risk assessment for decades. Research on disaster mitigation has shifted from exploring the causes and factors of disasters to reducing vulnerability based on the total result of disaster mitigation (Brice\u0026ntilde;o, 2015). In disaster management, vulnerability broadly refers to the possibility of suffering losses and injuries, including vulnerability based on social, economic, and political factors (Cutter, 2001). Children, pregnant women, the elderly, solitary seniors, and the disabled form vulnerable groups that increase the disaster risk through social vulnerability factors in the regions (Flanagan et al., 2011).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe formation and characteristics of isolated areas on the vulnerability of road networks have not yet been discussed enough. The interactive influence of social vulnerability in the isolated area needs further investigation. Thus, a vital contribution of this study is the integration of the concepts of PDIA and social vulnerability to render a feasible dividing of manageable regions for administrative assignment. By providing a disaster risk assessment for each region, we offer an insight into the disaster mitigation efforts, PDIA, and social vulnerability.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe investigation was performed on Tainan, Taiwan, a highly versatile metropolitan with highly asymmetric population density located at the joint of\u0026nbsp;four high-risk faults.\u0026nbsp;The city is a special municipality in southern Taiwan with 1.9 million registered residents living in an area of 2,191 km\u003csup\u003e2,\u003c/sup\u003e and the elevation starts from 31 m to\u0026nbsp;1241 m. The complex living styles are governed by\u0026nbsp;37 administrative districts, each governed by its own district office. The districts vary in population, area, and density, reflecting Tainan\u0026apos;s diverse history and culture. We acquired a list of demographic information from other data sources in 12,774 Basic Statistical Areas (BSA) and\u0026nbsp;information about 35 hospitals, as shown in Supplementary Fig. 1.\u003c/p\u003e\n\u003cp\u003eTainan is known for its rich cultural heritage but also suffers from its unstable\u0026nbsp;geological activities. The city\u0026nbsp;is located in a tectonically active region where several faults have been identified. The Kousiaoli Fault, which last moved 12,670 years ago, is a 21 km long fault that runs through the city. The Chegualin Fault, which last moved 7,500 years ago, is a 25 km long fault that extends from Kaohsiung to Tainan. The Chusiang Fault, which last moved 13,500 years ago, is a 20 km long fault that crosses Nantou County. These faults are considered active because they have shown evidence of movement within the past 100,000 years and could move again.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn practice, reliable records of road damage by earthquakes are difficult to collect. NCDR adopts a massive simulation platform to get the consequence of each event and frequently double-checks the simulation results with some limited reported damage cases. We collected earthquake data with magnitude five and above from 1993 to 2020, resulting in 2,800 events, and performed 100 simulations for each earthquake scenario to achieve significant statistical conclusions. A sample of the damage simulation before and after a historical earthquake event is compared in Fig. 1.\u003c/p\u003e\n\u003cp\u003eThe simulation illustrates the potential occurrence probabilities of PDIAs for each BSA in Fig. 2. PDIA with sporadic and higher PRI mostly overlapped. As shown in Fig. 3., regions with higher potential occurrence probability of PDIAs mostly have higher PRI values. We find that regions with high population density have higher PRI values. Therefore, the PRI captures the extreme conditions of exposure and vulnerability.\u003c/p\u003e\n\u003cp\u003eIt indicates that in addition to being located near rivers, reservoirs, and mountains, the lack of critical facilities can also cause the occurrence of PDIAs. Areas with fewer roads or areas near canals, such as Anping District, had a higher potential occurrence probability of PDIAs. We find that medical relief facilities are vital in rural areas. Due to a retraction from a hospital to a nursing home in Baihe District in 2022, a widespread and higher potential occurrence probability of PDIAs occurred in both Baihe and the nearby Houbi District.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe PRI analysis agrees with the existing Social Vulnerability Index for Disasters of NCDR in the level of administrative\u0026nbsp;demographics\u0026nbsp;in Tainan, which reveals that our index is acceptable to administration infrastructure, as shown in Fig. 4. However, our indicator can effectively highlight potential PDIAs when accurate area clustering is properly applied. The government should keep at least a facility located in a responsible area.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe conclude that regions\u0026nbsp;with high social vulnerability generally have high PRI values. Conversely, regions with a high potential for PDIAs usually have high PRI values. If the closure of a responsible area is adequately chosen such that the above two clauses are commutable, the PRI value can effectively foresee the occurrence of vulnerable PDIA.\u003c/p\u003e\n\u003cp\u003eTo investigate the proper boundaries and size of responsible areas for allocating medical facilities, we have tried K-means, DBSCAN, and Hierarchical clustering methods with evaluation metrics the total within-cluster sum of square, average silhouette score, Davies-Bouldin index, and Calinski-Harabasz index. From the analysis, K-means generally exhibit good clustering metric values for most indexes. It is noticeable that the clustering corresponds to the administrative demographics and population concentration areas of Tainan City. DBSCAN provides intensive and robust clustering results, implying applicability for targeting high-risk areas. The clustering methods, therefore, have been used to find the region closure with a high potential of PDIA. To be accepted by top management, large clustering can provide acceptable allocation aligning with existing administration boundaries.\u003c/p\u003e"},{"header":"Discussions","content":"\u003cp\u003ePast research on the interactive effect of PDIA and social vulnerability has not yet been discussed sufficiently (Soden, Lallemant, and Kalirai, 2023). The primary contribution of this study lies in combining the two concepts, offering data-driven managerial support, and pioneering a new perspective in post-disaster management.\u003c/p\u003e \u003cp\u003eOur findings indicate that PDIAs can be both highly and under-populated. Historical disaster events often expel residents from dangerous areas. However, superior economic factors often overwhelm risk factors such that positive population growth is often observed in some high-risk areas (Shu, Porter, Hauer, 2023). In contrast, inferior economic factors also hinder vulnerable dwellers from moving out of such areas (Rachunok and Nateghi, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Do, McBrien, Flores, 2023).\u003c/p\u003e \u003cp\u003eProper clustering incorporating social vulnerability in this study shall facilitate accurate disaster reduction planning. At the planning level, it would be possible to understand which areas have lower transportation vulnerability and to strengthen road connectivity (Le\u0026oacute;n, Ordaz, and Haddad, 2022; Yarveysi, Alipour, and Moftakhari, 2023).\u003c/p\u003e \u003cp\u003eBy carefully choosing region boundaries, our PRI can also match the government's existing administrative structure, which the government can quickly adopt. In practice, through our suggestions, medical resources can be allocated to highly potential PDIAs by the responsive agents with highly vulnerable groups if the government needs to conduct an effective mitigation plan before the disaster (Newell, Rakow, and Yechiam, 2016).\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eA risk-based area clustering model is developed in this study. First, based on earthquake scenarios, we estimated the disaster relief population demand and potential occurrence probability of PDIAs for each area. By integrating socioeconomic data, we propose a potential risk index (PRI) and evaluate the values for each area. Finally, we determine proper clustering responsive areas for administration using these regional PRI values. The workflow of the risk-based area clustering model is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eIsolated Area Simulation Model\u003c/h2\u003e \u003cp\u003eTo define whether a road section fails after an earthquake, we adopt the definition of NCDR for the state of road damage (Liu, Wu, Li et al., 2014), as shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eOnce a road section is determined to fail, it is considered inaccessible and becomes a potentially isolated area.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRoad Damage State\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDamage State\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDefinition\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReduction\u003c/p\u003e \u003cp\u003eefficacy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAccessibility\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFailure Road\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo damage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo deformation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePassable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSlightly\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSlightly deformation\u003c/p\u003e \u003cp\u003e(\u0026lt; 30cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePassable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModerately\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModerately deformation\u003c/p\u003e \u003cp\u003e(30-60cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eClosed for several days\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSeverely/Completely\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSeverely/Completely deformation(\u0026gt;60cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eClosed for several days or weeks\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe Federal Emergency Management Agency (FEMA) fragility curve (1) can approximate the probability of road sections in damaged states.\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$${f}_{x}\\left(x\\right)=\\frac{1}{\\sqrt{2\\pi }\\beta x}exp\\left[-\\frac{1}{2}{\\left(\\frac{\\text{ln}x-\\text{ln}m}{\\beta }\\right)}^{2}\\right], 0\\le x\\le \\infty ,$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(x, m, \\beta\\)\u003c/span\u003e\u003c/span\u003e are permanent ground deformation, PGD, the damage state's median parameter, and the damage state, and the variance parameter of the damage state, respectively. The simulation took parameters from Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e3\u003c/span\u003e, indicating the median and variance parameters corresponding to the damage state of different types of roads. Main roads are defined as those wider than 15 meters, and secondary roads are those less than 15 meters wide. We can obtain the probability of a road section being damaged by applying these parameters and the permanent ground deformation (PGD) to the fragility curve (1).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe median and variance parameters corresponding to the damage state of different types of roads\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRoad Type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDamage State\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMedian \u003cem\u003em\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVariance\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\varvec{\\beta }\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eMain Road\u003c/p\u003e \u003cp\u003e(Road width is more than 15m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSlightly\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModerately\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSeverely/Completely\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eSecondary roads(Road width is less than 15m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSlightly\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModerately\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSeverely/Completely\u003c/p\u003e \u003cp\u003edamaged\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTo obtain the PGD after an earthquake simulation, this study collaborated with the National Science and Technology Center for Disaster Reduction (NCDR). We adopt TERIA (Taiwan Earthquake Impact Research and Information Application Platform), developed by NCDR, to conduct earthquake simulations. We, therefore, acquired the PGD data of a region's road network by setting the earthquake scale, depth, and fault. Supplementary Fig.\u0026nbsp;5 illustrates the cumulative distribution function of the fragility curves for the main roads.\u003c/p\u003e \u003cp\u003eThe road damages were simulated based on the probability of the road being moderately damaged or above. For instance, if the probability of a road section being \"severely/completely damaged\" after the simulation is 30%, it implies a 30% chance that the road section will fail. The collection of road sections left after removing failed sections is the post-disaster road network.\u003c/p\u003e \u003cp\u003eOnce the post-disaster road network's damaged states were realized from the random simulation of the empirical distribution, we then determine whether an area becomes isolated. Because of aligning with the data collection for social vulnerability, this study uses the basic statistical area (BSA) defined by the Ministry of the Interior's National Land Information System for a basic unit of simulation.\u003c/p\u003e \u003cp\u003eIn the simulation, some road sections fail. If a network node cannot connect to key facilities (for example, hospitals), it is identified as an \u0026ldquo;isolated node.\u0026rdquo; If all network nodes in the BSA cannot reach key facilities, this BSA is deemed an isolated BSA.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAfter applying the rules for determining isolated BSA, we then obtained a set of isolated areas. Taken from 28 significant earthquakes of magnitude five or above in Tainan from 1993 to 2020, we simulated the earthquake scenarios to the TERIA platform and accumulated 100 road damages from each scenario. The generated 2800 sets of post-disaster road network data were sufficient to calculate the probability of being PDIA.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003ePotential Risk Index\u003c/h2\u003e \u003cp\u003eThe Potential Risk Index, as an assessment of the disaster risk of a region, requires not only the potential occurrence probability of PDIAs (environmental vulnerability) but also needs to incorporate social vulnerability factors such as population od elderlies. Yang et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) suggest that the social vulnerability index is divided into two stages. The first stage is standardizing variables, such as z-score and relative comparison. The second stage calculates the weights of each variable.\u003c/p\u003e \u003cp\u003eIn contrast to the risk assessment framework of \u0026ldquo;Risk\u0026thinsp;=\u0026thinsp;Frequency x Population x Vulnerability\u0026rdquo; proposed by the United Nations Environment Program (UNEP), this study replaces frequency with the potential occurrence probability of PDIAs. It defines the population as the potential casualty population. The product of the two variables represents the expected value of the exposed population to casualties. However, considering the heterogeneity of population composition between regions, regions with higher proportions of vulnerable groups require more medical facilities than those with lower ones. Therefore, according to the government's announced \u0026ldquo;Basic Plan for Disaster Prevention and Rescue\u0026rdquo; compiled by the national agent of the Central Disaster Prevention and Rescue Committee, we define the potential risk index \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({v}_{i}\\)\u003c/span\u003e\u003c/span\u003e for area \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(i\\)\u003c/span\u003e\u003c/span\u003e:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\({ v}_{i} ={p}_{i}\\)\u003c/span\u003e \u003c/span\u003e \u003cem\u003e*\u003c/em\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({ x}_{i }\\)\u003c/span\u003e\u003c/span\u003e \u003cem\u003e*\u003c/em\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\((\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({{w}_{o}O}_{i}+{{w}_{a}A}_{i}+{{w}_{c}C}_{i}+{w}_{d}{D}_{i}+{{w}_{p}P}_{i})\\)\u003c/span\u003e\u003c/span\u003e, (2)\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({p}_{i}{, x}_{i}\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({O}_{i},{ A}_{i}, {C}_{i}, {D}_{i}, {P}_{i}\\)\u003c/span\u003e\u003c/span\u003eare the estimated population number, probability of being isolated, population ratio over age 65, population ratio living alone, population ratio under 12, population ratio of disability, population ratio of pregnancy, in area \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(i\\)\u003c/span\u003e\u003c/span\u003e, respectively. The weighting \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({w}_{*}\\)\u003c/span\u003e\u003c/span\u003e of the corresponding indicator ranges from 0 to 1.\u003c/p\u003e \u003cp\u003eAdditionally, due to the varying degrees of casualties among different vulnerable groups, this study put entropy weights of the variables for the five vulnerable groups, allowing variables with higher data variability to have higher weights. The entropy and information for each variable are defined as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({e}_{j}=-\\frac{1}{\\text{ln}m}{\\sum }_{i=1}^{m}\\frac{{a}_{ij}}{{\\sum }_{i=1}^{m}{a}_{ij}}\\text{ln}\\frac{{a}_{ij}}{{\\sum }_{i=1}^{m}{a}_{ij}}\\)\u003c/span\u003e\u003c/span\u003e, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({d}_{j}=\\)\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({1-e}_{j}\\)\u003c/span\u003e\u003c/span\u003e, respectively.\u003c/p\u003e \u003cp\u003eA considerable number of BSAs are not manageable for administrative convenience. The management region or responsible area must be large enough to overlap with existing infrastructure and small enough to incorporate a relief facility.\u003c/p\u003e \u003cp\u003eThe PRI values in BSAs must be distributed as dots to implement the clustering algorithm, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAfter the dots are distributed, they can be clustered through the selected clustering algorithm in this study (such as K-means, DBSCAN. Subsequently, to obtain stable clustering results, we performed clustering with 1000 realization and recorded cluster assignment with the lowest p-value as the final clustering result, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e.\u003c/p\u003e\u003c/div\u003e "},{"header":"Declarations","content":"\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eData Availability\u003c/h2\u003e \u003cp\u003eNone of the data used in this study is publicly accessible. The data used in this study is provided by NCDR-S-111054.\u003c/p\u003e \u003c/div\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAdger, W. N. (2006). Vulnerability. \u003cem\u003eGlobal Environmental Change, 16,\u003c/em\u003e 268-281.\u003c/li\u003e\n\u003cli\u003eBohle, H. G., T. E. Downing \u0026amp; M. J. Watts (1994). Climate Change and Social Vulnerability. Toward a Sociology and Geography of Food Insecurity. \u003cem\u003eGlobal Environmental Change, 4\u003c/em\u003e(1), 37-48.\u003c/li\u003e\n\u003cli\u003eBrice\u0026ntilde;o, S. (2015). Looking back and beyond Sendai:25 Years of International Policy Experience on Disaster Risk Reduction. \u003cem\u003eInternational Journal of Disaster Risk Science,6\u003c/em\u003e(1)\u003cem\u003e,\u003c/em\u003e 1-7.\u003c/li\u003e\n\u003cli\u003eBurton, C., Rufat, S., \u0026amp; Tate, E. (2018). Social Vulnerability: Conceptual Foundations and Geospatial Modeling. In \u003cem\u003eVulnerability and Resilience to Natural Hazards\u003c/em\u003e. Cambridge University Press.\u003c/li\u003e\n\u003cli\u003eCutter, S. L., Ed. (2001). \u003cem\u003eAmerican hazardscapes: The Regionalization of Hazards and Disasters. \u003c/em\u003eWashington, D.C., Joseph Henry Press.\u003c/li\u003e\n\u003cli\u003eCutter, S. L., Mitchell, J. T., \u0026amp; Scott, M. S. 2000. Revealing the Vulnerability of People and Places: A Case Study of Georgetown County, South Carolina. \u003cem\u003eAnnals of the Association of American Geographers, 9\u003c/em\u003e, 713-737.\u003c/li\u003e\n\u003cli\u003eCutter, S. L., B. J. Boruff \u0026amp; W. L. Shirley (2003). Social Vulnerability to Environmental Hazards. \u003cem\u003eSocial Science Quarterly, 84\u003c/em\u003e(1), 242-261.\u003c/li\u003e\n\u003cli\u003eDo, V., McBrien, H., Flores, N.M. et al. Spatiotemporal Distribution of Power Outages with Climate Events and Social Vulnerability in the USA. Nat Commun 14, 2470 (2023). https://doi.org/10.1038/s41467-023-38084-6\u003c/li\u003e\n\u003cli\u003eLe\u0026oacute;n, J.A., Ordaz, M., Haddad, E. et al. Risk Caused by the Propagation of Earthquake Losses through the Economy. Nat Commun 13, 2908 (2022). https://doi.org/10.1038/s41467-022-30504-3\u003c/li\u003e\n\u003cli\u003eFlanagan, B. E., Gregory, E. W., Hallisey, E. J., Heitgerd, J. L., \u0026amp; Lewis, B. (2011). A Social Vulnerability Index for Disaster Management. \u003cem\u003eJournal of Homeland Security and Emergency Management, 8\u003c/em\u003e(1), 4-6.\u003c/li\u003e\n\u003cli\u003eLiu. Sheu-Yien, Wu. C. Carol, Li. Chin-Yen, Lee. et al. (2014). \u003cem\u003eThe Application of ArcGIS on Post-Earthquake Scenario Assessment for Road System, Power and Water Networks, and Emergency Relief and Response Systems. \u003c/em\u003eNational Science and Technology Center for Disaster Reduction.\u003c/li\u003e\n\u003cli\u003eNewell, B., Rakow, T., Yechiam, E. et al. Rare Disaster Information can Increase Risk-taking. Nature Clim Change 6, 158\u0026ndash;161 (2016). https://doi.org/10.1038/nclimate2822\u003c/li\u003e\n\u003cli\u003eRachunok, B., Nateghi, R. Overemphasis on Recovery Inhibits Community Transformation and Creates Resilience Traps. Nat Commun 12, 7331 (2021). https://doi.org/10.1038/s41467-021-27359-5\u003c/li\u003e\n\u003cli\u003eShu, E.G., Porter, J.R., Hauer, M.E. et al. Integrating Climate Change Induced Flood Risk into Future Population Projections. Nat Commun 14, 7870 (2023). https://doi.org/10.1038/s41467-023-43493-8\u003c/li\u003e\n\u003cli\u003eSoden, R., Lallemant, D., Kalirai, M. et al. The Importance of Accounting for Equity in Disaster Risk Models. Commun Earth Environ 4, 386 (2023). https://doi.org/10.1038/s43247-023-01039-2\u003c/li\u003e\n\u003cli\u003eSugiura, S., \u0026amp; Kurauchi, F. (2023). Isolation Vulnerability Analysis in Road Network: Edge Connectivity and Critical Link Sets. \u003cem\u003eTransportation Research Part D: Transport and Environment\u003c/em\u003e, \u003cem\u003e119\u003c/em\u003e, 103768.\u003c/li\u003e\n\u003cli\u003eWu, S.-Y., Yarnal, B., \u0026amp; Fisher, A. (2002). Vulnerability of Coastal Communities to Sea-level Rise: A Case Study of Cape May County, New Jersey, USA. \u003cem\u003eClimate Research, 22,\u003c/em\u003e 255-270.\u003c/li\u003e\n\u003cli\u003eYang, H. H., Chen, Y. C. \u0026amp; Li, H. C. (2014). Establishing and Assessing Social Vulnerability Index (SVI) of Natural Disasters: Town Level Application. \u003cem\u003eJournal of Disaster Management, 3\u003c/em\u003e(2), 71-93.\u003c/li\u003e\n\u003cli\u003eYarveysi, F., Alipour, A., Moftakhari, H. et al. Block-level Vulnerability Assessment Reveals Disproportionate Impacts of Natural Hazards across the Conterminous United States. Nat Commun 14, 4222 (2023). https://doi.org/10.1038/s41467-023-39853-z\u003c/li\u003e\n\u003cli\u003eZhou, M., Yuan, M., Yang, G., \u0026amp; Mei, G. (2023). Risk Analysis of Road Networks under the Influence of Landslides by Considering Landslide Susceptibility and Road Vulnerability: A Case Study. \u003cem\u003eNatural Hazards Research\u003c/em\u003e.\u003cstrong\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Intelligent disaster management, Social vulnerability, Syncretic disaster risk index, highly potential Post-disaster isolated area, Responsible Area, Data-driven spatial clustering, National Science and Technology Center for Disaster Reduction","lastPublishedDoi":"10.21203/rs.3.rs-3844488/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3844488/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIdentifying and characterizing post-disaster isolated areas are critical to the success of large-scale disaster management. A post-disaster isolated area (PDIA) refers to an area that can hardly be reached because of the destruction of traffic networks amid a disaster. Lacking relief and medical resources also inflicts psychological impacts on vulnerable dwellers in a PDIA. We believe humanitarian relief can be planned prior to disaster devastation. If a connected area has installed a relief facility, such as a hospital, the road damage may not severely affect the residents in PDIAs. This study enables the exploration of PDIAs characterized by the possibility of disaster occurrence and social vulnerability; and moreover, identifying the size of connected areas, also called responsible areas, to facilitate emergency relief supply and distribution in PDIAs amid a disaster. This research represents a joint venture with a national-level government agent, targeting a highly vulnerable territory that permits, efficiently and effectively, identifying and characterizing PDIAs from the perspective of social vulnerability. We adopt multi-data sources that incorporate socioeconomic, geographic, and disaster impact data gained and inputted from a national earthquake impact information platform. By conceptualizing and incorporating a syncretic disaster-risk index into the clustering metric, managerial endeavor becomes possible. We find that the chosen sizes of responsible areas of PDIAs are decisive, and by managing to maintain at least one relief facility in each PDIA, the impact on the dwellers can be mitigated.\u003c/p\u003e","manuscriptTitle":"Identification and Characterization of Highly Potential Post-Disaster Isolated Areas through Sorting Social Vulnerability","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-19 08:08:53","doi":"10.21203/rs.3.rs-3844488/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6fb1a7fc-c673-4ac3-b319-1073330300f8","owner":[],"postedDate":"January 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":28227739,"name":"Earth and environmental sciences/Natural hazards"},{"id":28227740,"name":"Scientific community and society/Social sciences/Government"}],"tags":[],"updatedAt":"2024-06-18T20:45:17+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-19 08:08:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3844488","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3844488","identity":"rs-3844488","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}